Multimodal Artificial Intelligence Models Predicting Glaucoma Progression Using Electronic Health Records and Retinal Nerve Fiber Layer Scans.

Purpose: The purpose of this study was to develop models that predict which patients with glaucoma will progress to require surgery, combining structured data from electronic health records (EHRs) and retinal fiber layer optical coherence tomography (RNFL OCT) scans. Methods: EHR data (demographics and clinical eye examinations) and RNFL OCT scans were identified for patients with glaucoma from an academic center (2008-2023). Comparing the novel TabNet deep learning architecture to a baseline XGBoost model, we trained and evaluated single modality models using either EHR or RNFL features, as well as fusion models combining both EHR and RNFL features as inputs, to predict glaucoma surgery within 12 months (binary). Results: We had 1472 patients with glaucoma who were included in this study, of which 29.9% (N = 367) progressed to glaucoma surgery. The TabNet fusion model achieved the highest performance on the test set with an area under the receiver operating characteristic curve (AUROC) of 0.832, compared to the XGBoost fusion model (AUROC = 0.747). EHR only models performed with an AUROC of 0.764 and 0.720 for the deep learning model and XGBoost models, respectively. RNFL only models performed with an AUROC of 0.624 and 0.633 for the deep learning and XGBoost models, respectively. Conclusions: Fusion models which integrate both RNFL with EHR data outperform models only utilizing one datatype or the other to predict glaucoma progression. The deep learning TabNet architecture demonstrated superior performance to traditional XGBoost models. Translational Relevance: Prediction models that utilize the wealth of structured clinical and imaging data to predict glaucoma progression could form the basis of future clinical decision support tools to personalize glaucoma care.