But, with the introduction of this new algorithm-based app, they can heave a sigh of relief soon as it can take some of the stress away.
Though there is still some work that needs to be done on its process, the idea behind this personalized technology is to predict the impact of each meal on a user’s blood sugar levels.
Type 2 diabetes now affects more than 29 million people in the United States. An additional 86 million adults are thought to have prediabetes, which can develop into type 2 diabetes if lifestyle changes are not implemented.
If levels are too high for prolonged periods of time, serious health complications can arise.
Medication is given to help manage sugar level fluctuations, but exercise and diet also play a substantial role.
Although the impact of specific food types on glucose levels can be estimated, it is not an exact science.
Effects can vary substantially between individuals and they can even vary within an individual dependent on a range of factors.
A report, published in PLOS Computational Biology this week, explains how a group of scientists have integrated an algorithm into an app called Glucoracle, which goes some way toward solving this problem.
David Albers, Ph.D., associate research scientist in biomedical informatics at Columbia University Medical Center (CUMC) in New York and lead author of the study, explains: “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.”
To tackle this problem, Albers and his team are attempting to design an algorithm that can help individuals to make more informed dietary decisions.
Predicting glucose levels
Albers explains how the app works: “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.”
The algorithm uses data assimilation, a technique that is utlized in a range of modern applications, including weather prediction.
Data assimilation takes regularly updated information – including blood sugar measurements and nutritional information – collates it, and then creates a mathematical model of an individual’s response to glucose.
Lena Mamykina, Ph.D., assistant professor of biomedical informatics at CUMC and a study co-author, explains: “The data assimilator is continually updated with the user’s food intake and blood glucose measurements, personalizing the model for that individual.”
Users of Glucoracle can upload pictures of a particular meal with rough estimates of its nutritional content, along with fingerstick blood measurements. The app can then provide an immediate prediction of post-meal blood sugar levels.
The app must be used for a week before it starts to generate predictions.
This allows the data assimilator to learn how the individual user responds to various types of food. The estimate and forecast are then adjusted for accuracy over time.
How well does it work?
In the non-diabetic participants, the readings quite accurately matched the genuine glucose measurements.
For the three participants with diabetes, the results were less accurate. The researchers believe that this might be due to physiological fluctuations in the patients or a parameter error.
However, the predictions were “still comparable” to those of certified diabetes educators.
Although the results are not perfect, Albers is not disheartened. Instead, he says:
“There’s certainly room for improvement. 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.”
A larger clinical trial is now planned, and the researchers hope that the app will be ready for widespread use in two years.