SSM-CGM: Interpretable State-Space Forecasting Model of Continuous Glucose Monitoring for Personalized Diabetes Management
This work addresses the need for interpretable forecasting models in personalized diabetes management, offering incremental improvements in accuracy and interpretability for clinical use.
The paper tackles the problem of forecasting continuous glucose monitoring (CGM) data for diabetes management by developing SSM-CGM, a Mamba-based neural state-space model that integrates CGM and wearable activity signals, improving short-term accuracy over a Temporal Fusion Transformer baseline and adding interpretability features like variable selection and counterfactual forecasts.
Continuous glucose monitoring (CGM) generates dense data streams critical for diabetes management, but most used forecasting models lack interpretability for clinical use. We present SSM-CGM, a Mamba-based neural state-space forecasting model that integrates CGM and wearable activity signals from the AI-READI cohort. SSM-CGM improves short-term accuracy over a Temporal Fusion Transformer baseline, adds interpretability through variable selection and temporal attribution, and enables counterfactual forecasts simulating how planned changes in physiological signals (e.g., heart rate, respiration) affect near-term glucose. Together, these features make SSM-CGM an interpretable, physiologically grounded framework for personalized diabetes management.