Body composition analysis on computed tomography scans using easy-to-access segmentation software

Computed tomography (CT) is an indirect and reference technique for body composition analysis with interesting possibilities in clinical nutrition. Segmentation is the automatic or semi-automatic software-based process by which different metabolic tissues of interest (muscle tissue, subcutaneous adipose tissue, visceral adipose tissue and intermuscular adipose tissue) that are part of the current diagnoses of malnutrition, sarcopenia and sarcopenic obesity are delimited, separated and quantified. The Alberta protocol is the most common segmentation guide, being applicable in most software. In this paper, we review the main characteristics of the most common open-source segmentation software, their degree of agreement, and some precautions and limitations of this process.