Hybrid Eye Power and Disease Prediction System: Integrating CNN-based Retinal Image Analysis with ML-based Perception Testing

Oviya S; Prashaanth S M; Rosario Joseph Stalin -; Suresh P

doi:10.36948/ijfmr.2026.v08i02.70709

Hybrid Eye Power and Disease Prediction System: Integrating CNN-based Retinal Image Analysis with ML-based Perception Testing

Author(s)	Ms. Oviya S, Mr. Prashaanth S M, Mr. Rosario Joseph Stalin -, Prof. Suresh P
Country	India
Abstract	Visual impairment caused by refractive errors and retinal diseases such as diabetic retinopathy and glaucoma remains a major global health concern. Early detection and accurate refractive assessment are essential to prevent irreversible vision loss; however, conventional diagnostic procedures require specialized equipment and trained professionals, limiting accessibility in remote and resource-constrained regions. This paper presents a Hybrid Eye Power and Disease Prediction System that integrates convolutional neural network (CNN)-based retinal image analysis with machine learning-driven perception testing. A DenseNet-121 model is employed for refractive error regression and classification from retinal fundus images, while a K-Nearest Neighbors (KNN) model analyzes structured visual acuity test responses to estimate eye power. A weighted fusion mechanism combines predictions from both modalities to enhance accuracy and robustness. Additionally, a MobileNetV2-based module performs multi-class retinal disease detection, further fused with symptom-based inputs to generate a comprehensive disease risk score. Experimental results demonstrate that the proposed hybrid framework outperforms single-modality approaches, achieving reduced regression error and improved classification metrics for both refractive estimation and disease detection. The multimodal design enhances stability against image variability and subjective response inconsistencies. The system is intended as a non-invasive, screening-support tool suitable for telemedicine, school health programs, and large-scale preliminary eye assessment. While not a replacement for clinical diagnosis, it offers a scalable and accessible solution for early ocular evaluation and risk stratification.
Keywords	Refractive Error, Retinal Image Analysis, CNN, KNN, Hybrid Fusion, Eye Disease Detection, Telemedicine.
Field	Biology > Medical / Physiology
Published In	Volume 8, Issue 2, March-April 2026
Published On	2026-03-07
DOI	https://doi.org/10.36948/ijfmr.2026.v08i02.70709

View / Download PDF File

E-ISSN 2582-2160

doi

CrossRef DOI is assigned to each research paper published in our journal.

IJFMR DOI prefix is
10.36948/ijfmr

Downloads

Research Paper Format Copyright Permission Form and Undertaking Form Cover Page Vol 8 Isu 2 Cover Page Vol 8 Isu 1 Cover Page Vol 7 Isu 6

All research papers published on this website are licensed under Creative Commons Attribution-ShareAlike 4.0 International License, and all rights belong to their respective authors/researchers.

CC-BY-SA

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Hybrid Eye Power and Disease Prediction System: Integrating CNN-based Retinal Image Analysis with ML-based Perception Testing

Share this