PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
January 17, 2019Journal of Diabetes Science and TechnologyOpen Access

Prediction of Hypoglycemia During Aerobic Exercise in Adults With Type 1 Diabetes

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Do machine learning algorithms accurately predict hypoglycemia during aerobic exercise in adults with Type 1 Diabetes?

Population

55 adults with Type 1 Diabetes using sensor augmented pump therapy, automated insulin delivery therapy, or…

Design

Other

Follow-up

during exercise

Authors

RRRavi ReddyNRNavid ResalatLWLeah M. Wilson

Discussion

Loading...

Member takes

Overview

May encourage exercise in T1D by predicting hypoglycemia; hypothesis-generating pending prospective validation.

Structured PICO

Do machine learning algorithms accurately predict hypoglycemia during aerobic exercise in adults with Type 1 Diabetes?

P
Population
55 adults with Type 1 Diabetes (43 in training set with 154 exercise observations, 12 in validation set with 90 exercise observations) using sensor augmented pump therapy, automated insulin delivery therapy, or automated insulin and glucagon therapy
I
Intervention
Hypoglycemia prediction algorithms (Model 1: decision tree using heart rate and glucose; Model 2: random forest model)
O
Outcome
Accuracy of predicting hypoglycemia during aerobic exercise

A simple heuristic (heart rate >121 bpm and glucose <182 mg/dL) and a more complex random forest model can accurately predict hypoglycemia during aerobic exercise in adults with Type 1 Diabetes.

Cite This Study

Reddy et al. (2019) studied this question.

synapsesocial.com/papers/6a76ec739d809400dabd4bf8https://doi.org/10.1177/1932296818823792
View Full Paper
Ask AI
Bookmark
Share