Tailoring Adverse Event Prediction in Type 1 Diabetes with Patient-Specific Deep Learning Models

• URL: http://arxiv.org/abs/2601.14917v1
• Date: Wed, 21 Jan 2026 11:57:51 GMT
• Title: Tailoring Adverse Event Prediction in Type 1 Diabetes with Patient-Specific Deep Learning Models
• Authors: Giorgia Rigamonti, Mirko Paolo Barbato, Davide Marelli, Paolo Napoletano,
• Abstract summary: This paper presents a deep learning-based approach for personalized blood glucose prediction.<n>Unlike traditional generalized models, our method accounts for individual variability, enabling more effective subject-specific predictions.<n>Results show that personalized models significantly improve the prediction of adverse events.
• Score: 4.201695852847293
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Effective management of Type 1 Diabetes requires continuous glucose monitoring and precise insulin adjustments to prevent hyperglycemia and hypoglycemia. With the growing adoption of wearable glucose monitors and mobile health applications, accurate blood glucose prediction is essential for enhancing automated insulin delivery and decision-support systems. This paper presents a deep learning-based approach for personalized blood glucose prediction, leveraging patient-specific data to improve prediction accuracy and responsiveness in real-world scenarios. Unlike traditional generalized models, our method accounts for individual variability, enabling more effective subject-specific predictions. We compare Leave-One-Subject-Out Cross-Validation with a fine-tuning strategy to evaluate their ability to model patient-specific dynamics. Results show that personalized models significantly improve the prediction of adverse events, enabling more precise and timely interventions in real-world scenarios. To assess the impact of patient-specific data, we conduct experiments comparing a multimodal, patient-specific approach against traditional CGM-only methods. Additionally, we perform an ablation study to investigate model performance with progressively smaller training sets, identifying the minimum data required for effective personalization-an essential consideration for real-world applications where extensive data collection is often challenging. Our findings underscore the potential of adaptive, personalized glucose prediction models for advancing next-generation diabetes management, particularly in wearable and mobile health platforms, enhancing consumer-oriented diabetes care solutions.

Related papers

• Beyond Traditional Diagnostics: Transforming Patient-Side Information into Predictive Insights with Knowledge Graphs and Prototypes [55.310195121276074]
  We propose a Knowledge graph-enhanced, Prototype-aware, and Interpretable (KPI) framework to predict diseases.<n>It integrates structured and trusted medical knowledge into a unified disease knowledge graph, constructs clinically meaningful disease prototypes, and employs contrastive learning to enhance predictive accuracy.<n>It provides clinically valid explanations that closely align with patient narratives, highlighting its practical value for patient-centered healthcare delivery.
  arXiv  Detail & Related papers  (2025-12-09T05:37:54Z)
• 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)
• Toward Short-Term Glucose Prediction Solely Based on CGM Time Series [4.7066018521459725]
  TimeGlu is an end-to-end pipeline for short-term glucose prediction based on CGM time series data. It achieves state-of-the-art performance without the need for additional personal data from patients.
  arXiv  Detail & Related papers  (2024-04-18T06:02:12Z)
• MedDiffusion: Boosting Health Risk Prediction via Diffusion-based Data Augmentation [58.93221876843639]
  This paper introduces a novel, end-to-end diffusion-based risk prediction model, named MedDiffusion. It enhances risk prediction performance by creating synthetic patient data during training to enlarge sample space. It discerns hidden relationships between patient visits using a step-wise attention mechanism, enabling the model to automatically retain the most vital information for generating high-quality data.
  arXiv  Detail & Related papers  (2023-10-04T01:36:30Z)
• Machine Learning based prediction of Glucose Levels in Type 1 Diabetes Patients with the use of Continuous Glucose Monitoring Data [0.0]
  Continuous Glucose Monitoring (CGM) devices offer detailed, non-intrusive and real time insights into a patient's blood glucose concentrations. Leveraging advanced Machine Learning (ML) Models as methods of prediction of future glucose levels, gives rise to substantial quality of life improvements.
  arXiv  Detail & Related papers  (2023-02-24T19:10:40Z)
• SANSformers: Self-Supervised Forecasting in Electronic Health Records with Attention-Free Models [48.07469930813923]
  This work aims to forecast the demand for healthcare services, by predicting the number of patient visits to healthcare facilities. We introduce SANSformer, an attention-free sequential model designed with specific inductive biases to cater for the unique characteristics of EHR data. Our results illuminate the promising potential of tailored attention-free models and self-supervised pretraining in refining healthcare utilization predictions across various patient demographics.
  arXiv  Detail & Related papers  (2021-08-31T08:23:56Z)
• On the Robustness of Pretraining and Self-Supervision for a Deep Learning-based Analysis of Diabetic Retinopathy [70.71457102672545]
  We compare the impact of different training procedures for diabetic retinopathy grading. We investigate different aspects such as quantitative performance, statistics of the learned feature representations, interpretability and robustness to image distortions. Our results indicate that models from ImageNet pretraining report a significant increase in performance, generalization and robustness to image distortions.
  arXiv  Detail & Related papers  (2021-06-25T08:32:45Z)
• Deep Personalized Glucose Level Forecasting Using Attention-based Recurrent Neural Networks [5.250950284616893]
  We study the problem of blood glucose forecasting and provide a deep personalized solution. We analyze the data and detect important patterns. We empirically show the efficacy of our model on a real dataset.
  arXiv  Detail & Related papers  (2021-06-02T01:36:53Z)
• Interpreting Deep Glucose Predictive Models for Diabetic People Using RETAIN [4.692400531340393]
  We study the RETAIN architecture for the forecasting of future glucose values for diabetic people. Thanks to its two-level attention mechanism, the RETAIN model is interpretable while remaining as efficient as standard neural networks.
  arXiv  Detail & Related papers  (2020-09-08T13:20:15Z)
• 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.