International Journal For Multidisciplinary Research

E-ISSN: 2582-2160   •   Impact Factor: 9.24

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

Call for Paper Volume 8, Issue 4 (July-August 2026) Submit your research before last 3 days of August to publish your research paper in the issue of July-August.
Submit Research Paper Join as a Reviewer

Plagiarism is checked by the leading plagiarism checker TurnItIn

Call for Paper

Volume 8 Issue 4
July-August 2026

Submit your research paper

Indexing Partners
Academia Advanced Sciences Index Bielefeld Academic Search Engine CiteSeer DRJI Google Scholar Independent Search Engine & Directory Network ISI (International Scientific Indexing) Issuu Mendeley Research Networks
RefSeek ResearcherId - Thomson Reuters ResearchGate Scirus Scribd Semantic Scholar UTeM - Universiti Teknikal Malaysia Melaka Wiki for Call for Papers WorldCat Zenodo

Multi-Class Cataract Detection from Pupil Images from Rural India

Author(s) Anoushka Agrawal, Jaden Shiju, Saiprathist Reganti, Vivaan Rao
Country United States
Abstract Cataracts are one of the leading causes of vision impairment worldwide, and their burden is disproportionately concentrated in low-resource settings such as rural India. Many existing state-of-the-art deep learning models for cataract detection struggle to localize the clinically relevant region, the pupil, and instead learn spurious correlations with irrelevant features such as skin color and eyebrow texture. In this work, it is proposed that cataract grading can be improved by restricting model input to the pupil region and addressing two systematic data quality issues: scale variability and specular reflection artifacts. We developed a two-path preprocessing pipeline that applies partial convolution-based inpainting to remove bright spot artifacts and randomized zoom augmentation via a MONAI pipeline to normalize scale. A ResNet-18 classifier is then trained on pupil-cropped images to perform six-class cataract grading (No Cataract, NS1–NS5). Experiments on 1,639 pupil images demonstrate that the combined training strategy, mixing inpainted and non-inpainted images with data augmentation, achieves the best multi-class AUC of 0.8919. It is observed that the model reliably distinguishes healthy from cataractous eyes, while confusion between adjacent intermediate grades (NS2–NS3) remains a challenge for future work.
Field Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In Volume 8, Issue 4, July-August 2026
Published On 2026-07-26
DOI https://doi.org/10.36948/ijfmr.2026.v08i04.83729
View / Download PDF File

Share this



E-ISSN 2582-2160
CrossRef

CrossRef DOI prefix of IJFMR is 10.36948/ijfmr

Downloads
Research Paper Format Copyright Permission Form and Undertaking Form

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 Open Access

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
Powered by Sky Research Publication and Journals