Prof. Lubomir M. Hadjiiski Profile

Prof. Lubomir M. Hadjiiski

Professor of Radiology at Michigan Medicine

SPIE Involvement:

Conference Program Committee | Author | Editor

Publications (168)

Proceedings Article | 4 October 2024 Presentation

Practical considerations for AI in medical imaging: tool development and validation

Lubomir Hadjiiski

Proceedings Volume PC13118, PC131180M (2024) https://doi.org/10.1117/12.3031759

KEYWORDS: Artificial intelligence, Standards development, Medical imaging, Systems modeling, Solid modeling, Reliability, Biomedical applications, Machine learning, Education and training, Decision support systems

Read Abstract +

Proceedings Article | 3 April 2024 Presentation

Use of natural language processing and semi-supervised machine learning to stratify STAT imaging studies by clinical urgency

Renaid Kim, L. Jordan Breyfogle, Benjamin Mervak, Lubomir Hadjiiski, Kenneth Buckwalter, Jessica Fried

Proceedings Volume 12931, 1293106 (2024) https://doi.org/10.1117/12.3005687

KEYWORDS: Machine learning, Education and training, Receivers

Read Abstract +

Proceedings Article | 3 April 2024 Poster + Paper

Large language model-assisted information extraction from clinical reports for survival prediction of bladder cancer patients

Di Sun, Lubomir Hadjiiski, John Gormley, Heang-Ping Chan, Elaine Caoili, Richard Cohan, Ajjai Alva, Rada Mihalcea, Chuan Zhou, Vikas Gulani

Proceedings Volume 12927, 129271V (2024) https://doi.org/10.1117/12.3008751

KEYWORDS: Deep learning, Radiomics, Bladder cancer, Chemotherapy, Data analysis

Read Abstract +

Proceedings Article | 7 April 2023 Presentation + Paper

Deep CNN task-based image quality assessment: application to digital breast tomosynthesis reconstruction and denoising

Mingjie Gao, Mark Helvie, Ravi Samala, Lubomir Hadjiiski, Jeffrey Fessler, Heang-Ping Chan

Proceedings Volume 12463, 1246319 (2023) https://doi.org/10.1117/12.2655419

KEYWORDS: Digital breast tomosynthesis, Image quality, Denoising, Human subjects, Reconstruction algorithms, Medical image reconstruction, Deep learning, 3D image reconstruction, Convolutional neural networks

Read Abstract +

Proceedings Article | 7 April 2023 Poster + Paper

Bladder cancer segmentation using U-Net-based deep-learning

Jonathan Clever, Lubomir Hadjiiski, Heang-Ping Chan, Richard Cohan, Elaine Caoili, Kenny Cha, Ravi Samala, Chuan Zhou

Proceedings Volume 12465, 1246532 (2023) https://doi.org/10.1117/12.2654650

KEYWORDS: Image segmentation, Bladder cancer, Education and training, Bladder, Performance modeling, 3D modeling, Deep learning, Cancer, Tumor growth modeling, Contour modeling

Read Abstract +

Proceedings Article | 7 April 2023 Poster + Paper

Bladder cancer treatment response assessment in CT urography by using deep-learning and radiomics

Di Sun, Lubomir Hadjiiski, Heang-Ping Chan, Richard Cohan, Elaine Caoili, Ajjai Alva, Kenny Cha, Ravi Samala, Chuan Zhou

Proceedings Volume 12465, 1246533 (2023) https://doi.org/10.1117/12.2654659

KEYWORDS: Radiomics, Bladder cancer, Education and training, Cancer, Chemotherapy, Oncology, Computed tomography, Artificial intelligence

Read Abstract +

Proceedings Article | 7 April 2023 Poster + Paper

Survival prediction for patients with metastatic urothelial cancer after immunotherapy using machine learning

Rain Tarango, Lubomir Hadjiiski, Ajjai Alva, Heang-Ping Chan, Richard Cohan, Elaine Caoili, Kenny Cha, Ravi Samala, Alon Weizer, Chuan Zhou, Monika Joshi, Yousef Zakharia

Proceedings Volume 12465, 124653A (2023) https://doi.org/10.1117/12.2655482

KEYWORDS: Tumors, Education and training, Machine learning, Cancer, Computed tomography, Artificial neural networks, Data modeling, Tumor growth modeling, Statistical analysis, Diseases and disorders

Read Abstract +

Proceedings Article | 7 April 2023 Poster + Paper

Segmentation of mouse tibia on MRI using deep learning U-Net models

Aman Kushwaha, Rami F. Mourad, Kevin Heist, Dariya Malyarenko, Heang-Ping Chan, Thomas Chenevert, Lubomir Hadjiiski

Proceedings Volume 12465, 124652O (2023) https://doi.org/10.1117/12.2654278

KEYWORDS: Image segmentation, Magnetic resonance imaging, Education and training, Bone, Performance modeling, Data modeling, Deep learning, 3D modeling, Animal model studies, Diseases and disorders

Read Abstract +

Proceedings Article | 4 April 2022 Paper

Fusion of multiple deep convolutional neural networks (DCNNs) for improved segmentation of lung nodules in CT images

Yifan Wang, Chuan Zhou, Heang-Ping Chan, Lubomir Hadjiiski, Aamer Chughtai

Proceedings Volume 12033, 1203329 (2022) https://doi.org/10.1117/12.2612360

KEYWORDS: Image segmentation, Lung, Image fusion, Computed tomography, Lung cancer, Computer programming, Convolutional neural networks, Databases, Cancer, Computer aided diagnosis and therapy

Read Abstract +

Proceedings Article | 4 April 2022 Presentation + Paper

Effect of computerized decision support on diagnostic accuracy and intra-observer variability in multi-institutional observer performance study for bladder cancer treatment response assessment in CT urography

Di Sun, Lubomir Hadjiiski, Rohan Garje, Yousef Zakharia, Lauren Pomerantz, Monika Joshi, Ajjai Alva, Heang-Ping Chan, Richard Cohan, Elaine Caoili, Kenny Cha, Galina Kirova-Nedyalkova, Matthew Davenport, Prasad Shankar, Isaac Francis, Kimberly Shampain, Nathaniel Meyer, Daniel Barkmeier, Sean Woolen, Phillip Palmbos, Alon Weizer, Ravi Samala, Chuan Zhou, Martha Matuszak

Proceedings Volume 12033, 120330G (2022) https://doi.org/10.1117/12.2611179

KEYWORDS: Bladder cancer, Cancer, Computed tomography, Statistical analysis, Diagnostics, Computer aided diagnosis and therapy, Computer aided design, Radiology, Decision support systems

Read Abstract +

Proceedings Article | 3 March 2021 Poster + Presentation + Paper

Deep learning based similarity-consistency abnormality detection (SCAD) model for classification of MRI patterns of multiple myeloma (MM) infiltration

Jinxin Zhou, Chuan Zhou, Heang-Ping Chan, Lubomir Hadjiiski, Qian Dong

Proceedings Volume 11595, 115954D (2021) https://doi.org/10.1117/12.2582041

KEYWORDS: Magnetic resonance imaging, Image classification, Computer programming, Gallium nitride

Read Abstract +

Proceedings Article | 3 March 2021 Presentation + Paper

Deep learning-based risk stratification for treatment management of multiple myeloma with sequential MRI scans

Chuan Zhou, Heang-Ping Chan, Lubomir Hadjiiski, Qian Dong

Proceedings Volume 11597, 1159716 (2021) https://doi.org/10.1117/12.2582203

KEYWORDS: Magnetic resonance imaging, Targeting Task Performance metric, Computer programming, Neural networks, Image classification, Convolutional neural networks, Bone

Read Abstract +

Proceedings Article | 3 March 2021 Presentation + Paper

Reinforced learning from serial CT to improve the early diagnosis of lung cancer in screening

Yifan Wang, Chuan Zhou, Lei Ying, Heang-Ping Chan, Lubomir Hadjiiski, Aamer Chughtai, Ella Kazerooni

Proceedings Volume 11597, 115971Q (2021) https://doi.org/10.1117/12.2582232

KEYWORDS: Lung cancer, Tumor growth modeling, Lung, Feature extraction, Computed tomography

Read Abstract +

Proceedings Article | 18 February 2021 Presentation + Paper

Severity assessment of COVID-19 using imaging descriptors: a deep-learning transfer learning approach from non-COVID-19 pneumonia

Ravi Samala, Lubomir Hadjiiski, Heang-Ping Chan, Chuan Zhou, Jadranka Stojanovska, Prachi Agarwal, Christopher Fung

Proceedings Volume 11597, 115971T (2021) https://doi.org/10.1117/12.2582115

KEYWORDS: Chest imaging, Image classification, Receivers, Diagnostics, Convolutional neural networks

Read Abstract +

Proceedings Article | 18 February 2021 Presentation + Paper

Multi-institutional observer performance study for bladder cancer treatment response assessment in CT urography with and without computerized decision support

Lubomir Hadjiiski, Monika Joshi, Ajjai Alva, Heang-Ping Chan, Richard Cohan, Elaine Caoili, Galina Kirova-Nedyalkova, Matthew Davenport, Prasad Shankar, Isaac Francis, Kenny Cha, Ravi Samala, Phillip Palmbos, Alon Weizer

Proceedings Volume 11597, 115971K (2021) https://doi.org/10.1117/12.2582331

KEYWORDS: Bladder cancer, Computed tomography, Oncology, Cancer, Human-machine interfaces, Decision support systems, Computing systems

Read Abstract +

Proceedings Article | 16 March 2020 Presentation + Paper

Explainable AI for medical imaging: deep-learning CNN ensemble for classification of estrogen receptor status from breast MRI

Zachary Papanastasopoulos, Ravi Samala, Heang-Ping Chan, Lubomir Hadjiiski, Chintana Paramagul, Mark Helvie, Colleen Neal

Proceedings Volume 11314, 113140Z (2020) https://doi.org/10.1117/12.2549298

KEYWORDS: Breast, Artificial intelligence, Magnetic resonance imaging, Tissues, Tumors, Medical imaging, Data modeling, Receptors, Performance modeling, Machine learning

Read Abstract +

Proceedings Article | 16 March 2020 Presentation + Paper

Convolutional neural network-based decision support system for bladder cancer staging in CT urography: decision threshold estimation and validation

Daniel Hoklai Chapman-Sung, Lubomir Hadjiiski, Dhanuj Gandikota, Heang-Ping Chan, Ravi Samala, Elaine Caoili, Richard Cohan, Alon Weizer, Ajjai Alva, Chuan Zhou

Proceedings Volume 11314, 113141T (2020) https://doi.org/10.1117/12.2551309

KEYWORDS: Bladder cancer, Cancer, Bladder, Neural networks, Decision support systems, Data modeling, Image segmentation, Statistical modeling, Computed tomography, Convolutional neural networks

Read Abstract +

Proceedings Article | 16 March 2020 Presentation + Paper

Hazards of data leakage in machine learning: a study on classification of breast cancer using deep neural networks

Ravi Samala, Heang-Ping Chan, Lubomir Hadjiiski, Sathvik Koneru

Proceedings Volume 11314, 1131416 (2020) https://doi.org/10.1117/12.2549313

KEYWORDS: Feature extraction, Machine learning, Feature selection, Mammography, Data modeling, Medical imaging, Digital mammography, Performance modeling, Breast cancer

Read Abstract +

Proceedings Article | 16 March 2020 Paper

Bladder wall segmentation using U-net based deep learning

Michael Ivanitskiy, Lubomir Hadjiiski, Heang-Ping Chan, Ravi Samala, Richard Cohan, Elaine Caoili, Alon Weizer, Ajjai Alva, Jun Wei, Chuan Zhou

Proceedings Volume 11314, 113143K (2020) https://doi.org/10.1117/12.2551343

KEYWORDS: Bladder, Image segmentation, Bladder cancer, Computed tomography, Convolution, Binary data, Tumor growth modeling, Neural networks, Cancer

Read Abstract +

Proceedings Article | 16 March 2020 Presentation + Paper

Hybrid deep-learning model for volume segmentation of lung nodules in CT images

Yifan Wang, Chuan Zhou, Heang-Ping Chan, Lubomir Hadjiiski, Jun Wei, Aamer Chughtai, Ella Kazerooni

Proceedings Volume 11314, 113141F (2020) https://doi.org/10.1117/12.2548994

KEYWORDS: Lung, Image segmentation, Lung cancer, Computed tomography, Volume rendering, Computer programming, Opacity, 3D modeling, Solids

Read Abstract +

Proceedings Article | 16 March 2020 Paper

Generating high resolution digital mammogram from digitized film mammogram with conditional generative adversarial network

Yuanpin Zhou, Jun Wei, Mark Helvie, Heang-ping Chan, Chuan Zhou, Lubomir Hadjiiski, Yao Lu

Proceedings Volume 11314, 1131426 (2020) https://doi.org/10.1117/12.2551278

KEYWORDS: Breast, Mammography, Image segmentation, Data modeling, Image resolution, Digital mammography, Breast cancer, Neural networks

Read Abstract +

Proceedings Article | 13 March 2019 Presentation + Paper

2D and 3D bladder segmentation using U-Net-based deep-learning

Xiangyuan Ma, Lubomir Hadjiiski, Jun Wei, Heang-Ping Chan, Kenny Cha, Richard Cohan, Elaine Caoili, Ravi Samala, Chuan Zhou, Yao Lu

Proceedings Volume 10950, 109500Y (2019) https://doi.org/10.1117/12.2511890

KEYWORDS: 3D modeling, Bladder, Image segmentation, Performance modeling, 3D image processing, Bladder cancer, Image resolution, Data modeling, Computed tomography, Computer aided diagnosis and therapy

Read Abstract +

Proceedings Article | 13 March 2019 Paper

Quantitative MRI biomarker for treatment response assessment of multiple myeloma: robustness evaluation using independent test set of prospective cases

Chuan Zhou, Qian Dong, Heang-Ping Chan, Erica Campagnaro, Jun Wei, Lubomir Hadjiiski

Proceedings Volume 10950, 1095047 (2019) https://doi.org/10.1117/12.2511898

KEYWORDS: Magnetic resonance imaging, Bone, Transform theory, Blood, Stem cells, Tumors, Transplantation, Scanners, Radiology, Plasma

Read Abstract +

Proceedings Article | 13 March 2019 Paper

Multi-path deep learning model for automated mammographic density categorization

Xiangyuan Ma, Caleb Fisher, Jun Wei, Mark Helvie, Heang-Ping Chan, Chuan Zhou, Lubomir Hadjiiski, Yao Lu

Proceedings Volume 10950, 109502E (2019) https://doi.org/10.1117/12.2511544

KEYWORDS: Breast, Mammography, Breast cancer, Convolutional neural networks, Data modeling, Lutetium, Cancer, Convolution, Neural networks, Radiology

Read Abstract +

Proceedings Article | 13 March 2019 Paper

Deep learning based bladder cancer treatment response assessment

Eric Wu, Lubomir Hadjiiski, Ravi Samala, Heang-Ping Chan, Kenny Cha, Caleb Richter, Richard Cohan, Elaine Caoili, Chintana Paramagul, Ajjai Alva, Alon Weizer

Proceedings Volume 10950, 109503D (2019) https://doi.org/10.1117/12.2512240

KEYWORDS: Bladder cancer, Computed tomography, Cancer, Image segmentation, Convolution, Bladder, Oncology, Performance modeling, Tumor growth modeling, Toxicity

Read Abstract +

Proceedings Article | 13 March 2019 Presentation + Paper

Analysis of deep convolutional features for detection of lung nodules in computed tomography

Ravi Samala, Heang-Ping Chan, Caleb Richter, Lubomir Hadjiiski, Chuan Zhou, Jun Wei

Proceedings Volume 10950, 109500Q (2019) https://doi.org/10.1117/12.2512208

KEYWORDS: Machine learning, Feature selection, Lung, Feature extraction, Computed tomography, Image segmentation, Convolutional neural networks, Image analysis, Computer-aided diagnosis

Read Abstract +

Proceedings Article | 13 March 2019 Presentation + Paper

Bladder cancer staging in CT urography: estimation and validation of decision thresholds for a radiomics-based decision support system

Dhanuj Gandikota, Lubomir Hadjiiski, Heang-Ping Chan, Kenny Cha, Ravi Samala, Elaine Caoili, Richard Cohan, Alon Weizer, Ajjai Alva, Chintana Paramagul, Jun Wei, Chuan Zhou

Proceedings Volume 10950, 109500W (2019) https://doi.org/10.1117/12.2513566

KEYWORDS: Bladder cancer, Cancer, Bladder, Decision support systems, Surface plasmons, Feature extraction, Data modeling, Neural networks, Pathology, Computed tomography, Machine learning, Image segmentation, Image classification

Read Abstract +

Proceedings Article | 13 March 2019 Presentation + Paper

Homogenization of breast MRI across imaging centers and feature analysis using unsupervised deep embedding

