Copyright © 2025 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{187134,
        author = {Lalit Kumar Rawat and Prof. Anil Kumar and Dr. Vijendra Pratap Singh},
        title = {Explainable AI for Diabetic Retinopathy Detection Using EfficientNetB4 and Swin Transformer},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {6},
        pages = {3257-3261},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=187134},
        abstract = {Diabetic Retinopathy (DR) is a leading cause of vision loss worldwide, and its early detection is crucial for preventing blindness. Recent advances in deep learning have revolutionized automated medical image analysis, but the lack of interpretability remains a significant barrier to clinical adoption. This study proposes an explainable AI framework using EfficientNetB4 and Swin Transformer architectures for automated detection of diabetic retinopathy from retinal fundus images. The model integrates Grad-CAM visual explanations to enhance transparency and assist clinicians in understanding model decisions. Experiments conducted on the APTOS 2019 and EyePACS datasets show an average classification accuracy of 95.3% and an area under the ROC curve (AUC) of 0.985. The explainability analysis demonstrates that the proposed hybrid model focuses on clinically relevant lesion regions such as microaneurysms and exudates. This work bridges the gap between performance and interpretability, providing a viable AI-based screening tool for early DR diagnosis, especially relevant to low-resource healthcare settings like India.},
        keywords = {Diabetic Retinopathy, Explainable AI, EfficientNet, Swin Transformer, Grad-CAM, Medical Imaging},
        month = {November},
        }