FDT-biotech:hybrid digital twins for predicting… (2625877) – Project Grant to Board Of Regents Of The…

Abstract: Skeletal muscle is essential for movement, strength, and healthy metabolism. When muscles do not respond well to aging, illness, or recovery from injury, people are more likely to become weak and have serious health problems. This project will develop a digital twin model, which is an accurate virtual replica, of skeletal muscle to better understand why muscle function declines in aging and disease, and how to improve recovery. The digital twin will be developed by combining machine learning and mathematical models with mouse studies and human measurements to predict how muscle adapts to various levels of activity and stress. In the future, the model will support strategies for rehabilitation, healthy aging, and precision health. The project may also reduce the need for some animal experiments by making computer simulations more informative. The educational goals of the project will increase workforce development in artificial intelligence and biomedical science through targeted course design and workshops. This project develops skeletal muscle adaptive analysis, assimilation, and augmented digital twin (SMA?-DT), a hybrid mechanistic-artificial intelligence (AI) framework for predicting skeletal muscle adaptation across activity and metabolic states. The project combines a dual physical twin design with computational modeling. A mouse physical twin provides controlled perturbations and multiscale molecular, structural, and functional measurements, while a human physical twin provides longitudinal physiological and behavioral data for translation and personalization. The digital twin is organized around five interacting biological modules: mechanical loading and mechanotransduction, protein turnover, inflammatory signaling, regeneration and fiber remodeling, and metabolic regulation. IT combines continuous-time mechanistic state-space modeling, bayesian data assimilation, latent-state inference, and learned residual dynamics to estimate partially observed physiological states from sparse, irregular, and multimodal data. The project also develops adaptive verification, validation, and uncertainty quantification (vvuq) methods, transferability assessment across species and perturbation regimes, and reproducible workflows for data integration, benchmarking, and reporting. Together, these components create a dynamic feedback framework in which physical observations update the digital twin and model predictions guide validation and refinement. More broadly, the work establishes skeletal muscle as a rigorous testbed for experimentally grounded biomedical digital twins and contributes reusable methods for latent-state inference, cross-scale validation, and adaptive model operation in partially observed physiological systems. 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.- subawards are planned for this award.

FDT-biotech:hybrid digital twins for predicting skeletal muscle adaptation across activity and metabolic states

Award facts

Description

Quick answers

Who received award 2625877?

How much is award 2625877 worth?

Which agency awarded 2625877?

See this award's full record