Ravi Samala, Heang-Ping Chan, Lubomir Hadjiiski, Chintana Paramagul, Mark Helvie, Colleen Neal

Proceedings Volume 10950, 1095020 (2019) https://doi.org/10.1117/12.2512209

KEYWORDS: Magnetic resonance imaging, Homogenization, Breast, Tumors, Machine learning, Tissues, Scanners, Mammography, Breast cancer, Data acquisition, Artificial intelligence

Read Abstract +

SPIE Journal Paper | 10 November 2018

PROSTATEx Challenges for computerized classification of prostate lesions from multiparametric magnetic resonance images

Samuel G. Armato, Henkjan Huisman, Karen Drukker, Lubomir Hadjiiski, Justin S. Kirby, Nicholas Petrick, George Redmond, Maryellen Giger, Kenny Cha, Artem Mamonov, Jayashree Kalpathy-Cramer, Keyvan Farahani

JMI, Vol. 5, Issue 04, 044501, (November 2018) https://doi.org/10.1117/12.10.1117/1.JMI.5.4.044501

KEYWORDS: Prostate, Magnetic resonance imaging, Cancer, Prostate cancer, Biopsy, Image classification, Magnetism, Image processing, Medical imaging, Medical research

Read Abstract +

Proceedings Article | 6 July 2018 Paper

Fully automated pectoral muscle identification on MLO-view mammograms with deep convolutional neural network

Xiangyuan Ma, Jun Wei, Chuan Zhou, Heang-Ping Chan, Lubomir Hadjiyski, Yao Lu

Proceedings Volume 10718, 1071818 (2018) https://doi.org/10.1117/12.2318124

KEYWORDS: Mammography, Design for manufacturing, Breast, Breast cancer, Convolutional neural networks, Image segmentation, Quantitative analysis, Binary data, Convolution, Medicine

Read Abstract +

Proceedings Article | 27 February 2018 Presentation + Paper

Bladder cancer treatment response assessment with radiomic, clinical, and radiologist semantic features

Marshall Gordon, Kenny Cha, Lubomir Hadjiiski, Heang-Ping Chan, Richard Cohan, Elaine Caoili, Chintana Paramagul, Ajjai Alva, Alon Weizer

Proceedings Volume 10575, 105751Y (2018) https://doi.org/10.1117/12.2294951

KEYWORDS: Bladder cancer, Cancer, Decision support systems, Computed tomography, Feature extraction, Classification systems, Machine learning, Pathology, Oncology, Feature selection

Read Abstract +

Proceedings Article | 27 February 2018 Paper

Computer-aided detection of bladder wall thickening in CT urography (CTU)

Kenny Cha, Lubomir Hadjiiski, Heang-Ping Chan, Elaine Caoili, Richard Cohan, Alon Weizer, Marshall Gordon, Ravi Samala

Proceedings Volume 10575, 1057529 (2018) https://doi.org/10.1117/12.2293844

KEYWORDS: Bladder, Image segmentation, Computer-aided diagnosis, Bladder cancer, Convolutional neural networks, Computer aided diagnosis and therapy, Computed tomography, Cancer

Read Abstract +

Proceedings Article | 27 February 2018 Paper

Bladder cancer staging in CT urography: effect of stage labels on statistical modeling of a decision support system

Dhanuj Gandikota, Lubomir Hadjiiski, Kenny Cha, Heang-Ping Chan, Elaine Caoili, Richard Cohan, Alon Weizer, Ajjai Alva, Chintana Paramagul, Jun Wei, Chuan Zhou

Proceedings Volume 10575, 105752B (2018) https://doi.org/10.1117/12.2295013

KEYWORDS: Bladder cancer, Decision support systems, Bladder, Statistical analysis, Feature extraction, Cancer, Data modeling, Computing systems, Image segmentation, Statistical modeling

Read Abstract +

Proceedings Article | 27 February 2018 Presentation + Paper

Generalization error analysis: deep convolutional neural network in mammography

Caleb Richter, Ravi Samala, Heang-Ping Chan, Lubomir Hadjiiski, Kenny Cha

Proceedings Volume 10575, 1057520 (2018) https://doi.org/10.1117/12.2292921

KEYWORDS: Mammography, Error analysis, Convolutional neural networks, Medical imaging, Image classification, Statistical analysis, Atomic force microscopy, Receivers, Image processing, Computer aided diagnosis and therapy

Read Abstract +

Proceedings Article | 27 February 2018 Presentation + Paper

Cross-domain and multi-task transfer learning of deep convolutional neural network for breast cancer diagnosis in digital breast tomosynthesis

Ravi Samala, Heang-Ping Chan, Lubomir Hadjiiski, Mark Helvie, Caleb Richter, Kenny Cha

Proceedings Volume 10575, 105750Q (2018) https://doi.org/10.1117/12.2293412

KEYWORDS: Digital breast tomosynthesis, Mammography, Convolutional neural networks, Breast cancer, Digital mammography, Computer aided diagnosis and therapy, Medical imaging, Machine learning

Read Abstract +

Proceedings Article | 27 February 2018 Presentation + Paper

Bladder cancer treatment response assessment in CT urography using two-channel deep-learning network

Kenny Cha, Lubomir Hadjiiski, Heang-Ping Chan, Ravi Samala, Richard Cohan, Elaine Caoili, Alon Weizer, Ajjai Alva

Proceedings Volume 10575, 105751V (2018) https://doi.org/10.1117/12.2292990

KEYWORDS: Bladder cancer, Computed tomography, Bladder, Image segmentation, Convolution, Neural networks, Cancer, Convolutional neural networks

Read Abstract +

Proceedings Article | 27 February 2018 Presentation + Paper

Compression of deep convolutional neural network for computer-aided diagnosis of masses in digital breast tomosynthesis

Ravi Samala, Heang-Ping Chan, Lubomir Hadjiiski, Mark Helvie, Caleb Richter, Kenny Cha

Proceedings Volume 10575, 1057521 (2018) https://doi.org/10.1117/12.2293400

KEYWORDS: Digital breast tomosynthesis, Mammography, Convolutional neural networks, Feature extraction, Computer aided diagnosis and therapy, Feature selection, Neurons, Neural networks, Data modeling, Atomic force microscopy

Read Abstract +

Proceedings Article | 27 February 2018 Presentation + Paper

Differentiating invasive and pre-invasive lung cancer by quantitative analysis of histopathologic images

Chuan Zhou, Hongliu Sun, Heang-Ping Chan, Aamer Chughtai, Jun Wei, Lubomir Hadjiiski, Ella Kazerooni

Proceedings Volume 10575, 105751H (2018) https://doi.org/10.1117/12.2293601

KEYWORDS: Tumors, Image segmentation, Lung cancer, Lung, Pathology, Quantitative analysis, Blob detection, Image classification, Feature extraction, Computed tomography

Read Abstract +

Proceedings Article | 27 February 2018 Paper

Deep convolutional neural network for mammographic density segmentation

Jun Wei, Songfeng Li, Heang-Ping Chan, Mark Helvie, Marilyn Roubidoux, Yao Lu, Chuan Zhou, Lubomir Hadjiiski, Ravi Samala

Proceedings Volume 10575, 105753I (2018) https://doi.org/10.1117/12.2293351

KEYWORDS: Breast, Image segmentation, Convolutional neural networks, Image processing, Mammography, Tissues, Machine learning, Computing systems, Feature extraction, Image enhancement

Read Abstract +

Proceedings Article | 3 March 2017 Presentation + Paper

Quantitative analysis of CT attenuation distribution patterns of nodule components for pathologic categorization of lung nodules

Chuan Zhou, Heang-Ping Chan, Jun Wei, Lubomir Hadjiiski, Aamer Chughtai, Ella Kazerooni

Proceedings Volume 10134, 1013422 (2017) https://doi.org/10.1117/12.2254155

KEYWORDS: Lung, Signal attenuation, Quantitative analysis, Computed tomography, Computer aided diagnosis and therapy, Machine vision, Computer vision technology, Image segmentation, Solids, Feature extraction, Lung cancer, Tumors, Tumor growth modeling, Expectation maximization algorithms

Read Abstract +

Proceedings Article | 3 March 2017 Presentation + Paper

Radiomics biomarkers for accurate tumor progression prediction of oropharyngeal cancer

Lubomir Hadjiiski, Heang-Ping Chan, Kenny Cha, Ashok Srinivasan, Jun Wei, Chuan Zhou, Mark Prince, Silvana Papagerakis

Proceedings Volume 10134, 101341Z (2017) https://doi.org/10.1117/12.2254973

KEYWORDS: Cancer, Feature extraction, Decision support systems, Head, Neck, Oncology, Statistical modeling, Machine learning, Image analysis, Tissues, Tumors, Computed tomography, Lymphatic system, Image segmentation

Read Abstract +

Proceedings Article | 3 March 2017 Presentation + Paper

Bladder cancer treatment response assessment using deep learning in CT with transfer learning

Kenny Cha, Lubomir Hadjiiski, Heang-Ping Chan, Ravi Samala, Richard Cohan, Elaine Caoili, Chintana Paramagul, Ajjai Alva, Alon Weizer

Proceedings Volume 10134, 1013404 (2017) https://doi.org/10.1117/12.2254977

KEYWORDS: Computer aided diagnosis and therapy, Toxicity, Computed tomography, Bladder, Bladder cancer, Image segmentation, CAD systems, Convolution, Neural networks, Machine learning

Read Abstract +

Proceedings Article | 3 March 2017 Presentation + Paper

Radiomic modeling of BI-RADS density categories

Jun Wei, Heang-Ping Chan, Mark Helvie, Marilyn Roubidoux, Chuan Zhou, Lubomir Hadjiiski

Proceedings Volume 10134, 101340P (2017) https://doi.org/10.1117/12.2255175

KEYWORDS: Breast cancer, Cancer, Mammography, Error analysis, Tumor growth modeling, X-rays, Machine learning, Breast, Digital mammography, Feature extraction

Read Abstract +

Proceedings Article | 3 March 2017 Presentation + Paper

Identifying key radiogenomic associations between DCE-MRI and micro-RNA expressions for breast cancer

Ravi Samala, Heang-Ping Chan, Lubomir Hadjiiski, Mark Helvie, Renaid Kim

Proceedings Volume 10134, 101340T (2017) https://doi.org/10.1117/12.2255512

KEYWORDS: Tumors, Magnetic resonance imaging, Data archive systems, Image segmentation, Receptors, Tumor growth modeling, Cancer, 3D acquisition, Breast cancer, Breast, 3D magnetic resonance imaging, Genomics, Proteins, Radiomics

Read Abstract +

Proceedings Article | 3 March 2017 Presentation + Paper

Segmentation of inner and outer bladder wall using deep-learning convolutional neural network in CT urography

Marshall Gordon, Lubomir Hadjiiski, Kenny Cha, Heang-Ping Chan, Ravi Samala, Richard Cohan, Elaine Caoili

Proceedings Volume 10134, 1013402 (2017) https://doi.org/10.1117/12.2255528

KEYWORDS: Convolutional neural networks, 3D imaging standards, Radiology, Cancer, Bladder, Neural networks, Computer aided diagnosis and therapy, Bladder cancer, Image segmentation, Machine learning

Read Abstract +

Proceedings Article | 3 March 2017 Presentation + Paper

Computer-aided detection of bladder masses in CT urography (CTU)

Kenny Cha, Lubomir Hadjiiski, Heang-Ping Chan, Elaine Caoili, Richard Cohan, Alon Weizer, Ravi Samala

Proceedings Volume 10134, 1013403 (2017) https://doi.org/10.1117/12.2255668

KEYWORDS: Convolutional neural networks, Computer aided design

Read Abstract +

Proceedings Article | 24 March 2016 Paper

First and second-order features for detection of masses in digital breast tomosynthesis

Ravi Samala, Jun Wei, Heang-Ping Chan, Lubomir Hadjiiski, Kenny Cha, Mark Helvie

Proceedings Volume 9785, 978523 (2016) https://doi.org/10.1117/12.2216327

KEYWORDS: Tomography, Digital breast tomosynthesis, Feature extraction, Computer-aided diagnosis, Breast, Feature selection, Machine learning, Breast cancer, Computer aided detection, Image classification, Pattern recognition, CAD systems, Computer aided diagnosis and therapy

Read Abstract +

Proceedings Article | 24 March 2016 Paper

Comparison of bladder segmentation using deep-learning convolutional neural network with and without level sets

Kenny Cha, Lubomir Hadjiiski, Ravi Samala, Heang-Ping Chan, Richard Cohan, Elaine Caoili

Proceedings Volume 9785, 978512 (2016) https://doi.org/10.1117/12.2216852

KEYWORDS: Convolutional neural networks, Bladder, Image segmentation, Bladder cancer, 3D imaging standards, CAD systems, Computer aided diagnosis and therapy, Computed tomography, Medical imaging

Read Abstract +

Proceedings Article | 24 March 2016 Paper

Computer-aided detection of bladder mass within non-contrast-enhanced region of CT Urography (CTU)

Kenny Cha, Lubomir Hadjiiski, Heang-Ping Chan, Elaine Caoili, Richard Cohan, Alon Weizer, Chuan Zhou

Proceedings Volume 9785, 97853W (2016) https://doi.org/10.1117/12.2216289

KEYWORDS: Bladder, Image segmentation, Computer-aided diagnosis, Feature extraction, Computer aided diagnosis and therapy, Computing systems, Bladder cancer, Neural networks, Convolutional neural networks, Computer aided design, Feature selection

Read Abstract +

Proceedings Article | 24 March 2016 Paper

Automatic detection of ureter lesions in CT urography

Trevor Exell, Lubomir Hadjiiski, Heang-Ping Chan, Kenny Cha, Elaine Caoili, Richard Cohan, Jun Wei, Chuan Zhou

Proceedings Volume 9785, 97851O (2016) https://doi.org/10.1117/12.2216866

KEYWORDS: CAD systems, Image segmentation, Feature extraction, Computer-aided diagnosis, Image analysis, Automatic tracking, Image filtering, Cancer, Bladder, Radiology, Computer aided diagnosis and therapy

Read Abstract +

Proceedings Article | 24 March 2016 Paper

Automatic staging of bladder cancer on CT urography

Sankeerth Garapati, Lubomir Hadjiiski, Kenny Cha, Heang-Ping Chan, Elaine Caoili, Richard Cohan, Alon Weizer, Ajjai Alva, Chintana Paramagul, Jun Wei, Chuan Zhou

Proceedings Volume 9785, 97851G (2016) https://doi.org/10.1117/12.2217004

KEYWORDS: Bladder cancer, Computed tomography, Feature extraction, Machine learning, Cancer, Computer aided diagnosis and therapy, Decision support systems, Tumor growth modeling, Bladder, Image segmentation

Read Abstract +

Proceedings Article | 24 March 2016 Paper

Deep-learning convolution neural network for computer-aided detection of microcalcifications in digital breast tomosynthesis

Ravi Samala, Heang-Ping Chan, Lubomir Hadjiiski, Kenny Cha, Mark Helvie

Proceedings Volume 9785, 97850Y (2016) https://doi.org/10.1117/12.2217092

KEYWORDS: Convolution, Neural networks, Computer-aided diagnosis, Digital breast tomosynthesis, Reconstruction algorithms, Computing systems, Digital imaging, Image filtering, CAD systems, Machine learning, Pattern recognition, Digital mammography, Computer aided detection, Computer architecture, Computer aided diagnosis and therapy

Read Abstract +

Proceedings Article | 24 March 2016 Paper

Reference state estimation of breast computed tomography for registration with digital mammography

Ravi Samala, Heang-Ping Chan, Lubomir Hadjiiski, Ruola Ning, Kenny Cha, Mark Helvie

Proceedings Volume 9785, 97851U (2016) https://doi.org/10.1117/12.2217080

KEYWORDS: Image registration, Breast, Finite element methods, Computed tomography, Digital mammography, Iterative methods, Image processing, Mammography, Computer simulations

Read Abstract +

Proceedings Article | 24 March 2016 Paper

Automated identification of best-quality coronary artery segments from multiple-phase coronary CT angiography (cCTA) for vessel analysis

Chuan Zhou, Heang-Ping Chan, Lubomir Hadjiiski, Aamer Chughtai, Jun Wei, Ella Kazerooni

Proceedings Volume 9785, 978503 (2016) https://doi.org/10.1117/12.2217261

KEYWORDS: Image segmentation, Arteries, Visualization, Angiography, Computer-aided diagnosis, Machine vision, Computer vision technology, Automatic tracking, Feature extraction, Computer aided design, Heart, Computer aided diagnosis and therapy, Optical tracking, Electroluminescence

Read Abstract +

Proceedings Article | 24 March 2016 Paper

Computerized flow and vessel wall analyses of coronary arteries for detection of non-calcified plaques in coronary CT angiography

Jun Wei, Chuan Zhou, Heang-Ping Chan, Aamer Chughtai, Prachi Agarwal, Lubomir Hadjiiski, Ella Kazerooni

