A new smartphone application with a personalized algorithm can help predict the impact of particular food based on an individual’s blood sugar levels.
The app—Glucoracle—developed by researchers from the Columbia University Medical Center (CUMC), will help individuals with type II diabetes to have a better handle on their glucose levels, which is crucial in preventing or controlling major complications tied to the disease.
“While we know the general effect of different types of food on blood glucose, the detailed effects can vary widely from one person to another and for the same person over time,” lead author David Albers, Ph.D., associate research scientist in Biomedical Informatics at CUMC, said in a statement. “Even with expert guidance, it’s difficult for people to understand the true impact of their dietary choices, particularly on a meal-to-meal basis.
“Our algorithm, integrated into an easy-to-use app, predicts the consequences of eating a specific meal before the food is eaten, allowing individuals to make better nutritional choices during mealtime,” he added.
The researchers developed the algorithm using data assimilation, where a mathematical model of a person’s response to glucose is regularly updated with blood sugar measurements and nutritional information to improve the model’s predictions.
“The data assimilator is continually updated with the user’s food intake and blood glucose measurements, personalizing the model for that individual,” co-study leader Lena Mamykina, Ph.D., assistant professor of biomedical informatics at Columbia, whose team designed and developed the Glucoracle app, said in a statement.
Users can upload fingerstick blood measurements and a photo of a particular meal with a rough estimate of the nutritional content of the meal to the app, which will provide the user with an immediate prediction of the post-meal blood sugar levels.
The estimate and forecast are then adjusted for accuracy and the app generates predictions after the data assimilator learns how the user responds to different foods after a week.
The researchers tested the app on five individuals—three of which had type II diabetes and two who did not have the disease. The app was slightly less accurate for the sufferers of the disease, possibly due to fluctuations in the physiology of patients with diabetes or parameter error. The app’s prediction was comparable to the actual glucose measurements for the two nondiabetic individuals.
“There’s certainly room for improvement,” Albers said. “This evaluation was designed to prove that it’s possible, using routine self-monitoring data, to generate real-time glucose forecasts that people could use to make better nutritional choices.
“We have been able to make an aspect of diabetes self-management that has been nearly impossible for people with type 2 diabetes more manageable. Now our task is to make the data assimilation tool powering the app even better,” he added.