Validation of a Body-Composition Segmentation Software (Soma) on a Diverse Cohort of Publicly Available CT Scans
Nucleo Research, Inc.
Summary
This study evaluates the standalone performance of Soma, a deep-learning software developed by Nucleo Research, Inc. for the automated segmentation of body-composition tissues (skeletal muscle, subcutaneous adipose tissue, visceral adipose tissue, and intramuscular adipose tissue) on whole-body computed tomography (CT) images. The aim is to confirm that Soma produces segmentations and tissue-area measurements that agree with a multi-rater expert reference standard, on a diverse cohort representative of demographic and clinical variation. A total of 200 CT scans are sampled by stratified design from a curated pool of 2,066 scans aggregated from six publicly available, de-identified imaging datasets (autoPET, AMOS, MSD Pancreas, CT-ORG, ENHANCE.PET, RATIC). Three board-certified radiologists independently annotate the reference standard at the L3 slice. Primary performance is assessed using the Dice similarity coefficient against the multi-rater reference, with predefined thresholds and BCa bootstrap confidence intervals, both in aggregate and within every demographic and clinical subgroup. Secondary endpoints include Bland-Altman analysis of tissue-area agreement, 95th-percentile Hausdorff distance, Pearson correlation of derived indices, and Cohen's kappa for sarcopenia classification using Skeletal Muscle Index (SMI). The study is fully retrospective on de-identified images, involves no patient contact, and has been determined exempt by Salus IRB (Salus Number 26328) under 45 CFR 46.104(d)(4).
Description
Background. Body composition derived from cross-sectional imaging is increasingly used to assess sarcopenia, cachexia, and metabolic risk across oncology, surgical, and metabolic conditions. Manual segmentation at the L3 vertebra is the established reference but is time-consuming and rater-dependent. Soma is a deep-learning software pipeline (U-Net segmentation of skeletal muscle, subcutaneous adipose tissue, visceral adipose tissue, and intramuscular adipose tissue; EfficientNet-Lite0 + BiLSTM for automated L3 slice detection) developed by Nucleo Research, Inc. to provide reproducible, automa…
Eligibility
- Age range
- 18+ years
- Sex
- All
- Healthy volunteers
- Yes
Inclusion Criteria: * Subjects above 16 years or older at the time the source imaging was acquired. * De-identified abdominal computed tomography (CT) scan available from one of the six predefined publicly available datasets (autoPET, AMOS, MSD Pancreas, CT-ORG, ENHANCE.PET, or RATIC). * Scan covers the third lumbar vertebra (L3) with a contiguous axial slice suitable for L3-level body-composition analysis. * Demographic metadata required for stratified sampling (age, sex; BMI where available; clinical context as encoded in source dataset) is present. Exclusion Criteria: * Subject under 16…
Interventions
- Diagnostic TestSoma Body-Composition Segmentation Software
Soma is a deep-learning software pipeline developed by Nucleo Research, Inc. for the automated quantitative analysis of body composition from abdominal CT. It comprises (i) a U-Net segmentation model that delineates skeletal muscle, subcutaneous adipose tissue, visceral adipose tissue, and intramuscular adipose tissue on each axial CT slice; and (ii) an EfficientNet-Lite0 + BiLSTM model for automated L3 vertebra detection from axial CT volumes. In this validation study, segmentation performance is assessed on every fifth axial slice across the full scan depth. Outputs include per-tissue segmentation masks, tissue cross-sectional areas (cm\^2), and derived indices including the Skeletal Muscle Index (SMI = muscle area / height\^2). In this study, Soma is applied as the index test in standalone mode, fully blinded to the multi-rater radiologist reference standard.
Location
- Nucleo Research, Inc.San Francisco, California