Proceedings Volume 9785, 97853U (2016) https://doi.org/10.1117/12.2216815

KEYWORDS: Computing systems, Blood circulation, Signal attenuation, Feature extraction, Angiography, Arteries, Computer-aided diagnosis, Computer aided design, Computer aided diagnosis and therapy, Tomography, Virtual colonoscopy, Image segmentation, Heart

Read Abstract +

SPIE Journal Paper | 25 June 2015 Open Access

Special Section Guest Editorial: LUNGx Challenge for computerized lung nodule classification: reflections and lessons learned

Samuel Armato, Lubomir Hadjiiski, Georgia Tourassi, Karen Drukker, Maryellen Giger, Feng Li, George Redmond, Keyvan Farahani, Justin S. Kirby, Laurence Clarke

JMI, Vol. 2, Issue 02, 020103, (June 2015) https://doi.org/10.1117/12.10.1117/1.JMI.2.2.020103

KEYWORDS: Medical imaging, Lung, Cancer, Calibration, Computed tomography, Oncology, Lung cancer, Databases, Algorithm development, Radiology

DOWNLOAD PAPER

Proceedings Article | 20 March 2015 Paper

Digital breast tomosynthesis: application of 2D digital mammography CAD to detection of microcalcification clusters on planar projection image

Ravi Samala, Heang-Ping Chan, Yao Lu, Lubomir Hadjiiski, Jun Wei, Mark Helvie

Proceedings Volume 9414, 941418 (2015) https://doi.org/10.1117/12.2076291

KEYWORDS: Digital breast tomosynthesis, Computer aided diagnosis and therapy, CAD systems, Digital mammography, Digital imaging, Statistical analysis, Computer aided design, Image enhancement, Image segmentation

Read Abstract +

Proceedings Article | 20 March 2015 Paper

Comparison of computer-aided detection of clustered microcalcifications in digital mammography and digital breast tomosynthesis

Ravi Samala, Heang-Ping Chan, Yao Lu, Lubomir Hadjiiski, Jun Wei, Mark Helvie

Proceedings Volume 9414, 94140K (2015) https://doi.org/10.1117/12.2081874

KEYWORDS: Digital breast tomosynthesis, CAD systems, Computer aided design, Digital mammography, Computer aided diagnosis and therapy, Breast, Computing systems, Convolution, Neural networks

Read Abstract +

Proceedings Article | 20 March 2015 Paper

Robustness evaluation of a computer-aided detection system for pulmonary embolism (PE) in CTPA using independent test set from multiple institutions

Chuan Zhou, Heang-Ping Chan, Aamer Chughtai, Jean Kuriakose, Ella Kazerooni, Lubomir Hadjiiski, Jun Wei, Smita Patel

Proceedings Volume 9414, 941408 (2015) https://doi.org/10.1117/12.2082015

KEYWORDS: CAD systems, Arteries, Image segmentation, Computer aided diagnosis and therapy, Computing systems, Feature extraction, Computed tomography, Lung, Standards development

Read Abstract +

Proceedings Article | 20 March 2015 Paper

Quantitative analysis of arterial flow properties for detection of non-calcified plaques in ECG-gated coronary CT angiography

Jun Wei, Chuan Zhou, Heang-Ping Chan, Aamer Chughtai, Prachi Agarwal, Jean Kuriakose, Lubomir Hadjiiski, Smita Patel, Ella Kazerooni

Proceedings Volume 9414, 94141H (2015) https://doi.org/10.1117/12.2082367

KEYWORDS: Arteries, Angiography, Blood circulation, Quantitative analysis, Heart, Computed tomography, Copper, Computer aided diagnosis and therapy, Computing systems

Read Abstract +

Proceedings Article | 20 March 2015 Paper

Computer-aided detection of bladder mass within contrast-enhanced region of CTU

Kenny Cha, Lubomir Hadjiiski, Heang-Ping Chan, Elaine Caoili, Richard Cohan, Chuan Zhou

Proceedings Volume 9414, 94141Q (2015) https://doi.org/10.1117/12.2081472

KEYWORDS: Bladder, Image segmentation, Computing systems, Bladder cancer, Computer aided design, Feature extraction, Computer aided diagnosis and therapy, Feature selection, CAD systems

Read Abstract +

Proceedings Article | 20 March 2015 Paper

Ureter segmentation in CT urography (CTU) by COMPASS with multiscale Hessian enhancement

Duncan Fairbanks, Lubomir Hadjiiski, Chuan Zhou, Heang-Ping Chan, Richard Cohan, Elaine Caoili, Kenny Cha

Proceedings Volume 9414, 94141S (2015) https://doi.org/10.1117/12.2082639

KEYWORDS: Image segmentation, Error analysis, Bladder, Solid modeling, Cancer, Tissues, Feature extraction, 3D image processing, Image filtering

Read Abstract +

Proceedings Article | 20 March 2015 Paper

Automatic selection of best quality vessels from multiple-phase coronary CT angiography (cCTA)

Jordan Liu, Lubomir Hadjiiski, Heang-Ping Chan, Chuan Zhou, Jun Wei, Aamer Chughtai, Jean Kuriakose, Prachi Agarwal, Ella Kazerooni

Proceedings Volume 9414, 94140E (2015) https://doi.org/10.1117/12.2082637

KEYWORDS: Visualization, Angiography, Image quality, Quality measurement, Arteries, Computer aided diagnosis and therapy, Composites, Image segmentation, CAD systems, Radiology

Read Abstract +

Proceedings Article | 24 March 2014 Paper

Automated identification of spinal cord and vertebras on sagittal MRI

Chuan Zhou, Heang-Ping Chan, Qian Dong, Bo He, Jun Wei, Lubomir Hadjiiski, Daniel Couriel

Proceedings Volume 9035, 903515 (2014) https://doi.org/10.1117/12.2043500

KEYWORDS: Spinal cord, Magnetic resonance imaging, Image segmentation, Image registration, Image filtering, Bone, Feature extraction, Image analysis, Magnetism, 3D image enhancement

Read Abstract +

Proceedings Article | 24 March 2014 Paper

Computerized luminal analysis for detection of non-calcified plaques in coronary CT angiography

Jun Wei, Chuan Zhou, Heang-Ping Chan, Aamer Chughtai, Smita Patel, Prachi Agarwal, Jean Kuriakose, Lubomir Hadjiiski, Ella Kazerooni

Proceedings Volume 9035, 90350R (2014) https://doi.org/10.1117/12.2043664

KEYWORDS: Arteries, Feature extraction, Angiography, Statistical analysis, Machine learning, Feature selection, Receivers, Computer aided diagnosis and therapy, Detector development, Heart

Read Abstract +

Proceedings Article | 18 March 2014 Paper

COMPASS-based ureter segmentation in CT urography (CTU)

David Zick, Lubomir Hadjiiyski, Heang-Ping Chan, Richard Cohan, Elaine Caoili, Chuan Zhou, Jun Wei

Proceedings Volume 9035, 90352S (2014) https://doi.org/10.1117/12.2042262

KEYWORDS: Image segmentation, Computing systems, Cancer, Bladder, Solid modeling, Computer aided diagnosis and therapy, Feature extraction, Computed tomography, Visualization, CAD systems

Read Abstract +

Proceedings Article | 18 March 2014 Paper

Segmentation of urinary bladder in CT urography (CTU) using CLASS with enhanced contour conjoint procedure

Kenny Cha, Lubomir Hadjiiski, Heang-Ping Chan, Richard Cohan, Elaine Caoili, Chuan Zhou

Proceedings Volume 9035, 90352P (2014) https://doi.org/10.1117/12.2042270

KEYWORDS: Bladder, Image segmentation, Bladder cancer, 3D imaging standards, Computer aided diagnosis and therapy, Computing systems, Error analysis, Radiology, 3D modeling, Computed tomography

Read Abstract +

Proceedings Article | 18 March 2014 Paper

Comparison of CLASS and ITK-SNAP in segmentation of urinary bladder in CT urography

Kenny Cha, Lubomir Hadjiiski, Heang-Ping Chan, Elaine Caoili, Richard Cohan, Chuan Zhou

Proceedings Volume 9035, 90351F (2014) https://doi.org/10.1117/12.2042272

KEYWORDS: Bladder, Image segmentation, Bladder cancer, Optical spheres, Medical imaging, 3D imaging standards, Computer aided diagnosis and therapy, Error analysis, 3D image processing, Radiology

Read Abstract +

Proceedings Article | 18 March 2014 Paper

Surgical retained foreign object (RFO) prevention by computer aided detection (CAD)

Theodore Marentis, Lubomir Hadjiiyski, Amrita Chaudhury, Lucas Rondon, Nikolaos Chronis, Heang-Ping Chan

Proceedings Volume 9035, 903529 (2014) https://doi.org/10.1117/12.2042469

KEYWORDS: Computer aided diagnosis and therapy, Radiography, Surgery, CAD systems, Computer aided design, Visibility, Computing systems, X-rays, Medical imaging, Abdomen

Read Abstract +

Proceedings Article | 18 March 2014 Paper

Digital breast tomosynthesis: effects of projection-view distribution on computer-aided detection of microcalcification clusters

Ravi Samala, Heang-Ping Chan, Yao Lu, Lubomir Hadjiiski, Jun Wei, Mark Helvie

Proceedings Volume 9035, 90350Y (2014) https://doi.org/10.1117/12.2043513

KEYWORDS: Digital breast tomosynthesis, Signal to noise ratio, Photovoltaics, Tomography, Reconstruction algorithms, Computer aided diagnosis and therapy, Computing systems, Computer aided design, Breast

Read Abstract +

Proceedings Article | 1 April 2013 Paper

Multiscale intensity homogeneity transformation method and its application to computer-aided detection of pulmonary embolism in computed tomographic pulmonary angiography (CTPA)

Yanhui Guo, Chuan Zhou, Heang-Ping Chan, Jun Wei, Aamer Chughtai, Baskaran Sundaram, Lubomir Hadjiiski, Smita Patel, Ella Kazerooni

Proceedings Volume 8670, 867036 (2013) https://doi.org/10.1117/12.2008053

KEYWORDS: CAD systems, Tomography, Angiography, Computer aided design, Statistical analysis, Computer aided diagnosis and therapy, Feature selection, Radiology, Medical imaging

Read Abstract +

Proceedings Article | 29 March 2013 Paper

Curved planar reformation and optimal path tracing (CROP) method for false positive reduction in computer-aided detection of pulmonary embolism in CTPA

Chuan Zhou, Heang-Ping Chan, Yanhui Guo, Jun Wei, Aamer Chughtai, Lubomir Hadjiiski, Baskaran Sundaram, Smita Patel, Jean Kuriakose, Ella Kazerooni

Proceedings Volume 8670, 867035 (2013) https://doi.org/10.1117/12.2008048

KEYWORDS: Feature extraction, Visualization, Image segmentation, CAD systems, Lung, Arteries, Computer aided diagnosis and therapy, Tissues, Electroluminescence

Read Abstract +

The curved planar reformation (CPR) method re-samples the vascular structures along the vessel centerline to generate longitudinal cross-section views. The CPR technique has been commonly used in coronary CTA workstation to facilitate radiologists’ visual assessment of coronary diseases, but has not yet been used for pulmonary vessel analysis in CTPA due to the complicated tree structures and the vast network of pulmonary vasculature. In this study, a new curved planar reformation and optimal path tracing (CROP) method was developed to facilitate feature extraction and false positive (FP) reduction and improve our PE detection system. PE candidates are first identified in the segmented pulmonary vessels at prescreening. Based on Dijkstra’s algorithm, the optimal path (OP) is traced from the pulmonary trunk bifurcation point to each PE candidate. The traced vessel is then straightened and a reformatted volume is generated using CPR. Eleven new features that characterize the intensity, gradient, and topology are extracted from the PE candidate in the CPR volume and combined with the previously developed 9 features to form a new feature space for FP classification. With IRB approval, CTPA of 59 PE cases were retrospectively collected from our patient files (UM set) and 69 PE cases from the PIOPED II data set with access permission. 595 and 800 PEs were manually marked by experienced radiologists as reference standard for the UM and PIOPED set, respectively. At a test sensitivity of 80%, the average FP rate was improved from 18.9 to 11.9 FPs/case with the new method for the PIOPED set when the UM set was used for training. The FP rate was improved from 22.6 to 14.2 FPs/case for the UM set when the PIOPED set was used for training. The improvement in the free response receiver operating characteristic (FROC) curves was statistically significant (p<0.05) by JAFROC analysis, indicating that the new features extracted from the CROP method are useful for FP reduction.

Proceedings Article | 29 March 2013 Paper

Computerized segmentation of ureters in CT urography (CTU) using COMPASS

Lubomir Hadjiiski, Heang-Ping Chan, Luke Niland, Richard Cohan, Elaine Caoili, Chuan Zhou, Jun Wei

Proceedings Volume 8670, 86703B (2013) https://doi.org/10.1117/12.2008244

KEYWORDS: Image segmentation, Cancer, Computed tomography, Bladder, Computing systems, Computer aided diagnosis and therapy, Solid modeling, Feature extraction, Visualization, Kidney

Read Abstract +

Proceedings Article | 29 March 2013 Paper

Automatic identification of origins of left and right coronary arteries in CT angiography for coronary arterial tree tracking and plaque detection

Chuan Zhou, Heang-Ping Chan, Aamer Chightai, Jun Wei, Lubomir Hadjiiski, Prachi Agarwal, Jean Kuriakose, Ella Kazerooni

Proceedings Volume 8670, 86703L (2013) https://doi.org/10.1117/12.2008046

KEYWORDS: Arteries, Heart, Automatic tracking, Image segmentation, Angiography, Visualization, Structural design, Optical tracking, Computer aided diagnosis and therapy

Read Abstract +

Proceedings Article | 29 March 2013 Paper

Automated registration of coronary arterial trees from multiple phases in coronary CT angiography (cCTA)

Lubomir Hadjiiski, Chuan Zhou, Heang-Ping Chan, Aamer Chughtai, Prachi Agarwal, Jean Kuriakose, Smita Patel, Jun Wei, Ella Kazerooni

Proceedings Volume 8670, 86703M (2013) https://doi.org/10.1117/12.2008058

KEYWORDS: Angiography, Computer aided diagnosis and therapy, Arteries, Composites, Computer aided design, Image registration, Radiology, Computed tomography, Heart

Read Abstract +

Proceedings Article | 19 March 2013 Paper

Study of image quality in digital breast tomosynthesis by subpixel reconstruction

Yao Lu, Heang-Ping Chan, Jun Wei, Lubomir Hadjiiski, Ravi Samala

Proceedings Volume 8668, 86680I (2013) https://doi.org/10.1117/12.2008069

KEYWORDS: Digital breast tomosynthesis, Image quality, Optical spheres, Sensors, Image acquisition, Computer simulations, Image resolution, Human subjects, Breast, X-rays

Read Abstract +

Proceedings Article | 18 March 2013 Paper

Computerized detection of non-calcified plaques in coronary CT angiography: topological soft-gradient detection method for plaque prescreening

Jun Wei, Chuan Zhou, Heang-Ping Chan, Aamer Chughtai, Smita Patel, Prachi Agarwal, Jean Kuriakose, Lubomir Hadjiiski, Ella Kazerooni

Proceedings Volume 8670, 867016 (2013) https://doi.org/10.1117/12.2008047

KEYWORDS: Arteries, Angiography, Detector development, Machine learning, Heart, Computer aided diagnosis and therapy, Computed tomography, Image segmentation, 3D modeling

Read Abstract +

Proceedings Article | 18 March 2013 Paper

Detection of microcalcifications in breast tomosynthesis reconstructed with multiscale bilateral filtering regularization

Ravi Samala, Heang-Ping Chan, Yao Lu, Lubomir Hadjiyski, Jun Wei, Berkman Sahiner, Mark Helvie

Proceedings Volume 8670, 86701L (2013) https://doi.org/10.1117/12.2008230

KEYWORDS: Digital breast tomosynthesis, CAD systems, Breast, Signal to noise ratio, Reconstruction algorithms, Mammography, Electronic filtering, Visibility, Breast cancer

Read Abstract +

Proceedings Article | 26 February 2013 Paper

Neural network training by maximization of the area under the ROC curve: application to characterization of masses on breast ultrasound as malignant or benign

Berkman Sahiner, Xin He, Weijie Chen, Heang-Ping Chan, Lubomir Hadjiiski, Nicholas Petrick

Proceedings Volume 8670, 86701M (2013) https://doi.org/10.1117/12.2007615

KEYWORDS: Neural networks, Breast, Ultrasonography, Artificial neural networks, Computer aided design, Computer aided diagnosis and therapy, Image segmentation, Evolutionary algorithms, Error analysis, Data acquisition

Read Abstract +

Back-propagation neural networks (BPNs) are traditionally trained using error measures such as sum-of-squares or cross-entropy. If the training sample size is small, and the neural network has a large number of hidden layer nodes, the BPN may be overtrained, i.e., it may fit the training data well, but may generalize poorly to independent test data. In this study, we investigated a training technique that maximized the approximate area under the ROC curve (AUC) to reduce overtraining. In general, the non-parametric AUC is a discontinuous and non-differentiable function of the neural network output, which makes it unsuitable for gradient descent algorithms such as back-propagation. We used a semidifferentiable approximation to AUC, which appeared to provide reasonable training for the data sets explored in this study. We performed a simulation study using synthetic data sets consisting of Gaussian mixtures to investigate the behavior of this new technique with respect to overtraining. Our results indicated that an artificial neural network trained using the AUC-maximization method is less prone to overtraining. The advantage of the AUC-maximization method was consistently observed over different values of hidden layer BPN nodes, training sample sizes, and the dimensionality of the feature spaces evaluated in our simulation study. For a five-hidden-node BPN trained using 50 training samples per class, the average test AUC was 0.896 (standard deviation (SD): 0.026) with AUC-maximization and 0.856 (SD: 0.028) with the sum-of-squares method. The gain in test performance by the AUC-maximization method over the traditional BPN training was greater when the training sample size was smaller. We also applied this new method to a data set previously acquired for characterization of masses on breast ultrasound as malignant or benign. Our results with this real-world data set had the same trend as with our simulation data sets in that the AUC-maximization technique was less prone to overtraining than the sum-of-squares method.

