This study developed and evaluated a deep learning model (DLM) to predict eyes at high risk of surgical intervention for uncontrolled glaucoma based on multimodal data from initial ophthalmology visits. The DLMs were trained to predict the occurrence of glaucoma surgery within various time horizons, from 3 months to 5 years. The inputs included spatially oriented visual field (VF) and optical coherence tomography (OCT) data, as well as clinical and demographic features. The model achieved clinically useful AUC values (>0.8) for all models predicting surgery within 3 years. Shapley additive explanations (SHAP) were used to assess the importance of different features, with intraocular pressure (IOP), mean deviation (MD), and average retinal nerve fiber layer (RNFL) thickness being among the top 5 most important features. The DLMs can successfully identify eyes requiring surgery for uncontrolled glaucoma within specific time horizons, with predictive performance decreasing as the time horizon increases. Implementing such prediction models in clinical settings may help identify patients who should be referred to a glaucoma specialist for surgical evaluation.This study developed and evaluated a deep learning model (DLM) to predict eyes at high risk of surgical intervention for uncontrolled glaucoma based on multimodal data from initial ophthalmology visits. The DLMs were trained to predict the occurrence of glaucoma surgery within various time horizons, from 3 months to 5 years. The inputs included spatially oriented visual field (VF) and optical coherence tomography (OCT) data, as well as clinical and demographic features. The model achieved clinically useful AUC values (>0.8) for all models predicting surgery within 3 years. Shapley additive explanations (SHAP) were used to assess the importance of different features, with intraocular pressure (IOP), mean deviation (MD), and average retinal nerve fiber layer (RNFL) thickness being among the top 5 most important features. The DLMs can successfully identify eyes requiring surgery for uncontrolled glaucoma within specific time horizons, with predictive performance decreasing as the time horizon increases. Implementing such prediction models in clinical settings may help identify patients who should be referred to a glaucoma specialist for surgical evaluation.