Wrist Photoplethysmography Predicts Dietary Information
This work addresses the problem of passive dietary monitoring for health and wellness applications, representing a novel application of existing methods to new data.
The study tackled the problem of whether wearable photoplethysmography (PPG) can predict dietary information, and found that PPG nontrivially predicts meal content, with predictability decreasing as PPGs are farther from meals, and it increased AUC by 11% for intake and satiety tasks across cohorts.
Whether wearable photoplethysmography (PPG) contains dietary information remains unknown. We trained a language model on 1.1M meals to predict meal descriptions from PPG, aligning PPG to text. PPG nontrivially predicts meal content; predictability decreases for PPGs farther from meals. This transfers to dietary tasks: PPG increases AUC by 11% for intake and satiety across held-out and independent cohorts, with gains robust to text degradation. Wearable PPG may enable passive dietary monitoring.