Attcatvgg-Net: an Explainable Multioutput Deep Learning Framework for Cataract Stage Classification and Visual Acuity Regression using Multicolor Fundus Images
Mostafa Nazarpour-Servak, Neda Taghinezhad, Tahereh Mahmoudi, Ali Azimi, Mohammad Hossein Nowroozzadeh
This paper develops AttCatVgg-Net, a deep learning model that classifies cataract stages and predicts visual acuity using multicolor fundus images, achieving high accuracy and strong correlation with actual VA.
bioRxiv · Oct 25, 2025 · PreprintBioengineering