Fundus Image
Fundus images, photographs of the back of the eye, are crucial for diagnosing various retinal diseases. Current research focuses on improving the accuracy and efficiency of automated analysis using deep learning models, including convolutional neural networks (CNNs), Vision Transformers (ViTs), and generative adversarial networks (GANs), often incorporating techniques like attention mechanisms and self-supervised learning to handle noisy or limited data. These advancements aim to improve diagnostic accuracy, reduce the need for expert interpretation, and enable large-scale screening programs for early detection and intervention of eye diseases, ultimately impacting patient care and public health. A significant area of ongoing work addresses the challenges of generalizability and bias in these models to ensure equitable performance across diverse populations and imaging modalities.
Papers
Grading and Anomaly Detection for Automated Retinal Image Analysis using Deep Learning
Syed Mohd Faisal Malik, Md Tabrez Nafis, Mohd Abdul Ahad, Safdar Tanweer
SSP-RACL: Classification of Noisy Fundus Images with Self-Supervised Pretraining and Robust Adaptive Credal Loss
Mengwen Ye, Yingzi Huangfu, You Li, Zekuan Yu
A Disease-Specific Foundation Model Using Over 100K Fundus Images: Release and Validation for Abnormality and Multi-Disease Classification on Downstream Tasks
Boa Jang, Youngbin Ahn, Eun Kyung Choe, Chang Ki Yoon, Hyuk Jin Choi, Young-Gon Kim
MM-UNet: A Mixed MLP Architecture for Improved Ophthalmic Image Segmentation
Zunjie Xiao, Xiaoqing Zhang, Risa Higashita, Jiang Liu