Fig. 5: The structure of 5 models of 3 types for visual field (VF) prediction. | npj Digital Medicine

Fig. 5: The structure of 5 models of 3 types for visual field (VF) prediction.

From: Interpretable longitudinal glaucoma visual field estimation deep learning system from fundus images and clinical narratives

Fig. 5

a–c basic VF estimation deep learning models (EDL) using color fundus photographs (CFPs), region of interest (ROI) of CFPs, ROI with optic disc and optic cup segmentation. d multi-modal VF estimation deep learning model (MEDL) using ROI and clinical information. e longitudinal VF estimation deep learning model (LEDL) using ROI and interval years for future VF prediction.

Back to article page