Presenting DiaData for Research on Type 1 Diabetes

2025-08-05

AI Summary

Early hypoglycemia prediction for type 1 diabetes (T1D) is hindered by the scarcity of high-quality, large-scale real-world data. Method: We systematically integrated 15 heterogeneous datasets to construct DiaData—the first standardized, large-scale T1D database—comprising 2,510 subjects and 149 million 5-minute-resolution glucose measurements. We developed a novel data quality assessment framework to address missingness and class imbalance; identified significant temporal correlations between glucose and heart rate 15–55 minutes prior to hypoglycemic events; and curated two structured sub-databases incorporating demographic and heart rate features. Contribution/Results: DiaData achieves balanced gender and age distributions, with hypoglycemic events constituting 4% of all glucose readings. This enables robust training and validation of machine learning models, substantially advancing the development of clinically deployable early-warning systems for T1D hypoglycemia.

Technology Category

• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data
• Machine Learning: Hardware-aware ML
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Web Mining and Content Analysis: Web data quality in the era of algorithmically-generated content
• Security and Privacy: Data transparency and provenance

Abstract

Type 1 diabetes (T1D) is an autoimmune disorder that leads to the destruction of insulin-producing cells, resulting in insulin deficiency, as to why the affected individuals depend on external insulin injections. However, insulin can decrease blood glucose levels and can cause hypoglycemia. Hypoglycemia is a severe event of low blood glucose levels ($le$70 mg/dL) with dangerous side effects of dizziness, coma, or death. Data analysis can significantly enhance diabetes care by identifying personal patterns and trends leading to adverse events. Especially, machine learning (ML) models can predict glucose levels and provide early alarms. However, diabetes and hypoglycemia research is limited by the unavailability of large datasets. Thus, this work systematically integrates 15 datasets to provide a large database of 2510 subjects with glucose measurements recorded every 5 minutes. In total, 149 million measurements are included, of which 4% represent values in the hypoglycemic range. Moreover, two sub-databases are extracted. Sub-database I includes demographics, and sub-database II includes heart rate data. The integrated dataset provides an equal distribution of sex and different age levels. As a further contribution, data quality is assessed, revealing that data imbalance and missing values present a significant challenge. Moreover, a correlation study on glucose levels and heart rate data is conducted, showing a relation between 15 and 55 minutes before hypoglycemia.

Problem

• Lack of large datasets for T1D and hypoglycemia research
• Need for accurate glucose level prediction to prevent hypoglycemia
• Challenges in data quality like imbalance and missing values

Innovation

• Integrates 15 datasets into large database
• Uses machine learning for glucose prediction
• Assesses data quality and missing values