Project Details
Projekt
Muscle matters: Analysing muscle quantity, quality and distribution as multisystem health biomarkers for cardiometabolic, musculoskeletal and functional outcomes
Subject Area
Medical Informatics and Medical Bioinformatics
Term
since 2025
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 566800501
Skeletal muscle is a critical player in overall health, influencing metabolic regulation, cardiovascular function, and physical performance. This project aims to harness artificial intelligence (AI) to automatically and accurately evaluate muscle quantity, quality, and distribution in large sets of MRI and CT scans, ultimately identifying new biomarkers for cardiometabolic risk, chronic back pain, and functional fitness. Our existing “MRSegmentator” prototype, which already segments multiple organs and initial muscle regions, serves as the foundation. We will enhance it to provide comprehensive 3D assessments of muscle mass, fat infiltration (both macroscopic and microscopic), and spatial distribution (e.g., left-right asymmetries, involvement of key muscle groups). These refinements will be developed and validated using extensive MRI data from the UK Biobank, the German National Cohort (NAKO), and a hospital-based dataset. By integrating these imaging-derived muscle parameters with rich clinical and lifestyle data—such as cardiovascular measures, metabolic markers, chronic back pain diagnoses, and activity levels—we aim to uncover how fatty degeneration and distribution patterns of skeletal muscle influence disease risk. We also plan to create a unified “muscle health index” that combines these parameters into a single score, offering a straightforward metric for gauging individual muscle status. In the long term, the project seeks not only to provide deeper insights into skeletal muscle as a “multisystem biomarker,” but also to pave the way for more personalized prevention and treatment strategies. Through automated segmentation and large-scale data analysis, clinicians could identify high-risk patients earlier and tailor interventions more effectively. The synergy of advanced AI methods, broad population datasets, and radiological expertise will facilitate swift translation of our findings into clinical practice.
DFG Programme
Research Grants