Proceedings Article | 3 March 2012 Paper

Multiscale regularized reconstruction for enhancing microcalcification in digital breast tomosynthesis

Yao Lu, Heang-Ping Chan, Jun Wei, Lubomir Hadjiiski, Chuan Zhou

Proceedings Volume 8313, 831322 (2012) https://doi.org/10.1117/12.912573

KEYWORDS: Digital breast tomosynthesis, Image quality, Tissues, Sensors, Breast, Reconstruction algorithms, X-ray sources, Image analysis, Human subjects, Image enhancement

Read Abstract +

Proceedings Article | 23 February 2012 Paper

Computer-aided detection of microcalcifications in digital breast tomosynthesis (DBT): a multichannel signal detection approach on projection views

Jun Wei, Heang-Ping Chan, Lubomir Hadjiiski, Mark Helvie, Chuan Zhou, Yao Lu

Proceedings Volume 8315, 83152D (2012) https://doi.org/10.1117/12.912501

KEYWORDS: Digital breast tomosynthesis, Signal detection, Computer aided diagnosis and therapy, Photovoltaics, Breast, 3D image processing, Mammography, Electronic filtering, Prototyping

Read Abstract +

Proceedings Article | 23 February 2012 Paper

Hyun-Chong Cho, Lubomir Hadjiiski, Heang-Ping Chan, Berkman Sahiner, Mark Helvie, Chintana Paramagul, Alexis Nees

Proceedings Volume 8315, 831528 (2012) https://doi.org/10.1117/12.912364

KEYWORDS: Feature extraction, Computer aided diagnosis and therapy, Breast, Ultrasonography, Content based image retrieval, Visualization, Distance measurement, Diagnostics, Pathology, Image segmentation

Read Abstract +

Proceedings Article | 23 February 2012 Paper

Automatic seed point identification and main artery segmentation for pulmonary vascular tree segmentation and tracking in computed tomographic pulmonary angiography (CTPA)

Yanhui Guo, Chuan Zhou, Heang-Ping Chan, Jean Kuriakose, Aamer Chughtai, Jun Wei, Lubomir Hadjiiski, Ella Kazerooni

Proceedings Volume 8315, 83152P (2012) https://doi.org/10.1117/12.912848

KEYWORDS: Arteries, Image segmentation, Heart, Automatic tracking, Tomography, Angiography, Lung, Computer aided diagnosis and therapy, Chest

Read Abstract +

Proceedings Article | 23 February 2012 Paper

Segmentation of urinary bladder in CT Urography (CTU) using CLASS

Lubomir Hadjiiski, Heang-Ping Chan, Yuen Law, Richard Cohan, Elaine Caoili, Hyun-Chong Cho, Chuan Zhou, Jun Wei

Proceedings Volume 8315, 83150J (2012) https://doi.org/10.1117/12.912847

KEYWORDS: Bladder, Image segmentation, Bladder cancer, Computing systems, Computer aided diagnosis and therapy, 3D image processing, Computed tomography, Cancer, Visualization, Image filtering

Read Abstract +

Proceedings Article | 23 February 2012 Paper

Pulmonary vessel segmentation utilizing curved planar reformation and optimal path finding (CROP) in computed tomographic pulmonary angiography (CTPA) for CAD applications

Chuan Zhou, Heang-Ping Chan, Jean Kuriakose, Aamer Chughtai, Jun Wei, Lubomir Hadjiiski, Yanhui Guo, Smita Patel, Ella Kazerooni

Proceedings Volume 8315, 83150N (2012) https://doi.org/10.1117/12.912446

KEYWORDS: Image segmentation, Lung, Computer aided diagnosis and therapy, Tissues, Angiography, Arteries, Tomography, Lymphatic system, Matrices, Radiology

Read Abstract +

Vessel segmentation is a fundamental step in an automated pulmonary embolism (PE) detection system. The purpose of this study is to improve the segmentation scheme for pulmonary vessels affected by PE and other lung diseases. We have developed a multiscale hierarchical vessel enhancement and segmentation (MHES) method for pulmonary vessel tree extraction based on the analysis of eigenvalues of Hessian matrices. However, it is difficult to segment the pulmonary vessels accurately under suboptimal conditions, such as vessels occluded by PEs, surrounded by lymphoid tissues or lung diseases, and crossing with other vessels. In this study, we developed a new vessel refinement method utilizing curved planar reformation (CPR) technique combined with optimal path finding method (MHES-CROP). The MHES segmented vessels straightened in the CPR volume was refined using adaptive gray level thresholding where the local threshold was obtained from least-square estimation of a spline curve fitted to the gray levels of the vessel along the straightened volume. An optimal path finding method based on Dijkstra's algorithm was finally used to trace the correct path for the vessel of interest. Two and eight CTPA scans were randomly selected as training and test data sets, respectively. Forty volumes of interest (VOIs) containing "representative" vessels were manually segmented by a radiologist experienced in CTPA interpretation and used as reference standard. The results show that, for the 32 test VOIs, the average percentage volume error relative to the reference standard was improved from 32.9±10.2% using the MHES method to 9.9±7.9% using the MHES-CROP method. The accuracy of vessel segmentation was improved significantly (p<0.05). The intraclass correlation coefficient (ICC) of the segmented vessel volume between the automated segmentation and the reference standard was improved from 0.919 to 0.988. Quantitative comparison of the MHES method and the MHES-CROP method with the reference standard was also evaluated by the Bland-Altman plot. This preliminary study indicates that the MHES-CROP method has the potential to improve PE detection.

Proceedings Article | 23 February 2012 Paper

Interactive content-based image retrieval (CBIR) computer-aided diagnosis (CADx) system for ultrasound breast masses using relevance feedback

Hyun-chong Cho, Lubomir Hadjiiski, Berkman Sahiner, Heang-Ping Chan, Chintana Paramagul, Mark Helvie, Alexis Nees

Proceedings Volume 8315, 831509 (2012) https://doi.org/10.1117/12.912164

KEYWORDS: Computer aided diagnosis and therapy, Feature extraction, Computing systems, Ultrasonography, Breast, Content based image retrieval, Visualization, Distance measurement, Pathology, Image segmentation

Read Abstract +

Proceedings Article | 17 March 2011 Paper

Adaptive diffusion regularization for enhancement of microcalcifications in digital breast tomosynthesis (DBT) reconstruction

Yao Lu, Heang-Ping Chan, Jeffrey Fessler, Lubomir Hadjiiski, Jun Wei, Mitchell Goodsitt

Proceedings Volume 7961, 796117 (2011) https://doi.org/10.1117/12.878096

KEYWORDS: Digital breast tomosynthesis, Reconstruction algorithms, Diffusion, Interference (communication), Breast, Sensors, Signal to noise ratio, Human subjects, Radiology, Mammography

Read Abstract +

Proceedings Article | 16 March 2011 Paper

Hyun-chong Cho, Lubomir Hadjiiski, Berkman Sahiner, Heang-Ping Chan, Mark Helvie, Alexis Nees, Chintana Paramagul

Proceedings Volume 7963, 79632F (2011) https://doi.org/10.1117/12.878117

KEYWORDS: Computer aided diagnosis and therapy, Breast, Feature extraction, Ultrasonography, Biopsy, Image segmentation, Content based image retrieval, Visualization, Distance measurement, Neural networks

Read Abstract +

Proceedings Article | 16 March 2011 Paper

Computer-aided detection of breast masses in digital breast tomosynthesis (DBT): improvement of false positive reduction by optimization of object segmentation

Jun Wei, Heang-Ping Chan, Berkman Sahiner, Lubomir Hadjiiski, Mark Helvie, Chuan Zhou, Yao Lu

Proceedings Volume 7963, 796311 (2011) https://doi.org/10.1117/12.878214

KEYWORDS: Digital breast tomosynthesis, Breast, Image segmentation, Reconstruction algorithms, Feature extraction, Mammography, Breast cancer, Computing systems, Prototyping

Read Abstract +

Proceedings Article | 16 March 2011 Paper

Computerized detection of pulmonary embolism in computed tomographic pulmonary angiography (CTPA): improvement of vessel segmentation

Chuan Zhou, Heang-Ping Chan, Jean Kuriakose, Aamer Chughtai, Lubomir Hadjiiski, Jun Wei, Smita Patel, Ella Kazerooni

Proceedings Volume 7963, 79630L (2011) https://doi.org/10.1117/12.878156

KEYWORDS: Image segmentation, Lung, Lymphatic system, Tissues, Angiography, Arteries, Mathematical modeling, Tomography, Computer aided diagnosis and therapy, Matrices

Read Abstract +

Proceedings Article | 9 March 2011 Paper

BI-RADS guided mammographic mass retrieval

Yimo Tao, Shih-Chung Lo, Lubomir Hadjiski, Heang-Ping Chan, Matthew Freedman

Proceedings Volume 7963, 79632H (2011) https://doi.org/10.1117/12.878274

KEYWORDS: Image retrieval, Feature extraction, Mammography, Visualization, Computer aided diagnosis and therapy, Computing systems, Image segmentation, Content based image retrieval, Biological research, Biopsy

Read Abstract +

Proceedings Article | 9 March 2011 Paper

Study of adaptability of breast density analysis system developed for screen film mammograms (SFMs) to full-field digital mammograms (FFDMs): robustness of parenchymal texture analysis

Jun Wei, Heang-Ping Chan, Mark Helvie, Chuan Zhou, Lubomir Hadjiiski

Proceedings Volume 7963, 796320 (2011) https://doi.org/10.1117/12.878235

KEYWORDS: Atomic force microscopy, Mammography, Breast cancer, Breast, Nipple, Digital mammography, Computing systems, Feature extraction, Computer aided diagnosis and therapy, Image analysis

Read Abstract +

Proceedings Article | 3 March 2011 Paper

Analysis of the number of distinct findings obtained by multiple readers in an MRMC study: When do findings obtained from the addition of new readers become redundant, or otherwise negligible?

Sophie Paquerault, Berkman Sahiner, Anna Kettermann, Laura Yarusso, Lubomir Hadjiiski, Heang-Ping Chan

Proceedings Volume 7966, 79661M (2011) https://doi.org/10.1117/12.878103

KEYWORDS: Data modeling, CAD systems, Lung, Computed tomography, Computer aided diagnosis and therapy, Computer aided design, Computing systems, Mathematical modeling, Solid modeling, Computer simulations

Read Abstract +

The ultimate goal of this project is to investigate whether the effect of a computer-aided detection (CAD) system on readers' performance (especially, in situation of an upgrade of the CAD system, or between two different CAD systems with similar design) can be accurately predicted without having to perform a multi-reader multi-case (MRMC) observer study and, if such prediction is possible, to establish the underlying methodology. Our current study is intended to provide evidence that would substantiate efforts toward such investigation. The objectives of this study were 1) to investigate the relationship between the number of radiologists reading a dataset of thoracic computed tomography (CT) images to identify lung nodules and the number of distinct findings and 2) to determine the number of readers needed to identify almost all clinically distinct findings in a dataset. We used data from a multi-reader multi-case (MRMC) observer study that consisted of six radiologists interpreting 85 thoracic CT examinations. To further illustrate our approach, we also utilized simulated data consisting of twelve readers interpreting 198 samples equally distributed between three levels of detection difficulty. For each possible reader grouping, the number of distinct findings identified by the readers in the group was calculated. Five types of regression models used to describe the relationship between the average number of distinct findings per case and the number of readers needed were compared. The result showed that the logistic model best fitted both the thoracic CT data and the simulated data. Our assumption is that adding more readers after a certain reader set size would mostly add redundant findings and, therefore, the benefit would be negligible. Using this model, the predicted number of readers was found to depend on the type of findings considered. Our study showed that the number of clinically distinct findings that can be identified by radiologists on CT lung examinations without the use of a CAD system may be limited and that identifying almost all of these findings may only require a limited number of readers.

Proceedings Article | 19 March 2010 Paper

Effects of projection-view distributions on image quality of calcifications in digital breast tomosynthesis (DBT) reconstruction

Yao Lu, Heang-Ping Chan, Mitch Goodsitt, Jun Wei, Lubomir Hadjiiski, Andrea Schmitz, Jeffrey Eberhard, Bernhard E. Claus

Proceedings Volume 7622, 76220D (2010) https://doi.org/10.1117/12.844481

KEYWORDS: Digital breast tomosynthesis, Photovoltaics, Image quality, Breast, X-ray sources, Reconstruction algorithms, Sensors, Tomography, Imaging systems, Field emission displays

Read Abstract +

Proceedings Article | 9 March 2010 Paper

Digital breast tomosynthesis: feasibility of automated detection of microcalcification clusters on projections views

Lubomir Hadjiiski, Heang-Ping Chan, Jun Wei, Berkman Sahiner, Chuan Zhou, Mark Helvie

Proceedings Volume 7624, 76241F (2010) https://doi.org/10.1117/12.844627

KEYWORDS: Digital breast tomosynthesis, Breast, Photovoltaics, Signal to noise ratio, Computer aided diagnosis and therapy, Breast cancer, Mammography, Algorithm development, Image filtering

Read Abstract +

Proceedings Article | 9 March 2010 Paper

Automated segmentation and tracking of coronary arteries in cardiac CT scans: comparison of performance with a clinically used commercial software

Chuan Zhou, Heang-Ping Chan, Aamer Chughtai, Smita Patel, Lubomir Hadjiiski, Berkman Sahiner, Jun Wei, Ella Kazerooni

Proceedings Volume 7624, 76240O (2010) https://doi.org/10.1117/12.844391

KEYWORDS: Arteries, Image segmentation, Detection and tracking algorithms, Visualization, Optical spheres, Heart, Digital breast tomosynthesis, Computed tomography, Computer aided diagnosis and therapy, Automatic tracking

Read Abstract +

Coronary CT angiography (cCTA) has been reported to be an effective means for diagnosis of coronary artery disease. We are investigating the feasibility of developing a computer-aided detection (CADe) system to assist radiologists in detection of non-calcified plaques in coronary arteries in ECG-gated cCTA scans. In this study, we developed a prototype vessel segmentation and tracking method to extract the coronary arterial trees which will define the search space for plaque detection. Vascular structures are first enhanced by 3D multi-scale filtering and analysis of the eigenvalues of Hessian matrices using a vessel enhancement response function specifically designed for coronary arteries. The enhanced vascular structures are then segmented by an EM estimation method. The segmented coronary arteries are tracked using a 3D dynamic balloon tracking (DBT) method. For this preliminary study, two starting seed points were manually identified at the origins of the left and right coronary artery (LCA and RCA). The DBT method automatically moves a sphere along the vessel whose diameter is adjusted dynamically based on the local vessel size, tracks the vessels, and identifies its branches to generate the left and right coronary arterial trees. The algorithm was applied to 20 cCTA scans that contained various degrees of coronary artery diseases. To evaluate the performance of vessel segmentation and tracking, the rendered volume of coronary arteries tracked by our algorithm was displayed on a PC, placed next to a GE Advantage workstation on which the coronary arterial trees tracked by the GE software and the original cCTA scan were displayed. Two experienced thoracic radiologists visually examined the coronary arteries on the cCTA scan and the segmented vessels to count untracked false-negative (FN) segments and false positives (FPs). The comparison was made by radiologists' visual judgment because the digital files for the segmented vessels were not accessible on the commercial system. A total of 19 and 38 artery segments were identified to be FNs, and 23 FPs and 20 FPs were found in the coronary trees tracked by our algorithm and the GE software, respectively. The preliminary results demonstrated the feasibility of our approach.

