Explainable AI (XAI) for Early Detection of Diabetic Retinopathy

     We have presented an end‑to‑end XAI framework for detecting diabetic retinopathy that combines solid classification performance with multimodal, human‑readable explanations. The complementary nature of Grad‑CAM, LIME, and SHAP gives clinicians a fuller picture of model reasoning, fosters trust, and helps flag potential errors. In the future, we plan to extend the system to video‑based OCT‑angiography and to incorporate concept‑based explanations grounded in medical lesion ontologies. Prospective clinical trials are on the horizon to test the impact in real screening workflows.

1. Gulshan, V., Peng, L., Coram, M., Stumpe, M.C., Wu, D., Narayanaswamy, A., Venugopalan, S., Widner, K., Madams, T., Cuadros, J., Kim, R., Raman, R., Nelson, P.C., Mega, J.L., Webster, D.R.: Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs. JAMA 316(22) (2016) 2402–2410 2. Gargeya, R., Leng, T.: Automated Identification of Diabetic Retinopathy Using Deep Learning. Ophthalmology 124(7) (2017) 962–969 3. Quellec, G., Charrière, K., Boudi, Y., Cochener, B., Lamard, M.: Deep Image Mining for Diabetic Retinopathy Screening. Medical Image Analysis 39 (2017) 178–193   4. Selvaraju, R.R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., Batra, D.: Grad‑CAM: Visual Explanations from Deep Networks via Gradient‑Based Localization. International Journal of Computer Vision 128(2) (2020) 336–359   5. Ribeiro, M.T., Singh, S., Guestrin, C.: “Why Should I Trust You?”: Explaining the Predictions of Any Classifier. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), San Francisco, CA, USA (2016) 1135–1144