Deep Retinal Image Analysis (DRIA)
A unified framework of retinal image analysis that perform two tasks; segmentation and classification. The segmentation part provides (1) blood vessel, (2) optic disc/cup segmentation. While the classification part provides classification for diabetic retinopathy (DR), glaucoma and age-related macular degeneration.
Related Paper
Multi-view Deep Mammograms Image Analysis (MV-DMIA)
A multi-view convolutional neural network based on multiple mammography views that extracts discriminative features from the mediolateral oblique (MLO) and craniocaudal (CC) views of the mammographic images for both left and right breast and can effectively incorporate these features for mass deletion and classification. Dataset for this task is publicly available.
Related Paper
GANs for Enhancing Retinal Fundus Image Segmentation
Given medical imaging's annotation scarcity, Computer Vision researchers need new ways of obtaining data. One way is through Generative Adverserial Networks (GANs). These improve the generalizeability of models. While researchers have used GANs to improve MRI and CT scan segmentation, occular fundus image segmentation with GANs is an untapped project.
Related Paper 1 Related Paper 2