Proceedings Article | 9 March 2010 Paper

Digital breast tomosynthesis: computerized detection of microcalcifications in reconstructed breast volume using a 3D approach

Heang-Ping Chan, Berkman Sahiner, Jun Wei, Lubomir Hadjiiski, Chuan Zhou, Mark Helvie

Proceedings Volume 7624, 76241D (2010) https://doi.org/10.1117/12.844511

KEYWORDS: Digital breast tomosynthesis, Breast, Mammography, CAD systems, 3D image processing, Signal to noise ratio, Photovoltaics, Prototyping, Computer aided diagnosis and therapy

Read Abstract +

Proceedings Article | 9 March 2010 Paper

Association of a mammographic parenchymal pattern (MPP) descriptor with breast cancer risk: a case-control study

Jun Wei, Heang-Ping Chan, Chuan Zhou, Mark Helvie, Lubomir Hadjiiski, Berkman Sahiner

Proceedings Volume 7624, 76240A (2010) https://doi.org/10.1117/12.844040

KEYWORDS: Breast cancer, Breast, Mammography, Cancer, Nipple, Statistical analysis, Received signal strength, Feature extraction, Image segmentation, Feature selection

Read Abstract +

Proceedings Article | 9 March 2010 Paper

Interobserver variability effects on computerized volume analysis of treatment response of head and neck lesions in CT

Lubomir Hadjiiski, Heang-Ping Chan, Mohannad Ibrahim, Berkman Sahiner, Sachin Gujar, Suresh Mukherji

Proceedings Volume 7624, 76243G (2010) https://doi.org/10.1117/12.844643

KEYWORDS: Image segmentation, Tumors, Head, Neck, Computed tomography, Computing systems, Cancer, Statistical analysis, Tissues, 3D modeling

Read Abstract +

A computerized system for segmenting lesions in head and neck CT scans was developed to assist radiologists in estimation of the response to treatment of malignant lesions. The system performs 3D segmentation based on a level set model and uses as input an approximate bounding box for the lesion of interest. We investigated the effect of the interobserver variability of radiologists' marking of the bounding box on the automatic segmentation performance. In this preliminary study, CT scans from a pre-treatment exam and a post one-cycle chemotherapy exam of 34 patients with primary site head and neck neoplasms were used. For each tumor, an experienced radiologist marked the lesion with a bounding box and provided a reference standard by outlining the full 3D contour on both the pre- and post treatment scans. A second radiologist independently marked each tumor again with another bounding box. The correlation between the automatic and manual estimates for both the pre-to-post-treatment volume change and the percent volume change was r=0.95. Based on the bounding boxes by the second radiologist, the correlation between the automatic and manual estimate for the pre-to-post-treatment volume change was r=0.89 and for the percent volume change was r=0.91. The correlation for the automatic estimates obtained from the bounding boxes by the two radiologists was as follows: (1) pretreatment volume r=0.92, (2) post-treatment volume r=0.88, (3) pre-to-post-treatment change r=0.89 and (4) percent preto- post-treatment change r=0.90. The difference between the automatic estimates based on the two sets of bounding boxes did not achieve statistical significance for any of the estimates (p>0.29). The preliminary results indicate that the automated segmentation system can reliably estimate tumor size change in response to treatment relative to radiologist's hand segmentation as reference standard, and that the performance was robust against inter-observer variability in marking the input bounding boxes.

Proceedings Article | 10 March 2009 Paper

Inter- and intra-observer variability in radiologists' assessment of mass similarity on mammograms

Berkman Sahiner, Lubomir Hadjiiski, Heang-Ping Chan, Jing Cui, Chintana Paramagul, Alexis Nees, Mark Helvie

Proceedings Volume 7263, 726315 (2009) https://doi.org/10.1117/12.813425

KEYWORDS: Mammography, Image retrieval, Breast, Biopsy, Computer aided diagnosis and therapy, Radiology, Feature extraction, Human-machine interfaces, Medical imaging, Decision support systems

Read Abstract +

Proceedings Article | 4 March 2009 Paper

Comparison of breast parenchymal pattern on prior mammograms of breast cancer patients and normal subjects

Jun Wei, Heang-Ping Chan, Chuan Zhou, Yi-ta Wu, Berkman Sahiner, Lubomir Hadjiiski, Mark Helvie

Proceedings Volume 7260, 72602H (2009) https://doi.org/10.1117/12.813565

KEYWORDS: Breast, Mammography, Cancer, Breast cancer, Statistical analysis, Feature selection, Feature extraction, Received signal strength, Tissues, Receivers

Read Abstract +

Proceedings Article | 3 March 2009 Paper

A computer-aided diagnosis system for prediction of the probability of malignancy of breast masses on ultrasound images

Jing Cui, Berkman Sahiner, Heang-ping Chan, Jiazheng Shi, Alexis Nees, Chintana Paramagul, Lubomir Hadjiiski

Proceedings Volume 7260, 72600L (2009) https://doi.org/10.1117/12.813722

KEYWORDS: Phase modulation, Computer aided diagnosis and therapy, Lawrencium, Image segmentation, Computing systems, Breast, Classification systems, Ultrasonography, Feature extraction, Image classification

Read Abstract +

Proceedings Article | 3 March 2009 Paper

Evaluation of computerized detection of pulmonary embolism in independent data sets of computed tomographic pulmonary angiographic (CTPA) scans

Chuan Zhou, Heang-Ping Chan, Berkman Sahiner, Lubomir Hadjiiski, Aamer Chughtai, Smita Patel, Jun Wei, Philip Cascade, Ella Kazerooni

Proceedings Volume 7260, 72600B (2009) https://doi.org/10.1117/12.813811

KEYWORDS: Image segmentation, Arteries, Computer aided diagnosis and therapy, Tissues, Tomography, Angiography, CAD systems, Lymphatic system, Feature selection, Feature extraction

Read Abstract +

Proceedings Article | 3 March 2009 Paper

Comparison of computerized mass detection in digital breast tomosynthesis (DBT) mammograms and conventional mammograms

Heang-Ping Chan, Jun Wei, Berkman Sahiner, Lubomir Hadjiiski, Mark Helvie

Proceedings Volume 7260, 72600S (2009) https://doi.org/10.1117/12.813851

KEYWORDS: Digital breast tomosynthesis, Mammography, CAD systems, 3D image processing, Atomic force microscopy, Breast, Photovoltaics, Feature extraction, Prototyping, Image segmentation

Read Abstract +

Proceedings Article | 28 February 2009 Paper

Automated segmentation of urinary bladder and detection of bladder lesions in multi-detector row CT urography

Lubomir Hadjiiski, Berkman Sahiner, Heang-Ping Chan, Elaine Caoili, Richard Cohan, Chuan Zhou

Proceedings Volume 7260, 72603R (2009) https://doi.org/10.1117/12.813864

KEYWORDS: Bladder, Image segmentation, Bladder cancer, CAD systems, Shape analysis, Computed tomography, Computer aided diagnosis and therapy, Cancer, Radiology, Optical tracking

Read Abstract +

Proceedings Article | 19 March 2008 Paper

Digital tomosynthesis mammography: improvement of artifact reduction method for high-attenuation objects on reconstructed slices

Jun Ge, Heang-Ping Chan, Berkman Sahiner, Yiheng Zhang, Jun Wei, Lubomir Hadjiiski, Chuan Zhou, Yi-Ta Wu, Jiazheng Shi

Proceedings Volume 6913, 69134O (2008) https://doi.org/10.1117/12.771539

KEYWORDS: Digital mammography, Imaging systems, Modulation transfer functions, Sensors, X-ray sources, X-rays, Breast, Data modeling, Visualization, X-ray imaging

Read Abstract +

Proceedings Article | 18 March 2008 Paper

Truncation artifact and boundary artifact reduction in breast tomosynthesis reconstruction

Yiheng Zhang, Heang-Ping Chan, Yi-Ta Wu, Berkman Sahiner, Chuan Zhou, Jun Wei, Jun Ge, Lubomir Hadjiiski, Jiazheng Shi

Proceedings Volume 6913, 69132Y (2008) https://doi.org/10.1117/12.769698

KEYWORDS: Breast, Photovoltaics, Sensors, Image quality, Digital mammography, Tissues, Reconstruction algorithms, Image processing, Signal attenuation, Mammography

Read Abstract +

Proceedings Article | 17 March 2008 Paper

Concordance of computer-extracted image features with BI-RADS descriptors for mammographic mass margin

Berkman Sahiner, Lubomir Hadjiiski, Heang-Ping Chan, Chintana Paramagul, Alexis Nees, Mark Helvie, Jiazheng Shi

Proceedings Volume 6915, 69151N (2008) https://doi.org/10.1117/12.770752

KEYWORDS: Mammography, Image segmentation, Feature extraction, Breast, Computer aided diagnosis and therapy, Computing systems, Radiology, Receivers, Image analysis, Medical imaging

Read Abstract +

The purpose of this study was to develop and evaluate computer-extracted features for characterizing mammographic mass margins according to BI-RADS spiculated and circumscribed categories. The mass was automatically segmented using an active contour model. A spiculation measure for a pixel on the mass boundary was defined by using the angular difference between the image gradient vector and the normal to the mass, averaged over pixels in a spiculation search region. For the circumscribed margin feature, the angular difference between the principal eigenvector of the Hessian matrix and the normal to the mass was estimated in a band of pixels centered at each point on the boundary, and the feature was extracted from the resulting profile along the boundary. Three MQSA radiologists provided BI-RADS margin ratings for a data set of 198 regions of interest containing breast masses. The features were evaluated with respect to the individual radiologists' characterization using receiver operating characteristic (ROC) analysis, as well as with respect to that from the majority rule, in which a mass was labeled as spiculated (circumscribed) if it was characterized as such by 2 or 3 radiologists, and non-spiculated (non-circumscribed) otherwise. We also investigated the performance of the features for consensus masses, defined as those labeled as spiculated (circumscribed) or nonspiculated (non-circumscribed) by all three radiologists. When masses were labeled according to radiologists R1, R2, and R3 individually, the spiculation feature had an area A_z under the ROC curve of 0.90±0.04, 0.90±0.03, 0.88±0.03, respectively, while the circumscribed margin feature had an A_z value of 0.77±0.04, 0.74±0.04, and 0.80±0.03, respectively. When masses were labeled according to the majority rule, the A_z values for the spiculation and the circumscribed margin features were 0.92±0.03 and 0.80±±0.03, respectively. When only the consensus masses were considered, the A_z values for the spiculation and the circumscribed margin features were 0.96±0.04 and 0.87±0.04, respectively. We conclude that the newly developed features had high accuracy for characterizing mass margins according to BI-RADS descriptors.

Proceedings Article | 17 March 2008 Paper

Breast mass segmentation on dynamic contrast-enhanced magnetic resonance scans using the level set method

Jiazheng Shi, Berkman Sahiner, Heang-Ping Chan, Chintana Paramagul, Lubomir Hadjiiski, Mark Helvie, Yi-Ta Wu, Jun Ge, Yiheng Zhang, Chuan Zhou, Jun Wei

Proceedings Volume 6915, 69152A (2008) https://doi.org/10.1117/12.771390

KEYWORDS: Image segmentation, Breast, Magnetic resonance imaging, 3D image processing, Magnetism, Breast cancer, 3D acquisition, 3D scanning, 3D modeling, Mathematical modeling

Read Abstract +

Proceedings Article | 17 March 2008 Paper

Design and evaluation of a new automated method for the segmentation and characterization of masses on ultrasound images

Jing Cui, Berkman Sahiner, Heang-Ping Chan, Alexis Nees, Chintana Paramagul, Lubomir Hadjiiski, Chuan Zhou, Jiazheng Shi

Proceedings Volume 6915, 69150H (2008) https://doi.org/10.1117/12.770300

KEYWORDS: Image segmentation, Feature extraction, Computer aided diagnosis and therapy, Ultrasonography, Breast, Computer aided design, Image classification, Computing systems, Feature selection, Receivers

Read Abstract +

Segmentation of masses is the first step in most computer-aided diagnosis (CAD) systems for characterization of breast masses as malignant or benign. In this study, we designed an automated method for segmentation of masses on ultrasound (US) images. The method automatically estimated an initial contour based on a manually-identified point approximately at the mass center. A two-stage active contour (AC) method iteratively refined the initial contour and performed self-examination and correction on the segmentation result. To evaluate our method, we compared it with manual segmentation by an experienced radiologists (R1) on a data set of 226 US images containing biopsy-proven masses from 121 patients (44 malignant and 77 benign). Four performance measures were used to evaluate the segmentation accuracy; two measures were related to the overlap between the computer and radiologist segmentation, and two were related to the area difference between the two segmentation results. To compare the difference between the segmentation results by the computer and R1 to inter-observer variation, a second radiologist (R2) also manually segmented all masses. The two overlap measures between the segmentation results by the computer and R1 were 0.87+ 0.16 and 0.73+ 0.17 respectively, indicating a high agreement. However, the segmentation results between two radiologists were more consistent. To evaluate the effect of the segmentation method on classification accuracy, three feature spaces were formed by extracting texture, width-to-height, and posterior shadowing features using the computer segmentation, R1's manual segmentation, and R2's manual segmentation. A linear discriminant analysis classifier using stepwise feature selection was tested and trained by a leave-one-case-out method to characterize the masses as malignant or benign. For case-based classification, the area A_z under the test receiver operating characteristic (ROC) curve was 0.90±0.03, 0.87±0.03 and 0.87±0.03 for the feature sets based on computer segmentation, R1's manual segmentation, and R2's manual segmentation, respectively.

Proceedings Article | 17 March 2008 Paper

Automated segmentation and tracking of coronary arteries in ECG-gated cardiac CT scans

Chuan Zhou, Heang-Ping Chan, Aamer Chughtai, Smita Patel, Prachi Agarwal, Lubomir Hadjiiski, Berkman Sahiner, Jun Wei, Jun Ge, Ella Kazerooni

Proceedings Volume 6915, 69150O (2008) https://doi.org/10.1117/12.770362

KEYWORDS: Arteries, Image segmentation, Heart, Computed tomography, Optical spheres, Visualization, Optical tracking, Expectation maximization algorithms, Blood, Image enhancement

Read Abstract +

Proceedings Article | 17 March 2008 Paper

Characterization of posterior acoustic features of breast masses on ultrasound images using artificial neural network

Jing Cui, Berkman Sahiner, Heang-Ping Chan, Chintana Paramagul, Alexis Nees, Lubomir Hadjiiski, Yi-Ta Wu

Proceedings Volume 6915, 691521 (2008) https://doi.org/10.1117/12.771401

KEYWORDS: Acoustics, Breast, Computing systems, Image segmentation, Ultrasonography, Artificial neural networks, Feature extraction, Computer aided diagnosis and therapy, Image classification, Computer aided design

Read Abstract +

Proceedings Article | 17 March 2008 Paper

Digital tomosynthesis mammography: comparison of mass classification using 3D slices and 2D projection views

Heang-Ping Chan, Yi-Ta Wu, Berkman Sahiner, Yiheng Zhang, Jun Wei, Richard Moore, Daniel Kopans, Mark Helvie, Lubomir Hadjiiski, Ted Way

Proceedings Volume 6915, 691506 (2008) https://doi.org/10.1117/12.770422

KEYWORDS: Digital breast tomosynthesis, Photovoltaics, Feature extraction, Digital mammography, Image segmentation, Computer aided diagnosis and therapy, Breast, Mammography, Reconstruction algorithms, Feature selection

Read Abstract +

Proceedings Article | 17 March 2008 Paper

Automated detection of breast vascular calcification on full-field digital mammograms

Jun Ge, Heang-Ping Chan, Berkman Sahiner, Chuan Zhou, Mark Helvie, Jun Wei, Lubomir Hadjiiski, Yiheng Zhang, Yi-Ta Wu, Jiazheng Shi

Proceedings Volume 6915, 691517 (2008) https://doi.org/10.1117/12.773096

KEYWORDS: Image segmentation, Computer aided diagnosis and therapy, Mammography, Breast, Algorithm development, Blood vessels, Feature selection, Computing systems, Arteries

Read Abstract +

Breast vascular calcifications (BVCs) are calcifications that line the blood vessel walls in the breast and appear as parallel or tubular tracks on mammograms. BVC is one of the major causes of the false positive (FP) marks from computer-aided detection (CADe) systems for screening mammography. With the detection of BVCs and the calcified vessels identified, these FP clusters can be excluded. Moreover, recent studies reported the increasing interests in the correlation between mammographically visible BVCs and the risk of coronary artery diseases. In this study, we developed an automated BVC detection method based on microcalcification prescreening and a new k-segments clustering algorithm. The mammogram is first processed with a difference-image filtering technique designed to enhance calcifications. The calcification candidates are selected by an iterative process that combines global thresholding and local thresholding. A new k-segments clustering algorithm is then used to find a set of line segments that may be caused by the presence of calcified vessels. A linear discriminant analysis (LDA) classifier was designed to reduce false segments that are not associated with BVCs. Four features for each segment selected with stepwise feature selection were used for this LDA classification. Finally, the neighboring segments were linked and dilated with morphological dilation to cover the regions of calcified vessels. A data set of 16 FFDM cases with vascular calcifications was collected for this preliminary study. Our preliminary result demonstrated that breast vascular calcifications can be accurately detected and the calcified vessels identified. It was found that the automated method can achieve a detection sensitivity of 65%, 70%, and 75% at 6.1 mm, 8.4 mm, and 12.6mm FP segments/image, respectively, without any true clustered microcalcifications being falsely marked. Further work is underway to improve this method and to incorporate it into our FFDM CADe system.

