Multi-Modal Wearable and Biological Data Integration Reveals Physiological Signatures of Weight Variation in Overweight and Obese Adults: An AI4Food Machine Learning Study

In 93 overweight and obese adults in a 30-day crossover at IMDEA Nutrition Institute, nocturnal respiratory rate and skin temperature are significantly associated with weight variation.

• Why it matters: Weight change in obesity is linked to complex physiology, but multi-source wearable and biological signals are not yet integrated to identify measurable physiological signatures.
• What they did: The study combined questionnaires and food diaries with Fitbit Sense and FreeStyle Libre 2 glucose sensing, plus sleep, skin temperature, nutrition, and gut microbiome data, then used clustering, association testing, and machine learning.
• The result: Random Forest models in participant-level five-fold cross-validation reached mean accuracies of 64.3% for thermoregulation and 61.7% for glucose phenotype prediction, while microbiome composition stayed almost stable over the short intervention.

AI summary of the abstract · may contain errors · not medical advice · read the paper.