dorsal/arxiv
View SchemaTemporal-Enhanced Interpretable Multi-Modal Prognosis and Risk Stratification Framework for Diabetic Retinopathy (TIMM-ProRS)
| Authors | Susmita Kar, A S M Ahsanul Sarkar Akib, Abdul Hasib, Samin Yaser, Anas Bin Azim |
|---|---|
| Categories | |
| ArXiv ID | 2601.08240vv1 |
| URL | https://arxiv.org/abs/2601.08240 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
Diabetic retinopathy (DR), affecting millions globally with projections indicating a significant rise, poses a severe blindness risk and strains healthcare systems. Diagnostic complexity arises from visual symptom overlap with conditions like age-related macular degeneration and hypertensive retinopathy, exacerbated by high misdiagnosis rates in underserved regions. This study introduces TIMM-ProRS, a novel deep learning framework integrating Vision Transformer (ViT), Convolutional Neural Network (CNN), and Graph Neural Network (GNN) with multi-modal fusion. TIMM-ProRS uniquely leverages both retinal images and temporal biomarkers (HbA1c, retinal thickness) to capture multi-modal and temporal dynamics. Evaluated comprehensively across diverse datasets including APTOS 2019 (trained), Messidor-2, RFMiD, EyePACS, and Messidor-1 (validated), the model achieves 97.8\% accuracy and an F1-score of 0.96, demonstrating state-of-the-art performance and outperforming existing methods like RSG-Net and DeepDR. This approach enables early, precise, and interpretable diagnosis, supporting scalable telemedical management and enhancing global eye health sustainability.
{
"annotation_id": "ae8c1825-56c9-4375-b273-2579bcc046c0",
"date_created": "2026-02-17T05:53:15.433000Z",
"date_modified": "2026-02-17T05:53:15.433000Z",
"file_hash": "301fb0ba5035e03a83ff94b59faaaf0a2c361f03195d6fa6dc15bc36ae407687",
"private": false,
"record": {
"abstract": "Diabetic retinopathy (DR), affecting millions globally with projections indicating a significant rise, poses a severe blindness risk and strains healthcare systems. Diagnostic complexity arises from visual symptom overlap with conditions like age-related macular degeneration and hypertensive retinopathy, exacerbated by high misdiagnosis rates in underserved regions. This study introduces TIMM-ProRS, a novel deep learning framework integrating Vision Transformer (ViT), Convolutional Neural Network (CNN), and Graph Neural Network (GNN) with multi-modal fusion. TIMM-ProRS uniquely leverages both retinal images and temporal biomarkers (HbA1c, retinal thickness) to capture multi-modal and temporal dynamics. Evaluated comprehensively across diverse datasets including APTOS 2019 (trained), Messidor-2, RFMiD, EyePACS, and Messidor-1 (validated), the model achieves 97.8\\% accuracy and an F1-score of 0.96, demonstrating state-of-the-art performance and outperforming existing methods like RSG-Net and DeepDR. This approach enables early, precise, and interpretable diagnosis, supporting scalable telemedical management and enhancing global eye health sustainability.",
"arxiv_id": "2601.08240",
"authors": [
"Susmita Kar",
"A S M Ahsanul Sarkar Akib",
"Abdul Hasib",
"Samin Yaser",
"Anas Bin Azim"
],
"categories": [
"eess.IV",
"cs.CV"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "Temporal-Enhanced Interpretable Multi-Modal Prognosis and Risk Stratification Framework for Diabetic Retinopathy (TIMM-ProRS)",
"url": "https://arxiv.org/abs/2601.08240",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "8c92edc3-8c18-4d8e-a89b-19c96349e8d4",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}