| Authors | Nowak, M., Nunez, L., Hill, C.E., Pagliaro, T., McGonigle, J., Niglas, M., Bell-Bradford, C., Robson, M.D., Thomaides Brears, H., Thomas, E.L. and Bell, J.D. |
|---|
| Abstract | Objectives: The escalating global incidence of obesity, cardiometabolic disease and sarcopenia necessitates reliable body composition measurement tools. MRI-based assessment is the gold standard for body composition analysis, with utility in both clinical and drug trial settings. This study aims to validate a new automated volumetric MRI method by comparing with manual segmentation, measurements from another vendor and against a new method for semi-automated single-slice area measurements. Methods: 4905 individuals from the UK Biobank with repeat whole-body Dixon MRI scans were selected. MRI data were processed automatically to derive volumetric and single-slice area measurements at L3 region for visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT) and abdominal skeletal muscle (SM). Automated volumetric measurements from another vendor were used for comparisons. A separate set of scans from 100 subjects were randomly selected and body composition volumes and areas were manually segmented by scientists with extensive anatomical expertise as the ground truth to validate the automated method. Results: The fully automated segmentations were found to have excellent agreement with manual segmentation with no substantial bias (ICC³ 0.96, CoV £3.3%). In 4905 individuals (49% male, mean age 62 years ± 8, BMI 26 kg/m2 ± 4), we confirmed that the volumetric methods of VAT, SAT, and SM measurement were very strongly correlated across vendors (VAT: ρ = 0.99; SAT: ρ = 1.0; both p<0.001). Single-slice L3 area measurements demonstrated very strong correlations with volume (for VAT ρ =0.97, for SAT ρ = 0.94, for SM ρ = 0.95, all p<0.001). Correlations between volumetric assessments and single-slice area assessments remained excellent (median rho 0.95) across sexes, ages, and cardiometabolic characteristics. Conclusion: We found robust correlations between manually segmented, automated volumetric and semi-automated single slice-body composition methods. Operational differences and granularity of measurements rather than technical performance define the applications most suitable for each method. |
|---|