Personalizing dietary interventions by predicting individual vulnerability to glucose excursions DOI Open Access
Victoria Brügger, Tobias Kowatsch, Mia Jovanova

et al.

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 7, 2024

Abstract Elevated postprandial glucose levels pose a global epidemic and are crucial in cardiometabolic disease management prevention. A major challenge is inter-individual variability, which limits the effectiveness of population-wide dietary interventions. To develop personalized interventions, it critical to first predict person’s vulnerability excursions—or elevated post-meal relative personal baseline—with minimal burden. We examined feasibility models future excursions daily lives 69 Chinese adults with type-2 diabetes ( M age=61.5; 50% women; 2’595 observations). developed machine learning models, trained on past individual context meal-based observations, employing low-burden (continuous monitoring) or additional high-burden (manual meal tracking) approaches. Personalized predicted (F1-score: =74%; median=78%), some individuals being more predictable than others. The low burden-models performed better for those consistent patterns healthier glycemic profiles. Notably, no two shared same context-based predictors. This study among sample diabetes. Findings can help personalize just-in-time-adaptive interventions unique live, thereby helping improve management.

Language: Английский

Personalizing dietary interventions by predicting individual vulnerability to glucose excursions DOI Open Access
Victoria Brügger, Tobias Kowatsch, Mia Jovanova

et al.

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 7, 2024

Abstract Elevated postprandial glucose levels pose a global epidemic and are crucial in cardiometabolic disease management prevention. A major challenge is inter-individual variability, which limits the effectiveness of population-wide dietary interventions. To develop personalized interventions, it critical to first predict person’s vulnerability excursions—or elevated post-meal relative personal baseline—with minimal burden. We examined feasibility models future excursions daily lives 69 Chinese adults with type-2 diabetes ( M age=61.5; 50% women; 2’595 observations). developed machine learning models, trained on past individual context meal-based observations, employing low-burden (continuous monitoring) or additional high-burden (manual meal tracking) approaches. Personalized predicted (F1-score: =74%; median=78%), some individuals being more predictable than others. The low burden-models performed better for those consistent patterns healthier glycemic profiles. Notably, no two shared same context-based predictors. This study among sample diabetes. Findings can help personalize just-in-time-adaptive interventions unique live, thereby helping improve management.

Language: Английский

Citations

1