A Large Sensor Foundation Model Pretrained on Continuous Glucose Monitor Data for Diabetes Management

TL;DR

This paper introduces CGM-LSM, a large Transformer-based model pretrained on extensive CGM data, significantly improving glucose prediction accuracy and robustness for diabetes management across diverse patient groups.

Contribution

The paper presents a novel large sensor model pretrained on diverse CGM data, enabling generalizable and accurate glucose forecasting for diabetes care.

Findings

48.51% reduction in root mean square error for 1-hour forecasts

Zero-shot prediction performance across different patient groups

Improved robustness and accuracy over prior models

Abstract

Continuous glucose monitoring (CGM) combined with AI offers new opportunities for proactive diabetes management through real-time glucose forecasting. However, most existing models are task-specific and lack generalization across patient populations. Inspired by the autoregressive paradigm of large language models, we introduce CGM-LSM, a Transformer decoder-based Large Sensor Model (LSM) pretrained on 1.6 million CGM records from patients with different diabetes types, ages, and genders. We model patients as sequences of glucose time steps to learn latent knowledge embedded in CGM data and apply it to the prediction of glucose readings for a 2-hour horizon. Compared with prior methods, CGM-LSM significantly improves prediction accuracy and robustness: a 48.51% reduction in root mean square error in one-hour horizon forecasting and consistent zero-shot prediction performance across held-out patient groups. We analyze model performance variations across patient subgroups and prediction scenarios and outline key opportunities and challenges for advancing CGM foundation models.