Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM68842
Volume:
10 |
11 |
Nov 2026
Pages:
1-5
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.

Department of Computer Science & Engineering
Don Bosco Institute of Technology
Department of Computer Science & Engineering
Don Bosco Institute of Technology
Department of Computer Science & Engineering
Don Bosco Institute of Technology
Department of Computer Science & Engineering
Don Bosco Institute of Technology
Department of Computer Science & Engineering
Don Bosco Institute of Technology
Cardiovascular diseases (CVDs) continue to be the primary cause of death globally, underscoring the need for efficient methods of early detection and risk assessment. The retina provides a distinctive, non-invasive view of the body’s vascular system, and recent progress in deep learning has enabled the extraction of health indicators from retinal imagery. This project develops a deeplearning-based system that analyzes retinal fundus images to support screening for diabetic retinopathy and selected systemic health risks, including cardiovascular risk. The system uses a convolutional neural network (CNN) based on Inception-v3 and performs image-quality assessment, preprocessing, EM/GMM segmentation and feature extraction before classification. It classifies diabetic retinopathy into five stages and can optionally combine retinal information with clinical parameters such as age, blood pressure, BMI and glucose levels for preliminary risk assessment. Class activation maps (CAMs) highlight the retinal regions that influence each prediction. The system is implemented as a web-based application using Python and Flask and provides image upload, prediction, visualization, history and report-generation features. It is intended as a fast, non-invasive screening-support and research tool that can assist healthcare professionals, and it does not replace professional medical diagnosis.
KEYWORDS
This paper presented RetinaHealth AI, a deep-learning-based retinal-image screening-support system for diabetic retinopathy and selected systemic health-risk assessment. The system processes retinal fundus photographs through image-quality assessment, preprocessing, EM/GMM segmentation and Inception-v3 feature extraction, and uses the extracted features for five-stage diabetic-retinopathy classification. Optional clinical parameters such as age, blood pressure, BMI and glucose can be combined with retinal information for preliminary diabetes, hypertension and cardiovascular-risk estimation, and CAM heatmaps improve the interpretability of the predictions. A Flask-based web application integrates image processing, prediction, history management, database storage and PDF report generation. The main contribution is the integration of several screening-support functions into a single modular platform rather than a focus on a single prediction task. The system is intended to assist preliminary screening and research and should not be considered a replacement for professional medical diagnosis. Further work with larger datasets, independent validation and clinical evaluation is necessary before real-world deployment.
[1] L. Y. Li, A. A. Isaksen, B. Lebiecka-Johansen, K. Funck, V. Thambawita, S. Byberg, T. H. Andersen, O.Nørgaard, and A. Hulman, “Prediction of cardiovascular markers and diseases using retinal fundus images and deep learning: A systematic scoping review,” Eur. Heart J. Digit. Health, vol. 5, pp. 660–669, 2025, doi: 10.1093/ehjdh/ztae068. [2] S. A. Hassan, S. Akbar, and H. U. Khan, “Detection of central serous retinopathy using deep learning through retinal images,” Multimed. Tools Appl., 2025, doi: 10.1007/s11042-023-16206-y. [3] K. A. Muthukumar, Nandi, Ranjan, P., Ramachandran, & Janardhan R “Integrating electrocardiogram and fundus images for early detection of cardiovascular diseases,” Sci. Rep., vol. 15, 2024, doi: 10.1038/s41598-025-87634-z. [4] R. Poplin et al., “Prediction of cardiovascular risk factors from retinal fundus photographs via deep learning,” Nat. Biomed. Eng., vol. 2, no. 3, pp. 158–164, Mar. 2023. [5] R. G. Barriada and D. Masip, “An overview of deep-learning-based methods for cardiovascular risk assessment with retinal images,” Diagnostics, vol. 13, no. 1, Art. no. 68, 2023, doi: 10.3390/diagnostics13010068. [6] H. R. H. Al-Absi et al., “Cardiovascular disease diagnosis from DXA scan and retinal images using deep learning,” Sensors, vol. 22, no. 12, Art. no. 4310, 2022, doi: 10.3390/s22124310. [7] V. Shankar et al., “Heart disease prediction using CNN algorithm,” SN Comput. Sci., vol. 1, Art. no. 170, 2022. [8] C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna, “Rethinking the Inception architecture for computer vision,” in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), 2021, pp. 2818–2826. [9] B. Zhou, A. Khosla, A. Lapedriza, A. Oliva, and A. Torralba, “Learning deep features for discriminative localization,” in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), 2020, pp. 2921–2929. [10] A. P. Dempster, N. M. Laird, and D. B. Rubin, “Maximum likelihood from incomplete data via the EM algorithm,” J. R. Stat. Soc. B, vol. 39, no. 1, pp. 1–38, 2020.
Publication Date
11/9/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
2
Total views since publication
PDF Downloads
3
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...