LLM-Powered Personalized Glycemic Assessment in Type 2 Diabetes with Wearable Sensor Data

2026-06-10

AI Summary

Existing approaches struggle to effectively integrate wearable device data with personalized contextual information, limiting the generalizability of glycemic assessment for individuals with type 2 diabetes. This work proposes GlyLLM, a novel framework that, for the first time, leverages pretrained large language models for personalized glucose modeling. GlyLLM jointly abstracts continuous glucose monitoring readings, multimodal wearable sensor data, and structured health metadata to enable both sequential modeling and semantic reasoning. Experimental results on the AI-READI dataset demonstrate that GlyLLM reduces the root mean square error (RMSE) of glucose prediction by an average of 13.66% and improves the area under the receiver operating characteristic curve (AUROC) for diabetes classification by 13.08% compared to conventional methods, underscoring the critical role of diabetes-related questionnaires and biomarker measurements in comprehensive glycemic evaluation.

Abstract

Type 2 Diabetes (T2D) poses an increasing global health threat, demanding effective glycemic assessment to support personalized and improved diabetes care. Wearable sensors such as continuous glucose monitors (CGM) and fitness trackers offer many valuable insights for glycemic assessment. However, effectively analyzing these data requires integration with essential individual-level context. Existing methods are often based on traditional machine learning (ML) and rely primarily on historical blood glucose measurements and overlook personalized information, which limits their performance across diverse diabetes populations. Recent advances in large language models (LLMs) have demonstrated their ability to integrate diverse data modalities while modeling sequential dependencies, motivating the exploration of their potential for personalized glycemic assessment. In this paper, we propose GlyLLM, an LLM-powered framework for modeling CGM-based glycemic dynamics through the integration of wearable sensor data and structured metadata. GlyLLM can leverage the extensive prior knowledge of pre-trained LLMs and achieve sensor-text semantic abstraction at decision time. Experiments on two related tasks on the AI-READI dataset demonstrate that our model outperforms traditional ML methods by an average of 13.66\% in Root Mean Squared Error (RMSE) for glucose forecasting and 13.08\% in Area Under the Receiver Operating Characteristic (AUROC) for diabetes categorization. Additionally, our ablation study shows that diabetes surveys and biometric tests are more critical than other health information for glycemic assessment. Our work presents a promising step toward harnessing the power of LLMs to advance personalized glycemic assessment in T2D care.

Problem

• personalized glycemic assessment
• Type 2 Diabetes
• wearable sensor data
• individual-level context
• glucose forecasting

Innovation

• Large Language Models
• Personalized Glycemic Assessment
• Wearable Sensors
• Continuous Glucose Monitoring
• Multimodal Integration