🤖 AI Summary
This study addresses the challenge that assessing glaucoma progression rates typically requires multiple visual field tests over years, hindering the rapid identification of high-risk patients. We propose GLAM, a model that leverages longitudinal Humphrey visual field data and clinical features via an attention-based fusion mechanism to predict progression rates while quantifying aleatoric uncertainty. The key innovation lies in achieving performance comparable to multimodal approaches using only routine unimodal visual field data, thereby substantially shortening the assessment period. Experimental results demonstrate that GLAM predicts mean deviation (MD) progression rates with a mean absolute error of 0.139 dB/year (R²=0.927), reducing the error by 73.5% compared to baselines. Furthermore, the model attains an AUC of 0.990 for detecting fast progressors, enabling highly accurate and cost-effective early screening.
📝 Abstract
Glaucoma is the leading cause of irreversible blindness, and timely identification of fast progressors is essential to prevent disability. Current practice estimates progression by ordinary least-squares regression of mean deviation (MD) on time, requiring 6--10 visual field (VF) tests over several years to obtain a reliable slope. We present GLAM (Glaucoma Longitudinal Analysis Model), a deep learning framework that ingests longitudinal Humphrey 24-2 total deviation sequences with five clinical features and predicts MD and visual field index progression rates using attention-based fusion and aleatoric uncertainty. On the open-access University of Washington Humphrey Visual Field dataset (4,276 patient-eyes), GLAM achieved an MD-rate mean absolute error of 0.139 dB yr$^{-1}$ ($R^2 = 0.927$; 73.5% reduction over a ridge baseline) and an AUC of 0.990 for fast-progressor detection. VF-only deep learning can match multimodal pipelines for progression prognostication using routinely collected perimetry alone.