DailyDose decision support recommender engine based on machine learning and artificial intelligence to improve glycemic control in type 1 diabetes (T1D)

Case ID:
2704
Web Published:
4/30/2020
Description:

Summary
Utilizing individualized behavioral trends, historical glucose measurement data (CGM), and insulin data, the DailyDose algorithm provides insulin dosage recommendations for patients with type 1 diabetes.

Technology Overview
The DailyDose algorithm provides daily and weekly insulin dosage recommendations for patients with type 1 diabetes who are using multiple daily injection or pump therapies. The algorithm utilizes machine-learning approaches to reach optimal dosage recommendations for a given set of inputs. Using personalized patient data, historical glucose measurement data (CGM), and insulin data, the algorithm can determine proper insulin dosages even if meals and other corrections are logged incorrectly. Additionally, the advanced meal bolus calculator can account for current glucose levels, insulin on board, carbohydrates consumed, the rate of glucose changed, along with anticipated and recent exercise. 

Studies have indicated that patients can improve time within a target glucose range from 63.18% to 77.5% after 12 weeks of use. In that time, hypoglycemia (<70 mh/dL) can also be reduced from 3.72% to 1.42%.

Patent Information:
Category(s):
Software
For Information, Contact:
Michele Gunness
Senior Technology Development Manager
Oregon Health & Science University
(503) 494-8200
gunnessm@ohsu.edu
Inventors:
Peter Jacobs
Jessica Castle
Joseph El Youssef
Clara Mosquera-Lopez
Nichole Tyler
Robert Dodier
Leah Wilson
Keywords:
Software
Software - Other
© 2020. All Rights Reserved. Powered by Inteum