Proceedings Article | 17 March 2008 Paper

Comparison of mammographic parenchymal patterns of normal subjects and breast cancer patients

Yi-Ta Wu, Berkman Sahiner, Heang-Ping Chan, Jun Wei, Lubomir Hadjiiski, Mark Helvie, Yiheng Zhang, Jiazheng Shi, Chuan Zhou, Jun Ge, Jing Cui

Proceedings Volume 6915, 691520 (2008) https://doi.org/10.1117/12.771278

KEYWORDS: Breast, Mammography, Breast cancer, Cancer, Image segmentation, Received signal strength, Feature extraction, Statistical analysis, Computer aided diagnosis and therapy, Computing systems

Read Abstract +

Proceedings Article | 17 March 2008 Paper

Automated detection of ureteral wall thickening on multi-detector row CT urography

Lubomir Hadjiiski, Berkman Sahiner, Elaine Caoili, Richard Cohan, Chuan Zhou, Heang-Ping Chan

Proceedings Volume 6915, 69152U (2008) https://doi.org/10.1117/12.771325

KEYWORDS: CAD systems, Bladder, Cancer, Computer aided diagnosis and therapy, Computed tomography, Tissues, Image segmentation, Biopsy, Visualization

Read Abstract +

Proceedings Article | 17 March 2008 Paper

Classification of breast masses and normal tissues in digital tomosynthesis mammography

Jun Wei, Heang-Ping Chan, Yiheng Zhang, Berkman Sahiner, Chuan Zhou, Jun Ge, Yi-Ta Wu, Lubomir Hadjiiski

Proceedings Volume 6915, 691508 (2008) https://doi.org/10.1117/12.771189

KEYWORDS: Breast, Digital mammography, Tissues, Statistical analysis, Computer aided diagnosis and therapy, Reconstruction algorithms, Feature extraction, Breast cancer, Photovoltaics, Image segmentation

Read Abstract +

Proceedings Article | 17 March 2008 Paper

Volume analysis of treatment response of head and neck lesions using 3D level set segmentation

Lubomir Hadjiiski, Ethan Street, Berkman Sahiner, Sachin Gujar, Mohannad Ibrahim, Heang-Ping Chan, Suresh Mukherji

Proceedings Volume 6915, 691512 (2008) https://doi.org/10.1117/12.771314

KEYWORDS: Image segmentation, Head, Neck, Computed tomography, Cancer, Tumors, Computing systems, 3D modeling, Image filtering, Lymphatic system

Read Abstract +

Proceedings Article | 5 April 2007 Paper

Pulmonary nodule registration in serial CT scans using rib anatomy and nodule template matching

Jiazheng Shi, Berkman Sahiner, Heang-Ping Chan, Lubomir Hadjiiski, Chuan Zhou, Yi-Ta Wu, Jun Wei

Proceedings Volume 6514, 65140R (2007) https://doi.org/10.1117/12.713409

KEYWORDS: Computed tomography, Lung, Image registration, 3D acquisition, Image segmentation, Spine, Expectation maximization algorithms, Bone, Detection and tracking algorithms, Error analysis

Read Abstract +

Proceedings Article | 31 March 2007 Paper

The effect of nodule segmentation on the accuracy of computerized lung nodule detection on CT scans: comparison on a data set annotated by multiple radiologists

Berkman Sahiner, Lubomir Hadjiiski, Heang-Ping Chan, Jiazheng Shi, Ted Way, Philip Cascade, Ella Kazerooni, Chuan Zhou, Jun Wei

Proceedings Volume 6514, 65140L (2007) https://doi.org/10.1117/12.713771

KEYWORDS: Roentgenium, Lung, Computed tomography, Image segmentation, Feature extraction, CAD systems, Computing systems, Space operations, 3D modeling, Detection and tracking algorithms

Read Abstract +

Proceedings Article | 30 March 2007 Paper

Automated detection of pulmonary embolism (PE) in computed tomographic pulmonary angiographic (CTPA) images: multiscale hierachical expectation-maximization segmentation of vessels and PEs

Chuan Zhou, Heang-Ping Chan, Lubomir Hadjiiski, Aamer Chughtai, Smita Patel, Philip Cascade, Berkman Sahiner, Jun Wei, Jun Ge, Ella Kazerooni

Proceedings Volume 6514, 65142F (2007) https://doi.org/10.1117/12.713769

KEYWORDS: Image segmentation, Arteries, Lung, Tissues, 3D image enhancement, Angiography, Computer aided diagnosis and therapy, 3D image processing, Chest, Expectation maximization algorithms

Read Abstract +

Proceedings Article | 29 March 2007 Paper

Computer-aided detection of masses in digital tomosynthesis mammography: combination of 3D and 2D detection information

Heang-Ping Chan, Jun Wei, Yiheng Zhang, Richard Moore, Daniel Kopans, Lubomir Hadjiiski, Berkman Sahiner, Marilyn Roubidoux, Mark Helvie

Proceedings Volume 6514, 651416 (2007) https://doi.org/10.1117/12.713640

KEYWORDS: Digital breast tomosynthesis, CAD systems, Photovoltaics, 3D image processing, Breast, Digital mammography, Mammography, Feature extraction, Image processing

Read Abstract +

Proceedings Article | 29 March 2007 Paper

Computer-aided detection of breast masses on prior mammograms

Jun Wei, Berkman Sahiner, Heang-Ping Chan, Lubomir Hadjiiski, Marilyn Roubidoux, Mark Helvie, Jun Ge, Chuan Zhou, Yi-Ta Wu

Proceedings Volume 6514, 651405 (2007) https://doi.org/10.1117/12.713768

KEYWORDS: Mammography, CAD systems, Computer aided design, Statistical analysis, Cancer, Computer aided diagnosis and therapy, Visibility, Breast, Feature extraction, Feature selection

Read Abstract +

Proceedings Article | 29 March 2007 Paper

Computer-aided diagnosis for interval change analysis of lung nodule features in serial CT examinations

Lubomir Hadjiiski, Ted Way, Berkman Sahiner, Heang-Ping Chan, Philip Cascade, Naama Bogot, Ella Kazerooni, Chuan Zhou

Proceedings Volume 6514, 651411 (2007) https://doi.org/10.1117/12.713770

KEYWORDS: Lung, Computed tomography, Feature extraction, Computer aided diagnosis and therapy, Lung cancer, Feature selection, CAD systems, 3D modeling, Classification systems, Computer aided design

Read Abstract +

Proceedings Article | 20 March 2007 Paper

Effect of CAD on radiologists’ detection of lung nodules on thoracic CT scans: observer performance study

Berkman Sahiner, Lubomir Hadjiiski, Heang Chan, Jiazheng Shi, Philip Cascade, Ella Kazerooni, Chuan Zhou, Jun Wei, Aamer Chughtai, Chad Poopat, Thomas Song, Jadranka Nojkova, Luba Frank, Anil Attili

Proceedings Volume 6515, 65151D (2007) https://doi.org/10.1117/12.709851

KEYWORDS: Computer aided diagnosis and therapy, Computing systems, Lung, Computed tomography, Computer aided design, Radiology, Statistical analysis, Gold, Lung imaging, Databases

Read Abstract +

Proceedings Article | 15 March 2007 Paper

Investigation of the Z-axis resolution of breast tomosynthesis mammography systems

Yiheng Zhang, Heang-Ping Chan, Berkman Sahiner, Jun Wei, Jun Ge, Lubomir Hadjiiski, Chuan Zhou

Proceedings Volume 6510, 65104A (2007) https://doi.org/10.1117/12.713816

KEYWORDS: Breast, Mammography, Reconstruction algorithms, Imaging systems, Digital mammography, Photovoltaics, Image resolution, Computer simulations, Image acquisition, Breast cancer

Read Abstract +

Proceedings Article | 8 March 2007 Paper

A dynamic multiple thresholding method for automated breast boundary detection in digitized mammograms

Yi-Ta Wu, Chuan Zhou, Lubomir Hadjiiski, Jiazheng Shi, Jun Wei, Chintana Paramagul, Berkman Sahiner, Heang-Ping Chan

Proceedings Volume 6512, 65122U (2007) https://doi.org/10.1117/12.710198

KEYWORDS: Breast, Mammography, Bragg cells, Image segmentation, Detection and tracking algorithms, Image analysis, Image filtering, Distance measurement, Radiology, Image processing algorithms and systems

Read Abstract +

Proceedings Article | 1 March 2007 Paper

Digital tomosynthesis mammography: intra- and interplane artifact reduction for high-contrast objects on reconstructed slices using a priori 3D geometrical information

Jun Ge, Heang-Ping Chan, Berkman Sahiner, Yiheng Zhang, Jun Wei, Lubomir Hadjiiski, Chuan Zhou

Proceedings Volume 6512, 65124Q (2007) https://doi.org/10.1117/12.708914

KEYWORDS: Digital mammography, Breast, Reconstruction algorithms, Signal to noise ratio, 3D image processing, Sensors, Metals, Iterative methods, Prototyping, Mammography

Read Abstract +

We are developing a computerized technique to reduce intra- and interplane ghosting artifacts caused by high-contrast objects such as dense microcalcifications (MCs) or metal markers on the reconstructed slices of digital tomosynthesis mammography (DTM). In this study, we designed a constrained iterative artifact reduction method based on a priori 3D information of individual MCs. We first segmented individual MCs on projection views (PVs) using an automated MC detection system. The centroid and the contrast profile of the individual MCs in the 3D breast volume were estimated from the backprojection of the segmented individual MCs on high-resolution (0.1 mm isotropic voxel size) reconstructed DTM slices. An isolated volume of interest (VOI) containing one or a few MCs is then modeled as a high-contrast object embedded in a local homogeneous background. A shift-variant 3D impulse response matrix (IRM) of the projection-reconstruction (PR) system for the extracted VOI was calculated using the DTM geometry and the reconstruction algorithm. The PR system for this VOI is characterized by a system of linear equations. A constrained iterative method was used to solve these equations for the effective linear attenuation coefficients (eLACs) within the isolated VOI. Spatial constraint and positivity constraint were used in this method. Finally, the intra- and interplane artifacts on the whole breast volume resulting from the MC were calculated using the corresponding impulse responses and subsequently subtracted from the original reconstructed slices. The performance of our artifact-reduction method was evaluated using a computer-simulated MC phantom, as well as phantom images and patient DTMs obtained with IRB approval. A GE prototype DTM system that acquires 21 PVs in 3º increments over a ±30º range was used for image acquisition in this study. For the computer-simulated MC phantom, the eLACs can be estimated accurately, thus the interplane artifacts were effectively removed. For MCs in phantom and patient DTMs, our method reduced the artifacts but also created small over-corrected areas in some cases. Potential reasons for this may include: the simplified mathematical modeling of the forward projection process, and the amplified noise in the solution of the system of linear equations.

Proceedings Article | 17 March 2006 Paper

Performance analysis of 3-class classifiers: properties of the 3D ROC surface and the normalized volume under the surface

Berkman Sahiner, Heang-Ping Chan, Lubomir Hadjiiski

Proceedings Volume 6146, 61460C (2006) https://doi.org/10.1117/12.654581

KEYWORDS: Electronic filtering, Monte Carlo methods, Matrices, Computer aided diagnosis and therapy, Analytical research, Integration, Computer aided design, Medical imaging, Radiology, 3D image processing

Read Abstract +

Many computer-aided diagnostic problems involve three-class classification (e.g., abnormal (malignant), normal, and benign classes). A general treatment of the performance of a 3-class classifier results in a complex sixdimensional (6D) ROC space for which no analysis tools are available at present. A practical paradigm for performance analysis and a figure of merit (FOM) are yet to be developed. The area Az under the ROC curve plays an important role in the evaluation of two-class classifiers. An analogous FOM for three-class classification is highly desirable. We have been investigating conditions for reducing the 6D ROC space to 3D by constraining the utilities of some of the decisions in the classification task. These assumptions lead to a 3D ROC space in which the true-positive fraction (TPF) of one class can be expressed as a function of the two types of false-positive fractions (FPFs) from the other two classes. In this study, we investigated the properties of the 3D ROC surface for a maximum-likelihood classifier under the condition that the three class distributions in the feature space are multivariate normal with equal covariance matrices. Under the same conditions, we studied the properties of the normalized volume under the surface (NVUS) that was previously proposed as an FOM for a 3-class classifier. Under the equal covariance Gaussian assumption, it can be shown that the probability density functions of the decision variables follow a bivariate log-normal distribution. By considering these probability density functions, we can express the TPF in terms of the FPFs, and numerically integrate over the domain of the TPF to obtain the NVUS. We have compared the NVUS value obtained by directly integrating the probability density functions to that obtained from a Monte Carlo simulation study in which the 3D ROC surface was generated by empirical "optimal" classification of case samples in the multi-dimensional feature space following the assumed distributions. The NVUS values obtained from the direct integration method and the Monte Carlo simulation were found to be in good agreement. Our results indicate that the NVUS exhibits many of the intuitive properties that a proper FOM should satisfy and may be used as a performance index under the conditions that we imposed on the utilities.

Proceedings Article | 17 March 2006 Paper

Computer-aided detection of clustered microcalcifications on full-field digital mammograms: a two-view information fusion scheme for FP reduction

Jun Ge, Berkman Sahiner, Heang-Ping Chan, Lubomir Hadjiiski, Mark Helvie, Chuan Zhou, Jun Wei, Yiheng Zhang

Proceedings Volume 6144, 61445R (2006) https://doi.org/10.1117/12.654997

KEYWORDS: Mammography, CAD systems, Information fusion, Breast, Computer aided design, Image segmentation, Feature extraction, Computer aided diagnosis and therapy, Signal to noise ratio

Read Abstract +

We are developing new techniques to improve the performance of our computer-aided detection (CAD) system for clustered microcalcifications on full-field digital mammograms (FFDMs). In this study, we designed an information fusion scheme by using joint two-view information on craniocaudal (CC) and mediolateral-oblique (MLO) views. After cluster candidates were detected using a single-view detection technique, candidates on CC and MLO views were paired using their geometrical information. Candidate pairs were classified as true and false pairs with a similarity classifier that used the joint information from both views. Each cluster candidate was also characterized by its single-view features. The outputs of the similarity classifier and the single-view classifier were fused and the cluster candidate was classified as a true microcalcification cluster or a false-positive (FP) using the fused two-view information. A data set of 192 FFDM images was collected from 96 patients at the University of Michigan. All patients had two mammographic views. This data set contained 96 microcalcification clusters, of which 28 clusters were proven by biopsy to be malignant and 68 were proven to be benign. For training and testing the classifiers, the data set was partitioned into two independent subsets with the malignant cases equally distributed to the two subsets. One subset was used for training and the other subset was used for testing. We compared three computerized methods for geometrically pairing cluster candidates on two mammographic views. The areas under the fitted ROC curves were 0.75±0.01, 0.74±0.01, and 0.76±0.01 for the three methods, respectively. The difference between any two methods measured by the area under the fitted ROC curve, A_z, was not statistically significant (p > 0.05). We also evaluated a new hybrid pairing scheme that used two different sensitivity levels for defining cluster pairs based on the single-view scores. The single-view CAD system achieved cluster-based sensitivities of 75%, 80%, and 85% at 0.48, 0.86, and 1.05 FPs/image, respectively. The joint two-view CAD system achieved the same sensitivity levels at 0.29, 0.46, and 0.89 FPs/image. When the hybrid pairing was used in the joint two-view CAD system, the same cluster-based sensitivities were achieved at 0.26, 0.37, and 0.88 FPs/image. Our results indicate that correspondence of cluster candidates on two different views provides valuable additional information for distinguishing FPs from true microcalcification clusters.

