The OhioT1DM Dataset

The OhioT1DM dataset is available to researchers interested in improving the health and wellbeing of people with type 1 diabetes. While it was originally developed for blood glucose level prediction, it is also well suited to other types of data analysis and knowledge discovery. For example, it could be used to examine the effects of food or exercise or to consider whether phyiological signals from fitness bands provide actionable information.

The OhioT1DM Dataset contains 8 weeks worth of data for each of 6 people with type 1 diabetes. These people were all on insulin pump therapy with continuous glucose monitoring (CGM). They reported life-event data and provided physiological data from a Basis Peak fitness band. The dataset includes: a CGM blood glucose level every 5 minutes; blood glucose levels from periodic self-monitoring of blood glucose (finger sticks); insulin doses, both bolus and basal; self-reported meal times with carbohydrate estimates; self-reported times of sleep, work, and exercise; and 5-minute aggregations of heart rate, galvanic skin response (GSR), skin temperature, air temperature, and step count. A paper fully describing the dataset is available: The OhioT1DM Dataset for Blood Glucose Level Prediction.

To protect the data and ensure that it is used only for research purposes, a Data Use Agreement (DUA) is required. The DUA must be signed by a legal signatory for the research institution as well as by the principal investigator. Researchers can click here to Request the Data Use Agreement for the OhioT1DM Dataset. Once a DUA is executed, the dataset will be released, along with the OhioT1DM Viewer, a tool for data visualization.

The OhioT1DM Dataset was first released for the Blood Glucose Level Prediction (BGLP) Challenge held at the Federated AI Meeting in Stockholm, Sweden, in July, 2018. The BGLP Challenge was part of The 3rd International Workshop on Knowledge Discovery in Healthcare Data. The Workshop Proceedings are available online from CEUR-WS.org.

 

This work is supported by grant 1R21EB022356 from the National Institutes of Health (NIH).