Paper
13 June 2024 Brain tumor image recognition method based on convolutional neural network
Ranran Tie, Shixiang Zhang, Zengying Chao, Genxiu Chen, Shuang Wang, Bocheng Liu
Author Affiliations +
Proceedings Volume 13180, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2024); 131805G (2024) https://doi.org/10.1117/12.3034124
Event: International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2024), 2024, Guangzhou, China
Abstract
This study focuses on brain tumors, a highly fatal malignant lesion of the cranial cavity, and proposes a brain tumor detection and automatic classification framework based on the VGG16 pre-trained model. By employing transfer learning, we incorporated global average pooling layers and fully connected layers into the model, and froze the last layer to prevent overfitting and underfitting. We used the focal loss function to optimize model training, enabling it to learn from difficult samples and down-weight simple samples. In performance evaluation, we compared the improved VGG-GAP model with the original VGG16, VGG19, and InceptionV3, and the results showed significant improvements in all performance metrics.
(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Ranran Tie, Shixiang Zhang, Zengying Chao, Genxiu Chen, Shuang Wang, and Bocheng Liu "Brain tumor image recognition method based on convolutional neural network", Proc. SPIE 13180, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2024), 131805G (13 June 2024); https://doi.org/10.1117/12.3034124
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Tumors

Brain

Neuroimaging

Magnetic resonance imaging

Education and training

Machine learning

Convolutional neural networks

Back to Top