SignificanceEarly detection of melanoma is crucial for improving patient outcomes, and dermoscopy is a critical tool for this purpose. However, hair presence in dermoscopic images can obscure important features, complicating the diagnostic process. Enhancing image clarity by removing hair without compromising lesion integrity can significantly aid dermatologists in accurate melanoma detection.AimWe aim to develop a novel synthetic hair dermoscopic image dataset and a deep learning model specifically designed for hair removal in melanoma dermoscopy images.ApproachTo address the challenge of hair in dermoscopic images, we created a comprehensive synthetic hair dataset that simulates various hair types and dimensions over melanoma lesions. We then designed a convolutional neural network (CNN)-based model that focuses on effective hair removal while preserving the integrity of the melanoma lesions.ResultsThe CNN-based model demonstrated significant improvements in the clarity and diagnostic utility of dermoscopic images. The enhanced images provided by our model offer a valuable tool for the dermatological community, aiding in more accurate and efficient melanoma detection.ConclusionsThe introduction of our synthetic hair dermoscopic image dataset and CNN-based model represents a significant advancement in medical image analysis for melanoma detection. By effectively removing hair from dermoscopic images while preserving lesion details, our approach enhances diagnostic accuracy and supports early melanoma detection efforts.
Scars usually do not show strong contrast under standard skin examination by using dermoscopes. We show that Mueller matrix polarimetry can provide strong contrast for in vivo scar imaging. Scars usually develop after skin injury when the body repairs the damaged tissue. They are causing multiple distresses such as movement restrictions, pain, itchiness, and the psychological impact of the associated cosmetic disfigurement. Scar treatment has significant economic impact as well. Mueller matrix polarimetry with integrated autofocus and automatic data registration can potentially improve the scar assessment by the dermatologists and help to objectify the evaluation of the treatment outcome.
KEYWORDS: Image fusion, Skin, Image segmentation, Machine learning, Melanoma, Data acquisition, Cancer detection, Digital imaging, Skin cancer, Medical imaging
We propose a method for digital hair removal from dermoscopic images that involves a new scheme for the acquisition of dermoscopic images to uncover hidden information by employing image fusion of multiple images of lesions. Classical approaches for the removal of hair from dermoscopic images are usually based on interpolation, pattern propagation or machine learning. These replace the hair pixels with calculated data aiming to reduce the impairment of the medical diagnostics. While these approaches are well established, the problem of information loss is not addressed. We show that our approach can lead to improved skin lesion assessment in dermoscopy.
Significance: Mueller matrix (MM) polarimetry is a promising tool for the detection of skin cancer. Polarimetric in vivo measurements often suffer from misalignment of the polarimetric images due to motion, which can lead to false results.Aim: We aim to provide an easy-to-implement polarimetric image data registration method to ensure proper image alignment.Approach: A feature-based image registration is implemented for an MM polarimeter for phantom and in vivo human skin measurements.Results: We show that the keypoint-based registration of polarimetric images is necessary for in vivo skin polarimetry to ensure reliable results. Further, we deliver an efficient semiautomated method for the registration of polarimetric images.Conclusions: Image registration for in vivo polarimetry of human skin is required for improved diagnostics and can be efficiently enhanced with a keypoint-based approach.
The orientation and concentration of structures like collagen within biological tissues can provide valuable information, for example, in skin disease diagnostics. Polarimetry lends itself for non-destructive investigation in various fields of research and development ranging from medical diagnostics to production monitoring, among others. We report on a system for polarimetric measurement of versatile targets in reflection and transmission mode. It efficiently determines the Mueller matrix (MM) of a sample under study and is also suited for in vivo applications. Generally, the Mueller matrix Mm allows to calculate the Stokes vector So of the light interacting with a sample, containing all information on its polarization properties, through So = Mm Si where Si is the Stokes vector of the illuminating light. The Mueller matrix can be derived from images taken with different polarization states of illuminating and observed light. In our setup we use liquid crystal retarders to precisely control the polarization states of the light. This enables fast measurement of the orientation of structures with high spatial resolution. In a first example, we demonstrate the capability of our system by characterizing electrospun fiber tissue implants and measuring the degree of alignment and orientation of the fibers in reflection mode. The results lead us to a deeper understanding of the signals which we expect from structures like collagen in skin. We were able to derive a correlation between the properties of the tissue structures, the parameters for production and the MM information, for the first time. This was possible by suitable decomposition of the MM into submatrices of known physical interpretation. In this work we present our latest results and discuss the next steps towards in vivo application in dermatology or tissue implant.
Optical systems have shown their potential in non-invasive medical diagnostics over the last years. While most imaging systems use information on wavelength or phase, e.g. OCT, in our approach we focus on the polarization properties of biotissue. We designed a Mueller matrix (MM) measurement setup for in vivo investigations on skin tissue. The MM describes the polarization-changing properties of a sample.. Thus, it is possible to calculate the MM from images taken with different polarization states of the illuminating and the observed light. For medical application, an important requirement is that the process is fast to enable in vivo measurement, avoid motion artifacts, and reduce stress for patients. In our setup, we use a combination of two polarizers and four liquid crystal retarders to quickly change between polarization states. The system is able to measure the location dependent MM of a target for different wavelengths. It is designed for measurement in reflection mode, however, upon simple modifications, it can be used in transmission mode as well. One interesting field of application is diagnostics for inflammatory skin diseases. Here, for example, changes in the structure and concentration of collagen could provide diagnostically valuable information. We evaluated our system on different skin phantoms to investigate the diagnostic advantages compared to standard approaches. In the future, our system could be part of a non-contact dermatoscopic device and provide extra information for the physician.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.