Artificial intelligence helps Americans gain additional insight into the quality of carbs they consume.
In a new study published in The American Journal of Clinical Nutrition, a team of researchers created the first national glycemic index (GI) and glycemic load (GL) database, providing valuable insights into the quality of carbohydrates consumed by Americans. The initiative is especially useful for diabetics and others who need to track their carb intake on a daily basis.
The glycemic index (GI) is a ranking system for carbohydrates that is based on their effect on blood glucose levels, which is a vital measurement for diabetics, who need to keep these levels as steady as possible, as well as others who may be carb counting to maintain a healthy weight. When ingested, carbohydrates are first met by enzyme amylase in saliva, which breaks down starches into simpler sugars. As the food travels to the stomach, it mixes with gastric juices, further breaking down the carbs. In the small intestine, pancreatic amylase continues the process, converting the carbohydrates into simple sugars like glucose. The sugars are then absorbed through the walls of the small intestine, making their way into the bloodstream, where they are transported to cells and used for energy.
This means, even when complex carbs are consumed, they converted via digestion into sugars that can cause dysregulation of blood glucose levels. With Type 1 diabetes, the body’s immune system mistakenly attacks and destroys insulin-producing beta cells in the pancreas, leading to little or no insulin production. Without insulin, glucose cannot enter cells and remains in the bloodstream, causing high blood sugar levels. In type 2 diabetes, the body either becomes resistant to insulin or does not produce enough of it, which prevents glucose from entering cells efficiently, also resulting in elevated blood sugar levels.
The new tool measures how quickly and how much a carbohydrate-containing food raises blood sugar levels compared to a standard reference food, usually glucose or white bread. Foods are scored on a scale from 0 to 100, with higher values indicating a more rapid increase in blood sugar. Low-GI foods (55 or less) cause slower, more gradual rises, while high-GI foods (70 or more) lead to quicker spikes. This insight allows for better blood glucose regulation.
To carry out this study, Karen Della Corte, a professor of nutrition and dietetics at Brigham Young University (BYU), collaborated with her husband, Dennis Della Corte, a professor of physics and astronomy at BYU. They developed an artificial intelligence (AI)-enabled model to analyze nearly 10,000 foods based on their GI and GL values. Utilizing data from the National Health and Nutrition Examination Survey (NHANES), which includes a comprehensive sample of various foods consumed daily by Americans, the AI matched food descriptions with corresponding GI/GL values, resulting in the first-ever national database.
“Using AI to create the GI database was a novel application for us and represents an advancement in nutritional research methodology,” said Karen Della Corte. “Moving forward, many new and important questions can now be investigated using this database, particularly regarding the role of GI and GL in chronic disease risk in the U.S.”
The team emphasized the importance of prioritizing low-GI carbohydrates in the diet. “Making swaps from refined grains to whole grains can help improve the healthfulness of the diet and lower the overall dietary GI,” said Della Corte, who suggested individuals add whole grains, beans, lentils, chickpeas, brown or wild rice, quinoa, barley, steel-cut or rolled oats, non-starchy vegetables, fruits, and nuts to their diet to better regulate blood sugar.
Sources:
Cracking the carb code: Researchers create new glycemic index database to improve dietary awareness
Researchers Use AI To Create National Glycemic Index Database
Join the conversation!