Proceedings Article | 17 March 2006 Paper

Regularized discriminate analysis for breast mass detection on full field digital mammograms

Jun Wei, Berkman Sahiner, Yiheng Zhang, Heang-Ping Chan, Lubomir Hadjiiski, Chuan Zhou, Jun Ge, Yi-Ta Wu

Proceedings Volume 6144, 61445P (2006) https://doi.org/10.1117/12.654897

KEYWORDS: Feature selection, Feature extraction, Mammography, CAD systems, Breast, Computer aided design, Tissues, Computer aided diagnosis and therapy, Classification systems

Read Abstract +

In computer-aided detection (CAD) applications, an important step is to design a classifier for the differentiation of the abnormal from the normal structures. We have previously developed a stepwise linear discriminant analysis (LDA) method with simplex optimization for this purpose. In this study, our goal was to investigate the performance of a regularized discriminant analysis (RDA) classifier in combination with a feature selection method for classification of the masses and normal tissues detected on full field digital mammograms (FFDM). The feature selection scheme combined a forward stepwise feature selection process and a backward stepwise feature elimination process to obtain the best feature subset. An RDA classifier and an LDA classifier in combination with this new feature selection method were compared to an LDA classifier with stepwise feature selection. A data set of 130 patients containing 260 mammograms with 130 biopsy-proven masses was used. All cases had two mammographic views. The true locations of the masses were identified by experienced radiologists. To evaluate the performance of the classifiers, we randomly divided the data set into two independent sets of approximately equal size for training and testing. The training and testing were performed using the 2-fold cross validation method. The detection performance of the CAD system was assessed by free response receiver operating characteristic (FROC) analysis. The average test FROC curve was obtained by averaging the FP rates at the same sensitivity along the two corresponding test FROC curves from the 2-fold cross validation. At the case-based sensitivities of 90%, 80% and 70% on the test set, our RDA classifier with the new feature selection scheme achieved an FP rate of 1.8, 1.1, and 0.6 FPs/image, respectively, compared to 2.1, 1.4, and 0.8 FPs/image with stepwise LDA with simplex optimization. Our results indicate that RDA in combination with the sequential forward inclusion-backward elimination feature selection method can improve the performance of mass detection on mammograms. Further work is underway to optimize the feature selection and classification scheme and to evaluate if this approach can be generalized to other CAD classification tasks.

Proceedings Article | 17 March 2006 Paper

Computerized lung nodule detection on screening CT scans: performance on juxta-pleural and internal nodules

Berkman Sahiner, Lubomir Hadjiiski, Heang-Ping Chan, Chuan Zhou, Jun Wei

Proceedings Volume 6144, 61445S (2006) https://doi.org/10.1117/12.655272

KEYWORDS: Lung, Computed tomography, Computer aided design, Feature extraction, CAD systems, Spherical lenses, Computing systems, Image segmentation, Computer aided diagnosis and therapy, Neural networks

Read Abstract +

We are developing a computer-aided detection (CAD) system for lung nodules in thoracic CT volumes. Our CAD system includes an adaptive 3D pre-screening algorithm to segment suspicious objects, and a false-positive (FP) reduction stage to classify the segmented objects as true nodules or normal lung structures. We found that the effectiveness of the FP reduction stage was limited by the different characteristics of the objects in the internal and the juxta-pleural (JP) regions. The purpose of this study was to evaluate object characteristics in the internal and JP regions of a lung CT scan, and to develop different FP reduction classifiers for JP and internal objects. Our FP reduction technique utilized shape, grayscale, and gradient features, as well as the scores of a newly-developed neural network trained on the eigenvalues of the Hessian matrix in a volume of interest containing the suspicious object. We designed an algorithm to automatically label the objects as internal or JP. Based on a training set of 75 CT scans containing internal and JP nodules, two FP classifiers were trained separately for objects in the two types of lung regions. The system performance was evaluated on an independent test set of 27 low dose screening scans. An experienced chest radiologist identified 64 solid nodules (mean diameter: 5.3 mm, range: 3.0-12.9 mm) on the test cases, of which 33 were internal and 31 were JP. Our adaptive 3D prescreening algorithm detected 28 internal and 29 JP nodules. At 80% sensitivity, the average number of FPs was 3.9 and 9.7 in the internal and JP regions per scan, respectively. In comparison, a classifier designed to work on both types of nodules had an average of 29.4 FPs per scan at the same sensitivity. Our results indicate that it is more effective to use two different classifiers for JP and internal nodules because of their different characteristics. FPs in the JP region were more difficult to distinguish from true nodules. Further investigation of task-specific FP reduction techniques is needed.

Proceedings Article | 17 March 2006 Paper

Characterization of corresponding microcalcification clusters on temporal pairs of mammograms for interval change analysis: comparison of classifiers

Lubomir Hadjiiski, Douglas Drouillard, Heang-Ping Chan, Berkman Sahiner, Mark Helvie, Marilyn Roubidoux, Chuan Zhou

Proceedings Volume 6144, 61445Q (2006) https://doi.org/10.1117/12.654920

KEYWORDS: Mammography, Feature extraction, Computer aided diagnosis and therapy, Computing systems, Feature selection, Computer aided design, Image classification, Image registration, Radiology, Breast cancer

Read Abstract +

Proceedings Article | 15 March 2006 Paper

Automatic pulmonary vessel segmentation in 3D computed tomographic pulmonary angiographic (CTPA) images

Chuan Zhou, Heang-Ping Chan, Lubomir Hadjiiski, Smita Patel, Philip Cascade, Berkman Sahiner, Jun Wei, Jun Ge, Ella Kazerooni

Proceedings Volume 6144, 61444Q (2006) https://doi.org/10.1117/12.655343

KEYWORDS: Image segmentation, Lung, Arteries, 3D image processing, Gold, Tissues, Tomography, Angiography, Lymphatic system, Computer aided diagnosis and therapy

Read Abstract +

Automatic and accurate segmentation of the pulmonary vessels in 3D computed tomographic angiographic images (CTPA) is an essential step for computerized detection of pulmonary embolism (PE) because PEs only occur inside the pulmonary arteries. We are developing an automated method to segment the pulmonary vessels in 3D CTPA images. The lung region is first extracted using thresholding and morphological operations. 3D multiscale filters in combination with a newly developed response function derived from the eigenvalues of Hessian matrices are used to enhance all vascular structures including the vessel bifurcations and suppress non-vessel structures such as the lymphoid tissues surrounding the vessels. At each scale, a volume of interest (VOI) containing the response function value at each voxel is defined. The voxels with a high response indicate that there is an enhanced vessel whose size matches the given filter scale. A hierarchical expectation-maximization (EM) estimation is then applied to the VOI to segment the vessel by extracting the high response voxels at this single scale. The vessel tree is finally reconstructed by combining the segmented vessels at all scales based on a "connected component" analysis. Two experienced thoracic radiologists provided the gold standard of pulmonary arteries by manually tracking the arterial tree and marking the center of the vessels using a computer graphical user interface. Two CTPA cases containing PEs were used to evaluate the performance. One of these two cases also contained other lung diseases. The accuracy of vessel tree segmentation was evaluated by the percentage of the "gold standard" vessel center points overlapping with the segmented vessels. The result shows that 97.3% (1868/1920) and 92.0% (2277/2476) of the manually marked center points overlapped with the segmented vessels for the cases without and with other lung disease, respectively. The results demonstrate that vessel segmentation using our method is not degraded by PE occlusion and the vessels can be accurately extracted.

Proceedings Article | 10 March 2006 Paper

Computer-aided detection of breast masses on mammograms: bilateral analysis for false positive reduction

Yi-Ta Wu, Lubomir Hadjiiski, Jun Wei, Chuan Zhou, Berkman Sahiner, Heang-Ping Chan

Proceedings Volume 6144, 614421 (2006) https://doi.org/10.1117/12.655262

KEYWORDS: Mammography, CAD systems, Breast, Tissues, Feature extraction, Computing systems, Breast cancer, Computer aided design, Computer aided diagnosis and therapy

Read Abstract +

Proceedings Article | 10 March 2006 Paper

Two-view information fusion for improvement of computer-aided detection (CAD) of breast masses on mammograms

Jun Wei, Berkman Sahiner, Lubomir Hadjiiski, Heang-Ping Chan, Mark Helvie, Marilyn Roubidoux, Chuan Zhou, Jun Ge, Yiheng Zhang

Proceedings Volume 6144, 614424 (2006) https://doi.org/10.1117/12.654593

KEYWORDS: CAD systems, Mammography, Information fusion, Computer aided design, Breast, Image fusion, Nipple, Computer aided diagnosis and therapy, Receivers

Read Abstract +

Proceedings Article | 10 March 2006 Paper

Automated detection of ureter abnormalities on multi-detector row CT urography

Lubomir Hadjiiski, Berkman Sahiner, Elaine Caoili, Richard Cohan, Heang-Ping Chan

Proceedings Volume 6144, 61441W (2006) https://doi.org/10.1117/12.654985

KEYWORDS: Cancer, Computed tomography, Bladder, CAD systems, Shape analysis, Computer aided diagnosis and therapy, Image segmentation, Rule based systems, Visualization, Radiology

Read Abstract +

Proceedings Article | 2 March 2006 Paper

Tomosynthesis reconstruction using the simultaneous algebraic reconstruction technique (SART) on breast phantom data

Yiheng Zhang, Heang-Ping Chan, Berkman Sahiner, Jun Wei, Mitchell Goodsitt, Lubomir Hadjiiski, Jun Ge, Chuan Zhou

Proceedings Volume 6142, 614249 (2006) https://doi.org/10.1117/12.654891

KEYWORDS: Breast, Reconstruction algorithms, Digital mammography, Tissues, Mammography, Sensors, Image quality, Breast cancer, Prototyping, 3D image processing

Read Abstract +

Proceedings Article | 29 April 2005 Paper

Computerized pectoral muscle identification on MLO-view mammograms for CAD applications

Chuan Zhou, Lubomir Hadjiiski, Chintana Paramagul, Berkman Sahiner, Heang-Ping Chan, Jun Wei

Proceedings Volume 5747, (2005) https://doi.org/10.1117/12.595693

KEYWORDS: Mammography, Image enhancement, Breast, Gold, Image segmentation, Image filtering, Computer aided design, Tissues, Computer aided diagnosis and therapy, Edge detection

Read Abstract +

Proceedings Article | 29 April 2005 Paper

Computer-aided detection of microcalcification clusters on full-field digital mammograms: multiscale pyramid enhancement and false positive reduction using an artificial neural network

Jun Ge, Jun Wei, Lubomir Hadjiiski, Berkman Sahiner, Heang-Ping Chan, Mark Helvie, Chuan Zhou

Proceedings Volume 5747, (2005) https://doi.org/10.1117/12.595753

KEYWORDS: Image enhancement, CAD systems, Mammography, Image filtering, Image segmentation, Nonlinear filtering, Artificial neural networks, Computer aided diagnosis and therapy, Breast cancer

Read Abstract +

We are developing a computer-aided detection (CAD) system to detect microcalcification clusters automatically on full field digital mammograms (FFDMs). The CAD system includes five stages: preprocessing, image enhancement and/or box-rim filtering, segmentation of microcalcification candidates, false positive (FP) reduction, and clustering. In this study, we investigated the performance of a nonlinear multiscale Laplacian pyramid enhancement method in comparison with a box-rim filter at the image enhancement stage and the use of a new error metric to improve the efficiency and robustness of the training of a convolution neural network (CNN) at the FP reduction stage of our CAD system. A data set of 96 cases with 200 images was collected at the University of Michigan. This data set contained 215 microcalcification clusters, of which 64 clusters were proven by biopsy to be malignant and 151 were proven to be benign. The data set was separated into two independent data sets. One data set was used to train and validate the CNN in our CAD system. The other data set was used to evaluate the detection performance. For this data set, Laplacian pyramid multiscale enhancement did not improve the performance of the microcalcification detection system in comparison with our box-rim filter previously optimized for digitized screen-film mammograms. With the new error metric, the training of CNN could be accelerated and the classification performance in validation was improved from an A_z value of 0.94 to 0.97 on average. The CNN in combination with rule-based classifiers could reduce FPs with a small tradeoff in sensitivity. By using the free-response receiver operating characteristic (FROC) methodology, it was found that our CAD system can achieve a cluster-based sensitivity of 70%, 80%, and 88% at 0.23, 0.39, and 0.71 FP marks/image, respectively. For case-based performance evaluation, a sensitivity of 80%, 90%, and 98% can be achieved at 0.17, 0.27, and 0.51 FP marks/image, respectively.

Proceedings Article | 29 April 2005 Paper

False-positive reduction using Hessian features in computer-aided detection of pulmonary nodules on thoracic CT images

Berkman Sahiner, Zhanyu Ge, Heang-Ping Chan, Lubomir Hadjiiski, Naama Bogot, Philip Cascade, Ella Kazerooni

Proceedings Volume 5747, (2005) https://doi.org/10.1117/12.595714

KEYWORDS: Computed tomography, Lung, Feature extraction, Lung cancer, Convolution, Computing systems, Computer aided diagnosis and therapy, CAD systems, 3D image processing

Read Abstract +

Proceedings Article | 29 April 2005 Paper

Computer-aided detection of breast masses on mammograms: performance improvement using a dual system

Jun Wei, Berkman Sahiner, Lubomir Hadjiiski, Heang-Ping Chan, Mark Helvie, Marilyn Roubidoux, Nicholas Petrick, Jun Ge, Chuan Zhou

Proceedings Volume 5747, (2005) https://doi.org/10.1117/12.595745

KEYWORDS: Mammography, CAD systems, Breast, Computing systems, Classification systems, Breast cancer, Computer aided diagnosis and therapy, Neural networks, Tissues

Read Abstract +

Proceedings Article | 29 April 2005 Paper

Comparison of decision tree classifiers with neural network and linear discriminant analysis classifiers for computer-aided diagnosis: a Monte Carlo simulation study

Berkman Sahiner, Lubomir Hadjiiski, Heang-Ping Chan, Chuan Zhou, Jun Wei

Proceedings Volume 5747, (2005) https://doi.org/10.1117/12.595761

KEYWORDS: Computer aided diagnosis and therapy, Statistical analysis, Monte Carlo methods, Computer aided design, Binary data, Neural networks, Artificial neural networks, Computer simulations, CAD systems, Matrices

Read Abstract +

Proceedings Article | 6 April 2005 Paper

Effects of the continuous and discrete confidence rating scales in ROC observer studies

Lubomir Hadjiiski, Heang-Ping Chan, Berkman Sahiner, Mark Helvie, Marilyn Roubidoux

Proceedings Volume 5749, (2005) https://doi.org/10.1117/12.595725

KEYWORDS: Computer aided diagnosis and therapy, Mammography, Computer aided design, Statistical analysis, Chromium, Biopsy, Analytical research, Monte Carlo methods, Medical imaging, Data modeling

Read Abstract +

Proceedings Article | 12 May 2004 Paper

Computer-aided detection of breast masses on full-field digital mammograms: false positive reduction using gradient field analysis

Jun Wei, Berkman Sahiner, Lubomir Hadjiiski, Heang-Ping Chan, Nicholas Petrick, Mark Helvie, Chuan Zhou, Zhanyu Ge

Proceedings Volume 5370, (2004) https://doi.org/10.1117/12.536126

KEYWORDS: CAD systems, Mammography, Breast, Computer aided diagnosis and therapy, Feature extraction, Digital mammography, Manufacturing, Breast cancer, Digital filtering, Image enhancement

Read Abstract +

Proceedings Article | 12 May 2004 Paper

ROC study of the effects of computer-aided interval change analysis on radiologists' characterization of breast masses in two-view serial mammograms

Lubomir Hadjiiski, Mark Helvie, Berkman Sahiner, Heang-Ping Chan, Marilyn Roubidoux, Alexis Nees, Nicholas Petrick, Caroline Blane, Chintana Paramagul, Janet Bailey, Stephanie Patterson, Katherine Klein, Dorit Adler, Michelle Foster, Joseph Shen

Proceedings Volume 5370, (2004) https://doi.org/10.1117/12.535892

KEYWORDS: Computer aided diagnosis and therapy, Mammography, Breast, Biopsy, CAD systems, Statistical analysis, Breast cancer, Computer aided design, Feature extraction, Breast imaging

Read Abstract +

Proceedings Article | 12 May 2004 Paper

Computerized multiple image analysis on mammograms: performance improvement of nipple identification for registration of multiple views using texture convergence analyses

Chuan Zhou, Heang-Ping Chan, Berkman Sahiner, Lubomir Hadjiiski, Chintana Paramagul

Proceedings Volume 5370, (2004) https://doi.org/10.1117/12.535879

KEYWORDS: Nipple, Breast, Mammography, Image analysis, Tissues, Image registration, Computer aided diagnosis and therapy, Feature extraction, Radiology, Automatic tracking

Read Abstract +

Automated registration of multiple mammograms for CAD depends on accurate nipple identification. We developed two new image analysis techniques based on geometric and texture convergence analyses to improve the performance of our previously developed nipple identification method. A gradient-based algorithm is used to automatically track the breast boundary. The nipple search region along the boundary is then defined by geometric convergence analysis of the breast shape. Three nipple candidates are identified by detecting the changes along the gray level profiles inside and outside the boundary and the changes in the boundary direction. A texture orientation-field analysis method is developed to estimate the fourth nipple candidate based on the convergence of the tissue texture pattern towards the nipple. The final nipple location is determined from the four nipple candidates by a confidence analysis. Our training and test data sets consisted of 419 and 368 randomly selected mammograms, respectively. The nipple location identified on each image by an experienced radiologist was used as the ground truth. For 118 of the training and 70 of the test images, the radiologist could not positively identify the nipple, but provided an estimate of its location. These were referred to as invisible nipple images. In the training data set, 89.37% (269/301) of the visible nipples and 81.36% (96/118) of the invisible nipples could be detected within 1 cm of the truth. In the test data set, 92.28% (275/298) of the visible nipples and 67.14% (47/70) of the invisible nipples were identified within 1 cm of the truth. In comparison, our previous nipple identification method without using the two convergence analysis techniques detected 82.39% (248/301), 77.12% (91/118), 89.93% (268/298) and 54.29% (38/70) of the nipples within 1 cm of the truth for the visible and invisible nipples in the training and test sets, respectively. The results indicate that the nipple on mammograms can be detected accurately. This will be an important step towards automatic multiple image analysis for CAD techniques.

