Description:
Summary
The DailyDose algorithm provides an artificial intelligence decision support system for improved management of blood glucose in patients with type 1 diabetes using multiple daily injection therapy.
Technology Overview
Up to 40% of patients with type 1 diabetes use multiple daily injections of insulin to manage their condition; however, dosing regimens can be complicated by daily changes in insulin sensitivity due to factors like meals and exercise. Failure to dose insulin properly can result in diabetic ketoacidosis and hypoglycemia, which may lead to coma or death. While many smartphone apps are available to help people better manage their diabetes, most of these have not shown clinical efficacy.
The DailyDose algorithm provides daily and weekly insulin dosage recommendations for people living 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.
Features of the algorithm include:
- Personalized-approach based on individual people’s data for improved insulin dosage recommendations.
- Advanced meal bolus calculator, which can account for current glucose levels, insulin on board, carbohydrates consumed, and the rate of glucose changed.
- Correction and adjustment for anticipated and recent exercise.
- Incorporation of an expert-knowledge quality control algorithm for improved user safety.
- Demonstrated alignment with clinician recommendations, with 97% of DailyDose recommendations delivered across 100 weeks of data passing an endocrinologist safety review.
Publications
Castle et al., "Assessment of a Decision Support System for Adults wtih Type 1 Diabetes on Multiple Daily Insulin Injections." Diabetes Technology & Therapeutics. 2022: 892-897. Link
Tyler et al., “An artificial intelligence decision support system for the management of type 1 diabetes.” Nature Metabolism 2(2020): 612-619. Link
Licensing Opportunity
This technology is available for licensing.