Evaluating Time-Series Foundation Models and Multimodal Dietary Context for CGM Forecasting

• URL: http://arxiv.org/abs/2609.11872v1
• Date: Thu, 10 Sep 2026 17:43:29 GMT
• Title: Evaluating Time-Series Foundation Models and Multimodal Dietary Context for CGM Forecasting
• Abstract summary: We conduct a study using eight public CGM datasets spanning Type 1 diabetes, Type 2 diabetes, and non-diabetes populations.<n>Foundation models did not consistently outperform strong task-specific baselines such as Elastic Net and PatchTST.<n>Fine-tuned Chronos-Bolt reduced RMSE by 6.5%-18.4% in the T1D cohort and by 8.6%-18.2% in the non-diabetes/T2D cohort.<n>A residual-based fusion framework reduced overall RMSE by approximately 3% and postprandial RMSE by approximately 15% relative to the CGM-only baseline.
• Score: 14.534275200164712
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Continuous glucose monitoring (CGM) provides high-frequency measurements of glucose dynamics and enables short-term glucose forecasting for diabetes management. Although time-series foundation models have shown strong general forecasting ability, their effectiveness for CGM prediction and the added value of multimodal dietary context remain unclear. We conduct a comprehensive empirical study using eight public CGM datasets spanning Type 1 diabetes, Type 2 diabetes, and non-diabetes populations. Under a unified protocol across multiple context lengths and prediction horizons, zero-shot foundation models did not consistently outperform strong task-specific baselines such as Elastic Net and PatchTST. In contrast, lightweight fine-tuning substantially improved forecasting performance. For example, fine-tuned Chronos-Bolt reduced RMSE by 6.5%-18.4% in the T1D cohort and by 8.6%-18.2% in the non-diabetes/T2D cohort, with comparable improvements in both in-distribution and out-of-distribution test settings. We further evaluate multimodal dietary context using CGMacros, which provides temporally aligned CGM signals, food images, and macronutrient records. A residual-based fusion framework reduced overall RMSE by approximately 3% and postprandial RMSE by approximately 15% relative to the CGM-only baseline. Moreover, Chronos-based CGM representations were more strongly correlated with observed postprandial glucose increments than representations from LSTM and CatBoost, even after those models incorporated additional dietary modalities, suggesting that pretrained temporal representations better preserve meal-induced excursion patterns. These findings show that foundation models require CGM-specific adaptation for reliable forecasting and that dietary context provides clinically meaningful signals beyond CGM alone, especially during postprandial periods.

Related papers

• GlucoFM-Bench: Benchmarking Time-Series Foundation Models for Blood Glucose Forecasting [6.427581905332425]
  We present GlucoFM-Bench, a benchmark evaluating state-of-the-art foundation models for blood glucose forecasting.<n>We assess eight representative architectures, including pre-trained TSFMs, time-series large language models, and task-specific deep learning models.<n>Across datasets, pre-trained TSFMs show strong zero-shot and few-shot transfer, with the best zero-shot model performing within 5% of the best full-shot supervised model.
  arXiv  Detail & Related papers  (2026-06-05T03:49:26Z)
• GlucoFM: A Dual-Stream Foundation Model for Continuous Glucose Monitoring [12.757891952959987]
  GlucoFM is a lightweight foundation model for continuous glucose monitoring (CGM)<n>It aligns irregular recordings to a 24-hour chronological grid, preserves observation masks, and decomposes glucose dynamics into slow physiological state and transient event streams.<n>It achieves the strongest subject-disjoint linear-probing performance among evaluated baselines.
  arXiv  Detail & Related papers  (2026-05-29T05:46:35Z)
• GlyRAG: Context-Aware Retrieval-Augmented Framework for Blood Glucose Forecasting [3.494950334697973]
  GlyRAG is a context-aware, retrieval-augmented forecasting framework that derives semantic understanding of blood glucose dynamics directly from CGM traces.<n>GlyRAG consistently outperforms state-of-the art methods, achieving up to 39% lower RMSE and a further 1.7% reduction in RMSE over the baseline.
  arXiv  Detail & Related papers  (2026-01-08T20:07:59Z)
• Integrating Neural Differential Forecasting with Safe Reinforcement Learning for Blood Glucose Regulation [51.12307713554633]
  TSODE is a safety-aware controller that integrates Thompson RL Sampling with a Neural Ordinary Differential Equation forecaster.<n>In the FDA-approved UVa/Padova simulator (adult cohort), TSODE achieved 87.9% time-in-range with less than 10% time below 70 mg/dL.
  arXiv  Detail & Related papers  (2025-11-16T02:11:33Z)
• A Large Sensor Foundation Model Pretrained on Continuous Glucose Monitor Data for Diabetes Management [3.8195320624847833]
  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.<n>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.<n>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.
  arXiv  Detail & Related papers  (2024-12-12T21:35:13Z)
• From Glucose Patterns to Health Outcomes: A Generalizable Foundation Model for Continuous Glucose Monitor Data Analysis [47.23780364438969]
  We present GluFormer, a generative foundation model for CGM data that learns nuanced glycemic patterns and translates them into predictive representations of metabolic health.<n>GluFormer generalizes to 19 external cohorts spanning different ethnicities and ages, 5 countries, 8 CGM devices, and diverse pathophysiological states.<n>In a longitudinal study of 580 adults with CGM data and 12-year follow-up, GluFormer identifies individuals at elevated risk of developing diabetes more effectively than blood HbA1C%.
  arXiv  Detail & Related papers  (2024-08-20T13:19:06Z)
• Using Pre-training and Interaction Modeling for ancestry-specific disease prediction in UK Biobank [69.90493129893112]
  Recent genome-wide association studies (GWAS) have uncovered the genetic basis of complex traits, but show an under-representation of non-European descent individuals. Here, we assess whether we can improve disease prediction across diverse ancestries using multiomic data.
  arXiv  Detail & Related papers  (2024-04-26T16:39:50Z)
• Forecast reconciliation for vaccine supply chain optimization [61.13962963550403]
  Vaccine supply chain optimization can benefit from hierarchical time series forecasting. Forecasts of different hierarchy levels become incoherent when higher levels do not match the sum of the lower levels forecasts. We tackle the vaccine sale forecasting problem by modeling sales data from GSK between 2010 and 2021 as a hierarchical time series.
  arXiv  Detail & Related papers  (2023-05-02T14:34:34Z)
• STELAR: Spatio-temporal Tensor Factorization with Latent Epidemiological Regularization [76.57716281104938]
  We develop a tensor method to predict the evolution of epidemic trends for many regions simultaneously. STELAR enables long-term prediction by incorporating latent temporal regularization through a system of discrete-time difference equations. We conduct experiments using both county- and state-level COVID-19 data and show that our model can identify interesting latent patterns of the epidemic.
  arXiv  Detail & Related papers  (2020-12-08T21:21:47Z)
• Short Term Blood Glucose Prediction based on Continuous Glucose Monitoring Data [53.01543207478818]
  This study explores the use of Continuous Glucose Monitoring (CGM) data as input for digital decision support tools. We investigate how Recurrent Neural Networks (RNNs) can be used for Short Term Blood Glucose (STBG) prediction.
  arXiv  Detail & Related papers  (2020-02-06T16:39:44Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.