Automated analysis of abdominal body composition using MRI: algorithm development and validation via CT comparison

The authors aimed to develop and validate a deep learning algorithm for automated segmentation of abdominal body composition from MRI data, comparing its effectiveness to CT-based analysis. Their results showed strong correlation and agreement between MRI and CT measurements for abdominal visceral fat, subcutaneous fat, and skeletal muscle, although significant absolute differences were noted, particularly for visceral fat. The study concludes that while the algorithm provides accurate MRI segmentation, clinical application requires modality-specific reference values due to systematic measurement discrepancies.