Sequential Deep Learning to Predict Non-Central to Central Geographic Atrophy Progression from OCT Imaging

Authors

Abstract

PurposeTo develop and validate a temporal deep learning framework for predicting geographic atrophy (GA) progression across multi-year horizons using longitudinal optical coherence tomography (OCT) sequences. DesignRetrospective longitudinal cohort study. Subjects, Participants, and/or ControlsA total of 91 patients with dry age-related macular degeneration (AMD) were identified from Wake Forest University School of Medicine (2013-2023), yielding 455 OCT volumes. Two prediction cohorts were defined: 32 patients with no GA (NGA) at baseline who subsequently developed GA, and 35 patients whose earliest GA manifestation was non-central GA (NCGA). Non-progressing patients served as negative controls. MethodsOCT B-scan volumes were encoded into visit-level feature representations using three pretrained architectures (ResNet-18, ResNet-50, ViT-B/16). Chronologically ordered visit embeddings, optionally augmented with inter-visit time intervals ({Delta}t), were processed through recurrent neural networks (RNN), long short-term memory networks (LSTM), and Transformer encoders to model longitudinal disease trajectories. Models were trained and evaluated independently for prediction horizons of 2, 3, 4, 5, and 6 years using patient-level stratified splits (80/20). Performance was assessed across five random seeds. Main Outcome MeasuresArea under the receiver operating characteristic curve (ROC-AUC), F1-score, and accuracy for predicting two clinically critical transitions: NGA to GA onset and NCGA to central GA (CGA) involvement. ResultsFor NGA to GA prediction, models achieved ROC-AUC of 0.84-0.94 at 2-4 years and 1.00 at 5-6 years. For NCGA to CGA prediction, Transformer-based models achieved peak AUC of 0.95 ({+/-}0.11) at 4 years and 0.96 ({+/-}0.06) at 5 years. Longer input sequences (8 visits vs. 4 visits) consistently improved NCGA to CGA performance at extended horizons. Temporal interval encoding improved stability in several LSTM configurations. ConclusionsTemporal deep learning applied to longitudinal OCT sequences can predict GA progression across clinically meaningful 2-6 year horizons without pixel-level annotations. These findings support the feasibility of automated, individualized risk stratification to guide complement inhibitor therapy decisions in patients with GA.

Tags

Topics