Advanced retinal disease detection using RHT-Net: a hybrid deep learning approach with augmented fundus imaging
Yaoyang Cheng, Yiling Pan, Guodao Zhang, Wei Wu, Yisu Ge, Jing Wang · 2026
Read Advanced retinal disease detection using RHT-Net: a hybrid deep learning approach with augmented fundus imaging by Yaoyang Cheng, Yiling Pan, Guodao Zhang, Wei Wu, Yisu Ge, Jing Wang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13341505.txt.
Description
A hybrid neural network sorts nine retinal diseases from fundus photos, in a 2026 study by Yaoyang Cheng and colleagues. RHT-Net fuses residual convolutional layers for local detail with transformer encoders that catch long-range patterns across the image.
Training used 5,318 real fundus images from Bengali patients, expanded by rotation, flipping, and noise to 21,272 images, then contrast-enhanced and resized. The model hit 97.93% training accuracy and 96.12% on held-out tests, with an overall F1-score of 95.31%. Heat map visualizations supported its decisions. Scalable screening like this could ease specialist shortages in eye care.
Catalog record
| CREATOR | Yaoyang Cheng, Yiling Pan, Guodao Zhang, Wei Wu, Yisu Ge, Jing Wang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 60797 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |