Blood Glucose Level Prediction Challenge
UPDATE: The second Blood Glucose Level Prediction (BGLP) Challenge concluded on August 30, 2020. Results and papers are now available here. The rest of this webpage remains as it was, describing the past event.
Blood glucose level prediction is a challenging task for AI researchers, with the potential to improve the health and wellbeing of people with diabetes. Knowing in advance when blood glucose is approaching unsafe levels provides time to pro-actively avoid hypo- and hyperglycemia and their concomitant complications. The drive to perfect an artificial pancreas has increased the interest in using machine learning (ML) approaches to improve prediction accuracy. Work in this area has been hindered, however, by a lack of real patient data; some researchers are only able to work on simulated patient data.
In this second Blood Glucose Level Prediction (BGLP) Challenge, researchers will come together to compare the efficacy of different prediction approaches on a standard set of real patient data, the OhioT1DM Dataset. Participants will use the same data with their own diverse prediction approaches, and we will come together as a group to compare results and share experiences. We aim to foster continuing collaboration and accelerate the pace of progress in the field.
The OhioT1DM Dataset
The OhioT1DM Dataset contains 8 weeks worth of data for each of 12 people with type 1 diabetes. These people were all on insulin pump therapy with continuous glucose monitoring (CGM). They provided blood glucose data, insulin data, self-reported life-event data, and data from physiological fitness bands. Data for the first 6 patients was released for the first BGLP Challenge, which was held in conjunction with the 3rd International Workshop on Knowledge Discovery in Healthcare Data, at IJCAI-ECAI 2018, in Stockholm, Sweden. Data for the second group of 6 patients is being released for this BGLP Challenge.
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 exercise, sleep, work, stress, and illness; and physiological data from fitness bands. A paper fully describing the dataset is available here: The OhioT1DM Dataset for Blood Glucose Level Prediction: Update 2020.
To protect the data and ensure that it is used only for research purposes, a Data Use Agreement (DUA) is required to obtain the data. Additional information, with a link to the DUA request form, is available at: The OhioT1DM Dataset.
Participate in the Challenge!
Participants will receive training data upon execution of a DUA and will receive test data two weeks prior to the due date for results. To facilitate "apples to apples" comparisons, participants are asked to report results for 30 and 60-minute prediction horizons, as specified in The BGLP Challenge Rules. Note that, to foster the aims of promoting collaboration and accelerating progress in the field, at least one participant from each team must attend the KDH workshop. To facilitate future work and replication of published results, we require that participants make the code publicly available, by including a link in the camera ready paper. The code will also be linked from this website immediately following the workshop.
UPDATE: ECAI 2020 is now a virtual conference, so the BGLP Challenge will be held online. Presentations will be pre-recorded, but there will be interactive question and answer sessions following each talk.
Submission requirements are detailed in the Call for Papers.
To receive notices about the BGLP Challenge and/or to participate in discussions, you can subscribe to the BGLP Challenge News group.
Training and Development Data Release: January 10, 2020
Test Data Release: March 10, 2020
Results Submission Deadline: April 26, 2020
System Description Paper Submission Deadline: May 10, 2020
Notification Date: June 15, 2020
Camera-Ready Deadline: July 15, 2020
Workshop: Saturday, August 29 and Sunday, August 30, 2020
The BGLP Challenge is supported by grant 1R21EB022356 from the National Institutes of Health (NIH).