Personalized glucose prediction using in situ data only

Abstract The worldwide rise in blood glucose levels is a major health concern, as various metabolic diseases become increasingly common. Diet, a modifiable health behavior, is a primary target for the preventive management of glucose levels. Recent studies have shown that blood glucose responses after meals (post-prandial glucose responses, PPGR) can vary greatly among individuals, even with identical food consumption, and suggested that the gut microbiota might play an important role in these differences. While accurate glucose response prediction has been achieved using various features like microbiome data and blood parameters, the exact influence of each individual factor on the prediction is still not clear. Here, we show that a machine learning algorithm with data collected from a digital cohort with over 1,000 participants can achieve high accuracy in PPRG prediction. Interestingly, we find that the best PPGR prediction model only requires glycemic and temporally resolved diet data. The demonstrated ability to predict PPGR with high accuracy using only data collected in situ, without the need for biological lab analysis, offers a path towards highly scalable personalized nutrition and glucose management strategies.