Collaborative Research: FDT-BioTech: NOURISH: Nutritional Optimization Using Real-time Integrated Sensors and Health Digital Twins

About This Grant

Metabolic diseases, such as diabetes, affect millions of people worldwide, profoundly impacting health, quality of life, and healthcare costs. However, traditional approaches to dietary management rely on generalized guidelines and intermittent blood tests, which limit their effectiveness in capturing the real-time metabolic changes necessary for personalized care. To address this, researchers at Purdue University and Oregon Health & Science University (OHSU) are developing NOURISH, an innovative digital twin technology that continuously tracks multiple metabolic indicators using wearable biosensors. By pairing these real-time measurements with advanced computational models, NOURISH simulates whole-body metabolism in individual patients and provides tailored dietary recommendations. The initial validation of these sensors will focus on healthy volunteers with controlled metabolic challenges, providing foundational data necessary for future clinical applications. NOURISH aims to significantly improve personalized dietary interventions and metabolic health outcomes. The project will also provide interdisciplinary training opportunities in advanced technologies and prepare a diverse and skilled workforce to meet critical national needs for healthcare innovation. The NOURISH project directly aligns with the NSF’s Foundations for Digital Twins (FDT-BioTech) program by developing advanced biosensors and integrating them into a physics-informed whole-body metabolism digital twin (WBM-DT). The team will retrofit FDA-approved continuous glucose monitors (CGMs) with tellurene-based nanosensors, enabling real-time measurement of multiple clinically relevant biomarkers, including glucose, lactate, β-hydroxybutyrate, branched-chain amino acids, glutamate, glutamine, acetoacetate, and glycerol. Sensor validation will be conducted initially in healthy adult volunteers using mixed-meal tests and standardized metabolic challenges. Data will be assimilated using an ensemble Kalman filter and Bayesian uncertainty quantification (UQ) to calibrate mechanistic metabolic models that accurately simulate systemic body-level metabolic fluxes. This framework will be combined with probabilistic AI-based control algorithms to deliver precise nutritional guidance tailored to individual physiological states. Additionally, large-scale, bias-audited synthetic cohorts will validate the model's accuracy, reliability, and fairness across diverse populations. These foundational methodological advancements will provide essential regulatory science toolkits for precision nutrition, facilitating broader biomedical applications for managing metabolic and chronic diseases. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.