Deep Learning–Based Detection of Early Renal Function Impairment Using Retinal Fundus Images: Model Development and Validation
importance 6/10
ai assisted
Retinal Imaging and Analysis renal diagnostics stage: research
Problem
Detecting early renal function impairment non-invasively, using retinal fundus images instead of blood tests.
Why it matters
Earlier detection of kidney impairment could enable earlier intervention before overt disease.
What the AI did
A deep learning model was trained on retrospective fundus images (2001–2019) to detect eGFR <90 mL/min/1.73 m2, evaluated via ROC/AUC.
Results
not stated in source
Impact
Evidence supports model development and validation only; performance figures absent from abstract.
Record
| Venue | JMIR Medical Informatics |
|---|---|
| Published | 2020 |
| Work type | article |
| Authors | Eugene Yu‐Chuan Kang, Yi‐Ting Hsieh, Chien‐Hung Li, Yijin Huang, Chang‐Fu Kuo, Je‐Ho Kang, Kuan‐Jen Chen, Chi‐Chun Lai et al. |
| Citations | 56 (OpenAlex, at import time) |
| References | 37 |
| Model or method | not stated in source |
| Organization | not stated in source |
| Primary source | https://doi.org/10.2196/23472 |
| Open access | gold |
| Keywords | deep learning, retinal fundus images, renal function, eGFR, medical imaging |
Importance rating
6
significant — thousands to millions affected
Non-invasive early detection of renal impairment via retinal images is a genuinely new capability enabling earlier intervention; abstract gives no disease prevalence or impact figures.