Magnetic resonance imaging-based quantitative assessment of fatty infiltration in skeletal muscle and myocardium: a narrative review of technical principles, differences in clinical applications, and future directions.
Authors
Affiliations (2)
Affiliations (2)
- Department of Radiology, West China Hospital, Sichuan University, Chengdu, China.
- Department of Radiology, West China Second University Hospital, Sichuan University, Chengdu, China.
Abstract
Fatty infiltration of skeletal muscle and myocardium is associated with muscle dysfunction, myocardial fibrosis, and arrhythmias. Magnetic resonance imaging (MRI)-based fat quantification techniques are being increasingly applied in the skeletal muscle and myocardium, but no review has systematically compared measurements in these areas. The purpose of this review was to compare the quantification of skeletal muscle and myocardial fat in terms of technical principles, clinical applications, and research paradigms. PubMed, Web of Science, relevant journal databases, and conference proceedings (e.g., Proceedings of the International Society for Magnetic Resonance in Medicine) were searched up to December 2025 for English-language literature with the following keywords: "MRI", "fat quantification", "Dixon", "magnetic resonance spectroscopy", "T1 mapping", "T2 mapping", "skeletal muscle", "myocardium", "epicardial adipose tissue", and "artificial intelligence". Original research, reviews, and consensus statements were included. Literature onthe Dixon method, magnetic resonance spectroscopy (MRS), mapping techniques, magnetic resonance fingerprinting (MRF), and artificial intelligence (AI) was summarized. In skeletal muscle, the Dixon method is the practical standard, demonstrating robust multicenter reproducibility and established functional correlations in muscular dystrophy, sarcopenia, and inflammatory myopathies. In the myocardium, fat quantification relies mainly on mapping techniques, especially T1 mapping and epicardial adipose tissue (EAT) quantification, as Dixon and MRS techniques are limited by cardiac motion, thin walls, and low-fat content. Myocardial mapping detects subclinical injury, and increased EAT volume is independently associated with myocardial fibrosis. These techniques aid diagnosis and risk stratification in patients with dilated cardiomyopathy, arrhythmogenic right ventricular cardiomyopathy, diabetic cardiomyopathy, and Fabry disease. AI is driving automated segmentation and prognostic prediction in both fields. Skeletal muscle fat quantification is predominantly conducted with the Dixon method and is relatively mature, whereas myocardial fat quantification relies mainly on mapping techniques and lacks standardization. The multicenter validation of approaches related to skeletal muscle and related experiences in automated segmentation and AI application can inform practices in the quantification of myocardial fat. Combined cardiac-skeletal muscle assessment and integrated multiparametric imaging are promising future directions in the broader field.