Papers for

diabetes care teams

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Streaming model improves real-time personalized diabetes forecasting

SSMCGM-Stream: A Streaming State-Space Model for Personalized Diabetes Management

Abstract: Continuous glucose monitoring supports diabetes management but requires forecasting models that are accurate, interpretable, and efficient enough for real-time deployment. We present SSMCGM-Stream, a streaming state-space model that maintains a personalized hidden state updated with each incoming glucose reading to provide constant-time forecasting, physiologically structured counterfactual reasoning, and interpretable forecasts through hidden-state attribution. Built on this foundation, SSMCGM-Agent delivers risk-aware, natural-language decision support by translating forecasts, explanations, and counterfactuals into grounded recommendations through a guarded language-model interface. Evaluated on the T1DEXI cohort, SSMCGM-Stream achieves forecasting performance comparable to strong neural baselines while substantially reducing computational cost, enabling continuous on-device diabetes decision support.

Tue 20 OctMachine LearningArtificial Intelligence

The gist

Managing diabetes requires predicting blood sugar levels continuously and accurately. The authors developed a model that updates its understanding of a person’s glucose in real time as new data arrives, making it fast and personalized. This model also explains its predictions and suggests how different actions might affect glucose levels. Their approach performs as well as complex neural networks but is much faster and can run directly on a user’s device.

Open → 2610.28976v1