In the Journals Plus

Speech recognition for insulin bolus calculator improves postprandial glycemia

Show Citation

January 15, 2018

Patients with type 1 diabetes on insulin pump therapy who used an automatic bolus calculator designed to recognize voice descriptions of meals were more likely than those without the technology to have 2-hour postprandial glucose levels within the target range, according to findings reported in the Journal of Diabetes.

The VoiceDiab system consists of an Android-system smartphone with an app that communicates with a set of three servers, according to Piotr Foltynski, PhD, of the Nalecz Institute of Biocybernetics and Biomedical Engineering at the Polish Academy of Sciences in Warsaw, and colleagues. The servers are responsible for automatic speech recognition and transformation of the verbal description of meals into text, analysis of the textual description of the meal to determine its composition (ie, amount of protein, fat and carbohydrate), and calculation of the insulin dose compensating the meal, according to an algorithm.

“The primary reasons for the use of software applications in bolus calculation are technological opportunities and the reduction of errors in dose calculation by patients using bolus calculators,” Foltynski and colleagues wrote. “Low numeracy is associated with poor diabetes control; therefore, the use of bolus advisors may bring benefits to the patients.”

In a randomized, crossover study conducted between March 2015 and January 2016, researchers analyzed data from 44 adults with type 1 diabetes treated with continuous subcutaneous insulin infusion, assigned to a group in which the VoiceDiab system supported the bolus calculation for 4 days, or to an unsupported group, in which patients or their caregivers calculated boluses. Participants switched groups after a 14-day washout period, again completing a 4-day intervention period.

Patients used the VoiceDiab system on average 6.1 times per day.

Researchers found that more participants using the speech recognition technology experienced 2-hour postprandial glycemia between 70 mg/dL and 180 mg/dL vs. the unsupported group (58.6% in range vs. 46.6%; P = .031). There were no between-group differences for time spent in hypoglycemia, time spent in hyperglycemia, variations in glucose level or mean glucose level (P > .05 for all).

“The VoiceDiab system allows improving postprandial glucose control without increasing the time of hyper- or hypoglycemia, and therefore, it seems to be a promising tool for use by patients with diabetes treated with insulin,” the researchers wrote. – by Regina Schaffer

Disclosures: The Institute of Diabetology in Warsaw conducted this study. The authors report no relevant financial disclosures.