Credit score: CC0 Public Area
In case you eat a snack—a meatball, say, or a marshmallow—how will it have an effect on your blood sugar? It is a surprisingly tough query; the physique’s glycemic response to completely different meals varies primarily based on particular person genetics, microbiomes, hormonal fluctuations, and extra. Due to that, offering customized dietary recommendation—which may help sufferers handle diabetes, weight problems, and cardiovascular illnesses, amongst different circumstances—requires pricey and intrusive testing, making it exhausting to ship efficient care at scale.
In a paper within the Journal of Diabetes Science and Know-how, researchers at Stevens Institute of Know-how provide a brand new method: a data-sparse mannequin able to precisely predicting particular person glycemic responses without having for blood attracts, stool samples, or different disagreeable testing. The important thing to their method? Maintaining observe of what individuals really eat.
“It might sound obvious, but until now most research has focused on macronutrients, such as grams of carbohydrates, instead of the specific foods that people are eating,” explains Dr. Samantha Kleinberg, Farber Chair Professor of Laptop Science. “We’ve shown that by analyzing food types, it’s possible to make highly accurate predictions with far less data.”
Dr. Kleinberg’s workforce studied two datasets that embody each detailed meals diaries and steady glucose monitor knowledge for nearly 500 individuals with diabetes (each sorts 1 and a pair of) primarily based in the US and China. Utilizing present meals databases and ChatGPT, they labeled every meal based on macronutrient content material and in addition leveraged the construction of meals (so meats are extra related to one another than to cheeses), enabling them to distinguish between nutritionally equal meals.
By coaching an algorithm utilizing each dietary knowledge and meals options, plus a number of demographic particulars, the workforce was capable of predict every particular person’s glycemic response to every meals with just about the identical ranges of accuracy present in prior research that included detailed microbiome knowledge and different hard-to-collect info.
“We still don’t know why including the food features makes such a big difference,” Dr. Kleinberg says. It is potential that meals info is a proxy for micronutrients that drive glycemic responses, or that the bodily properties of sure meals lead individuals to eat or digest them otherwise.
“What’s clear, though, is that when it comes to blood sugar, there’s more at work than just macronutrients,” Dr. Kleinberg says.
By specializing in meals sorts, the workforce was additionally capable of discover particular person variations in glycemic responses.
“Because people eat the same meals again and again, the data gives us visibility into the way that individual responses to specific foods change over time,” Dr. Kleinberg explains.
The workforce discovered that together with knowledge about menstrual cycles of their mannequin accounted for a lot of the intra-subject variation, suggesting that shifting hormone ranges might play an vital position in mediating particular person glycemic responses.
The workforce’s mannequin additionally precisely predicts glycemic response for each U.S. and Chinese language populations—an vital discovering, since microbiome-based fashions typically battle to ship correct outcomes throughout completely different cultural contexts.
“We don’t need data on a specific regional population to be able to make predictions there,” Dr. Kleinberg explains.
The brand new mannequin can be highly effective sufficient to foretell a person’s glycemic responses primarily based on demographic knowledge, with out custom-made coaching on meals logs or different customized knowledge. Because of this, clinicians might doubtlessly use the mannequin to supply dietary recommendation throughout an preliminary assembly with a affected person, with out the necessity for laborious meals logging or intrusive testing.
“We can offer better recommendations if we have more data, but we can get very good results with no personalized information at all,” Dr. Kleinberg explains. “That means we can give patients useful advice right away—and hopefully that will motivate them to keep going.”
Subsequent, the workforce plans to refine their mannequin utilizing bigger datasets, and to discover whether or not including microbiome knowledge will increase their mannequin’s accuracy.
“That’s the big question, because if food information alone gives us everything we need, there might be no need to collect stool samples or do other tests,” Dr. Kleinberg says. “That could make personalized nutrition more affordable and accessible for everyone.”
Extra info:
Yiheng Shen et al, Predicting Postprandial Glycemic Responses With Restricted Information in Kind 1 and Kind 2 Diabetes, Journal of Diabetes Science and Know-how (2025). DOI: 10.1177/19322968251321508
Offered by
Stevens Institute of Know-how
Quotation:
Information-sparse mannequin opens door to customized diet—with out the necessity for pesky samples (2025, April 10)
retrieved 10 April 2025
from https://medicalxpress.com/information/2025-04-sparse-door-personalized-nutrition-pesky.html
This doc is topic to copyright. Aside from any honest dealing for the aim of personal examine or analysis, no
half could also be reproduced with out the written permission. The content material is offered for info functions solely.
