Authors : Dr. Dwiti Krishna Bebarta; N. Srujana; R. R. V. Kusuma; A. Lavanya

Volume/Issue : Volume 11 - 2026, Issue 4 - April

Google Scholar : https://tinyurl.com/yp8473mz

Scribd : https://tinyurl.com/msx4j6uv

DOI : https://doi.org/10.38124/ijisrt/26apr932

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : Diabetic retinopathy, glaucoma and cataracts are the eye diseases that cause serious impairment of the vision in the world. Early signs can avoid the severe deterioration of vision, and in order to do it, it is vital to diagnose a person using only skilled ophthalmologists and specialized equipment, which is not always available. In this paper, the authors introduce an AI-based Systemic Disease screening with the help of eye images and multiagent Large Language Models (LLMs). The system makes use of deep learning model, which is DenseNet121 to extract features of eye pictures, and LightGBM to give precise classification of the disease with probability scores. GradyCAM visualization is also added to show critical areas that affect the model predictions to help in improving interpretability. A multiagent LLM framework is proposed to improve analysis and usability, and the Diagnosis, Validation, Risk Assessment, Explanation, and Report agents make up a multi agent LLM. Such agents offer a complete set of insights, reports that are easy to understand by patients and doctor-level data and enhance decision-making. Multi-language translation, and voice output are also enabled in the system which allows it to be used by a broader group of users. The suggested system shows greater accuracy, interpretability, and accessibility in the screening of eye diseases. It may help medical practitioners to diagnose early and give improved awareness to the patients, particularly in facilities with limited resources.

Keywords : Eye Disease Detection, Deep Learning, DenseNet121, LightGBM, Medical Image Analysis, Multi-Agent LLMs, GradCAM, Artificial Intelligence in Healthcare, Image Processing, Automated Diagnosis.

References :

1. Lavric, A., Anchidin, L., Popa, V., Al-Timemy, A. H., Alyasseri, Z., Takahashi, H., Yousefi, S., & Hazarbassanov, R. M. (2021). Keratoconus severity detection from elevation, topography and pachymetry raw data using a machine learning approach. IEEE Access, 9, 84344–84355.
2. Ismail, A. M., et al. (2024). Ensemble transfer learning networks for disease classification from retinal optical coherence tomography images. Journal of Optics, 1–16.
3. Kourou, K., Exarchos, T. P., Exarchos, K. P., Karamouzis, M. V., & Fotiadis, D. I. (2021). Applied machine learning in cancer research: A systematic review for patient diagnosis, classification and prognosis. Computational and Structural Biotechnology Journal, 19, 5546–5555.
4. Huang, G., Liu, Z., Van Der Maaten, L., & Weinberger, K. Q. (2017). Densely connected convolutional networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 4700–4708.
5. Bhattacharyya, S., De, S., Bhattacharjee, S., & Yoshida, K. (Eds.). (2020). Hybrid Machine Intelligence for Medical Image Analysis. Springer, Berlin. No. 172536.
6. Gulshan, V., Peng, L., Coram, M., Stumpe, M. C., Wu, D., Narayanaswamy, A., ... & Webster, D. R. (2016). Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA, 316(22), 2402–2410.
7. Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., ... & Liu, T. Y. (2017). LightGBM: A highly efficient gradient boosting decision tree. Advances in Neural Information Processing Systems (NeurIPS), 30.
8. Li, T., et al. (2019). Patients with suspected ocular disease intelligent recognition from retinal images using deep learning. In Proceedings of IEEE BIBM 2019 International Conference on Bioinformatics and Biomedicine.
9. Wang, X., et al. (2020). Attention-based multi-label fundus image classification. Medical Image Analysis, 60, 101572.
10. Rajpurkar, P., Irvin, J., Ball, R. L., Zhu, K., Yang, B., Mehta, H., ... & Lungren, M. P. (2017). CheXNet: Radiologist-level pneumonia detection on chest X-rays with deep learning. arXiv preprint arXiv:1711.05225.
11. Ting, D. S. W., Cheung, C. Y. L., Lim, G., Tan, G. S. W., Quang, N. D., Gan, A., ... & Wong, T. Y. (2017). Development and validation of a deep learning system for diabetic retinopathy and related eye diseases using retinal images from multiethnic populations with diabetes. JAMA, 318(22), 2211–2223.
12. Qian, R., et al. (2019). A new approach for diabetic retinopathy screening using optical coherence tomography and deep learning. International Journal of Ophthalmology, 12(12), 1883–1890.
13. Wang, L., et al. (2023). Unleashing cognitive synergy in large language models: A task-solving agent through multi-persona self-collaboration. In NeurIPS Workshop on Generative AI for Education (GAIED).
14. ODIR-5K Dataset. (2019). Ocular Disease Intelligent Recognition: A real-life ophthalmic database for multimodal multi-label diagnosis of ophthalmic conditions. Grand Challenge. Retrieved from https://odir2019.grand-challenge.org/
15. He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 770–778.
16. Simonyan, K., & Zisserman, A. (2015). Very deep convolutional networks for large-scale image recognition. In Proceedings of the International Conference on Learning Representations (ICLR).
17. Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., ... & Amodei, D. (2020). Language models are few-shot learners. Advances in Neural Information Processing Systems (NeurIPS), 33, 1877–1901.
18. Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2017). Grad-CAM: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), 618–626.
19. Abramoff, M. D., Lavin, P. T., Birch, M., Shah, N., & Folk, J. C. (2018). Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices. NPJ Digital Medicine, 1(1), 1–8.
20. Grinsztajn, L., Oyallon, E., & Varoquaux, G. (2022). Why tree-based models still outperform deep learning on tabular data. Advances in Neural Information Processing Systems (NeurIP