Proceedings Article | 12 May 2004 Paper

Multimodality CAD: combination of computerized classification techniques based on mammograms and 3D ultrasound volumes for improved accuracy in breast mass characterization

Berkman Sahiner, Heang-Ping Chan, Lubomir Hadjiiski, Marilyn Roubidoux, Chintana Paramagul, Mark Helvie, Chuan Zhou

Proceedings Volume 5370, (2004) https://doi.org/10.1117/12.535868

KEYWORDS: Mammography, Feature extraction, Breast, Image segmentation, 3D image processing, Biopsy, Ultrasonography, Image classification, Computer aided diagnosis and therapy, 3D acquisition

Read Abstract +

Proceedings Article | 12 May 2004 Paper

Computer-aided detection of lung nodules: false positive reduction using a 3D gradient field method

Zhanyu Ge, Berkman Sahiner, Heang-Ping Chan, Lubomir Hadjiiski, Jun Wei, Naama Bogot, Philip Cascade, Ella Kazerooni, Chuan Zhou

Proceedings Volume 5370, (2004) https://doi.org/10.1117/12.535943

KEYWORDS: Lung, Spherical lenses, 3D image processing, Computed tomography, Feature extraction, Computer aided diagnosis and therapy, Lung cancer, Optical spheres, CAD systems

Read Abstract +

Proceedings Article | 6 May 2004 Paper

Assessment of breast lesions on stereoscopic and monoscopic digital specimen mammograms: an ROC study

Heang-Ping Chan, Mitch Goodsitt, Mark Helvie, Lubomir Hadjiiski, Justin Lydick, Marilyn Roubidoux, Janet Bailey, Alexis Nees, Caroline Blane, Joseph Shen, Michelle Foster, Berkman Sahiner

Proceedings Volume 5368, (2004) https://doi.org/10.1117/12.535998

KEYWORDS: Tissues, Breast, Mammography, Digital imaging, Digital mammography, LCDs, Radiography, Medical imaging, Visualization, Eye

Read Abstract +

An ROC study was conducted to evaluate the usefulness of assessing breast lesion characteristics with stereomammography. Stereoscopic image pairs of 158 breast biopsy tissue specimens were acquired with a GE digital mammography system. Two stereo image pairs were taken at 1.8X magnification geometry and at approximately orthogonal orientations for each specimen. Display software was developed for a high resolution MegaScan CRT monitor driven by a DOME stereo display board. The specimens contained either a mass, microcalcifications, both, or normal tissue. About 40% of the specimens were found to contain malignancy by pathological analysis. Five MQSA radiologists participated in the observer performance experiment. The two views of each specimen were read independently and were separated by a large number of other specimen images to reduce any effects of memorization. Each observer read 316 specimen images in a systematically randomized order. The observer first read the monoscopic image and entered his/her assessment in terms of the confidence ratings on the presence of microcalcifications and/or masses, margin clearance, BI-RADS assessment, and the likelihood of malignancy. The corresponding stereoscopic images were then displayed on the same monitor and were viewed through stereoscopic LCD glasses. The observer was free to change the ratings in every category after stereoscopic reading. The ratings of the observers were analyzed by ROC methodology. For the 5 MQSA radiologists, the average Az value for estimation of the likelihood of malignancy of the lesions improved from 0.70 for monoscopic viewing to 0.72 (p<0.05) after stereoscopic viewing, and the average Az value for the presence of microcalcifications improved from 0.94 to 0.95 (p<0.05). The Az value for the presence of masses improved from 0.80 to 0.82 after stereoscopic viewing, but the difference fell short of statistical significance (p=0.08). The visual assessment of margin clearance was found to have very low correlation with pathological analysis. This study demonstrates the potential of using stereomammography to improve the detection and characterization of mammographic lesions.

Proceedings Article | 5 June 2003 Paper

Digital indirect-detection x-ray imagers with microlens focusing: effects of Fresnel reflections from the microlens layer

Nicholas Petrick, Aldo Badano, Heang-Ping Chan, Berkman Sahiner, Lubomir Hadjiiski

Proceedings Volume 5030, (2003) https://doi.org/10.1117/12.479995

KEYWORDS: Microlens, Imaging systems, Photodetectors, X-rays, X-ray imaging, Photons, X-ray detectors, Lenses, Prototyping, Monte Carlo methods

Read Abstract +

Proceedings Article | 15 May 2003 Paper

Design of three-class classifiers in computer-aided diagnosis: Monte Carlo simulation study

Heang-Ping Chan, Berkman Sahiner, Lubomir Hadjiiski, Nicholas Petrick, Chuan Zhou

Proceedings Volume 5032, (2003) https://doi.org/10.1117/12.481879

KEYWORDS: Computer aided design, Computer aided diagnosis and therapy, Lawrencium, Monte Carlo methods, Matrices, Tissues, Electronic filtering, Artificial neural networks, Barium, Sodium

Read Abstract +

For the development of computer-aided diagnosis (CAD) systems, a classifier that can effectively differentiate more than two classes is often needed. For example, a detected object on an image may need to be classified as a malignant lesion, a benign lesion, or normal tissue. Currently, a three-class problem is usually treated as a two-stage, two-class problem, in which the detected object is first differentiated as a lesion or normal tissue, and, in the second stage, the lesion is further classified as malignant or benign. In this work, we explored methods for classification of an object into one of the three classes, and compared the three-class approach with the common two-class approach. We conducted Monte Carlo simulation studies to evaluate the dependence of the performance of 3-class classification schemes on design sample size and feature space configurations. A k-dimensional multivariate normal feature space with three classes having different means was assumed. Linear classifiers and artificial neural networks (ANNs) were examined. ROC analysis for the 3-class approach was explored under simplifying conditions. A performance index representing the normalized volume under the ROC surface (NVUS) was defined. Linear classifiers for classification of three classes and two classes were compared. We found that a 3-class approach with a linear classifier can achieve a higher NVUS than that of a 2-class approach. We further compared the performance of an ANN having three or one output nodes with a linear classifier. At large sample sizes, a 3-output-node ANN was basically the same as that of a one-output-node ANN. When the three class distributions had equal covariance matrices and the distances between pairs of class means were equal, the linear classifiers could reach a higher performance for the test samples than the ANN when the design sample size was small; the linear classifier and the ANNs approached the same performance in the limit of large design sample size. However, under complex feature space configurations such as the class means located along a line, the class in the middle was poorly differentiated from the other two classes by the linear classifiers for any dimensionality; the ANN outperformed the linear classifier at all design sample size studied. This simulation study may provide some useful information to guide the design of 3-class classifiers for various CAD applications.

Proceedings Article | 15 May 2003 Paper

Computerized detection of pulmonary embolism in 3D computed tomographic (CT) images: vessel tracking and segmentation techniques

Chuan Zhou, Lubomir Hadjiiski, Berkman Sahiner, Heang-Ping Chan, Smita Patel, Philip Cascade, Ella Kazerooni, Jun Wei

Proceedings Volume 5032, (2003) https://doi.org/10.1117/12.481369

KEYWORDS: Image segmentation, Tissues, Arteries, Lung, Chest, Computed tomography, 3D image processing, Tomography, Computer aided diagnosis and therapy, Optical tracking

Read Abstract +

Proceedings Article | 15 May 2003 Paper

ROC study: effects of computer-aided diagnosis on radiologists' characterization of malignant and benign breast masses in temporal pairs of mammograms

Lubomir Hadjiiski, Heang-Ping Chan, Berkman Sahiner, Mark Helvie, Marilyn Roubidoux, Caroline Blane, Chintana Paramagul, Nicholas Petrick, Janet Bailey, Katherine Klein, Michelle Foster, Stephanie Patterson, Dorit Adler, Alexis Nees, Joseph Shen

Proceedings Volume 5032, (2003) https://doi.org/10.1117/12.483549

KEYWORDS: Computer aided diagnosis and therapy, Mammography, Breast, CAD systems, Biopsy, Feature extraction, Breast imaging, Breast cancer, Image segmentation, Receivers

Read Abstract +

Proceedings Article | 15 May 2003 Paper

Three-dimensional active contour model for characterization of solid breast masses on three-dimensional ultrasound images

Berkman Sahiner, Aditya Ramachandran, Heang-Ping Chan, Marilyn Roubidoux, Lubomir Hadjiiski, Mark Helvie, Nicholas Petrick, Chuan Zhou

Proceedings Volume 5032, (2003) https://doi.org/10.1117/12.483548

KEYWORDS: 3D modeling, 3D image processing, Image segmentation, Ultrasonography, Breast, Solids, Feature extraction, Tumor growth modeling, Data modeling, 3D acquisition

Read Abstract +

Proceedings Article | 9 May 2002 Paper

Optimal neural network architecture selection: effects on computer-aided detection of mammographic microcalcifications

Metin Gurcan, Heang-Ping Chan, Berkman Sahiner, Lubomir Hadjiiski, Nicholas Petrick, Mark Helvie

Proceedings Volume 4684, (2002) https://doi.org/10.1117/12.467095

KEYWORDS: Mammography, Computer aided diagnosis and therapy, Neural networks, Signal to noise ratio, CAD systems, Computer architecture, Databases, Convolution, Algorithms, Computer aided design

Read Abstract +

Proceedings Article | 9 May 2002 Paper

Use of joint two-view information for computerized lesion detection on mammograms: improvement of microcalcification detection accuracy

Berkman Sahiner, Metin Gurcan, Heang-Ping Chan, Lubomir Hadjiiski, Nicholas Petrick, Mark Helvie

Proceedings Volume 4684, (2002) https://doi.org/10.1117/12.467220

KEYWORDS: Mammography, Breast, Signal to noise ratio, Image segmentation, Databases, Image processing, Nipple, Detection and tracking algorithms, Computer aided diagnosis and therapy, Analytical research

Read Abstract +

Proceedings Article | 9 May 2002 Paper

Computer-aided characterization of malignant and benign microcalcification clusters based on the analysis of temporal change of mammographic features

Lubomir Hadjiiski, Heang-Ping Chan, Berkman Sahiner, Nicholas Petrick, Mark Helvie, Marilyn Roubidoux, Metin Gurcan

Proceedings Volume 4684, (2002) https://doi.org/10.1117/12.467218

KEYWORDS: Mammography, Feature extraction, Image filtering, Computer aided diagnosis and therapy, Feature selection, Computer aided design, Biopsy, Statistical analysis, Breast, Image classification

Read Abstract +

Proceedings Article | 3 May 2002 Paper

Digital stereomammography: observer performance study of the effects of magnification and zooming on depth perception

Heang-Ping Chan, Mitchell Goodsitt, Lubomir Hadjiiski, Janet Bailey, Katherine Klein, Katie Darner, Chintana Paramagul

Proceedings Volume 4682, (2002) https://doi.org/10.1117/12.465554

KEYWORDS: Visualization, Mammography, Digital imaging, Digital mammography, X-rays, Stereoscopy, Tissues, LCDs, Sensors, Glasses

Read Abstract +

Proceedings Article | 3 July 2001 Paper

Analysis of temporal change of mammographic features for computer-aided characterization of malignant and benign masses

Lubomir Hadjiiski, Berkman Sahiner, Heang-Ping Chan, Nicholas Petrick, Mark Helvie, Metin Gurcan

Proceedings Volume 4322, (2001) https://doi.org/10.1117/12.431142

KEYWORDS: Mammography, Feature extraction, Image filtering, Computer aided diagnosis and therapy, Image segmentation, Feature selection, Image classification, Biopsy, Computer aided design, Detector development

Read Abstract +

Proceedings Article | 3 July 2001 Paper

Recognition of lesion correspondence on two mammographic views: a new method of false-positive reduction for computerized mass detection

Berkman Sahiner, Nicholas Petrick, Heang-Ping Chan, Sophie Paquerault, Mark Helvie, Lubomir Hadjiiski

Proceedings Volume 4322, (2001) https://doi.org/10.1117/12.431139

KEYWORDS: Mammography, Breast, Computer aided diagnosis and therapy, Image filtering, Feature extraction, Image segmentation, Computer aided design, Feature selection, Image enhancement, Data fusion

Read Abstract +

Proceedings Article | 3 July 2001 Paper

Computerized lung nodule detection on thoracic CT images: combined rule-based and statistical classifier for false-positive reduction

Metin Gurcan, Nicholas Petrick, Berkman Sahiner, Heang-Ping Chan, Philip Cascade, Ella Kazerooni, Lubomir Hadjiiski

Proceedings Volume 4322, (2001) https://doi.org/10.1117/12.431145

KEYWORDS: Lung, Computed tomography, Image segmentation, Computer aided diagnosis and therapy, CAD systems, Lung cancer, Computing systems, Classification systems, Evolutionary algorithms, Computer aided design

Read Abstract +

Proceedings Article | 6 June 2000 Paper

Interval change analysis in temporal pairs of mammograms using a local affine transformation

Lubomir Hadjiiski, Heang-Ping Chan, Berkman Sahiner, Nicholas Petrick, Mark Helvie, Sophie Paquerault, Chuan Zhou

Proceedings Volume 3979, (2000) https://doi.org/10.1117/12.387748

KEYWORDS: Mammography, Breast, Image registration, Computer aided diagnosis and therapy, Tissues, Nipple, Calibration, Detector development, Image processing, Laser scanners

Read Abstract +

Proceedings Article | 6 June 2000 Paper

Computerized image analysis: estimation of breast density on mammograms

Chuan Zhou, Heang-Ping Chan, Nicholas Petrick, Berkman Sahiner, Mark Helvie, Marilyn Roubidoux, Lubomir Hadjiiski, Mitchell Goodsitt

Proceedings Volume 3979, (2000) https://doi.org/10.1117/12.387676

KEYWORDS: Breast, Image segmentation, Mammography, Image enhancement, Visualization, Image analysis, Tissues, Image compression, Breast cancer, Computer aided diagnosis and therapy

Read Abstract +

Proceedings Article | 28 May 1999 Paper

Digital mammographic imaging using microlens focusing: estimation of light collection and x-ray utilization

Nicholas Petrick, Heang-Ping Chan, Mitchell Goodsitt, Berkman Sahiner, Lubomir Hadjiiski

Proceedings Volume 3659, (1999) https://doi.org/10.1117/12.349502

KEYWORDS: X-rays, Imaging systems, Photodetectors, X-ray imaging, Microlens array, Microlens, Photons, Digital imaging, Digital x-ray imaging, X-ray detectors

Read Abstract +

Proceedings Article | 28 May 1999 Paper

Digital mammography: observer performance study of the effects of pixel size on radiologists' characterization of malignant and benign microcalcifications

Heang-Ping Chan, Mark Helvie, Nicholas Petrick, Berkman Sahiner, Dorit Adler, Caroline Blane, Lynn Joynt, Chintana Paramagul, Marilyn Roubidoux, Todd Wilson, Lubomir Hadjiiski, Mitchell Goodsitt

Proceedings Volume 3659, (1999) https://doi.org/10.1117/12.349514

KEYWORDS: Mammography, Digital mammography, Image classification, Statistical analysis, Sensors, Imaging systems, Laser scanners, 3D scanning, Feature extraction, Image quality

Read Abstract +

Proceedings Article | 21 May 1999 Paper

Hybrid unsupervised-supervised approach for computerized classification of malignant and benign masses on mammograms

Lubomir Hadjiiski, Berkman Sahiner, Heang-Ping Chan, Nicholas Petrick, Mark Helvie

Proceedings Volume 3661, (1999) https://doi.org/10.1117/12.348601

KEYWORDS: Mammography, Computer aided design, Feature selection, Feature extraction, Chlorine, Image filtering, Tumor growth modeling, Computer aided diagnosis and therapy, Visibility, Receivers

Read Abstract +

Proceedings Article | 21 May 1999 Paper

Stepwise linear discriminant analysis in computer-aided diagnosis: the effect of finite sample size

Berkman Sahiner, Heang-Ping Chan, Nicholas Petrick, Robert Wagner, Lubomir Hadjiiski

Proceedings Volume 3661, (1999) https://doi.org/10.1117/12.348606

KEYWORDS: Feature selection, Statistical analysis, Computer aided design, Matrices, Computer aided diagnosis and therapy, Feature extraction, Mammography, Computer simulations, Mahalanobis distance, Receivers

Read Abstract +