Routine computed tomography examinations inherently contain detailed quantitative information concerning skeletal muscle volume, adipose tissue distribution, and bone integrity that is strongly associated with patient clinical outcomes across oncology, cardiovascular disease, surgery, and critical care. In a narrative literature review published in the Journal of Medical Imaging and Interventional Radiology, researchers synthesize recent technological advances in artificial intelligence–enabled body composition analysis. The authors outline a comprehensive translational pathway toward establishing these automated computational evaluations as an opportunistic, value-added imaging biomarker during standard radiological care.
The investigators examined the technical foundations of artificial intelligence segmentation models, multidimensional biomarkers encompassing skeletal muscle, adipose tissue, ectopic fat depots, and bone, as well as necessary validation strategies. Furthermore, the review evaluated how these analytical tools achieve real-world workflow integration within existing radiology reporting systems and hospital electronic health records.
Recent advances in deep learning algorithms have enabled fast, reproducible, and scalable segmentation of multiple anatomical tissues directly from routine computed tomography examinations. These deep learning techniques segment skeletal muscle, visceral and subcutaneous adipose tissue, ectopic fat depots, and vertebral bone. Importantly, because these artificial intelligence models operate on routine diagnostic computed tomography scans that have already been acquired, detailed body composition assessments can be performed without requiring any additional imaging examinations or exposing patients to supplementary radiation.
The literature review emphasizes that multidimensional and longitudinal body composition phenotypes further improve risk stratification.
To translate automated body composition analysis into daily medical practice, emerging clinical workflows are establishing structured reporting frameworks, automated risk flags, and direct integration with clinical decision support systems. Through these integrated mechanisms, artificial intelligence–enabled body composition analysis demonstrates clear potential to transform standard computed tomography examinations into an active platform for comprehensive physiologic phenotyping and predictive health assessment.
Despite these clinical and technical opportunities, the review notes that meaningful obstacles remain before widespread adoption can occur. Key challenges include establishing standardization, completing external validation, harmonizing measurements across diverse computed tomography scanners and imaging protocols, and providing definitive demonstrations of clinical utility. The authors conclude that successful clinical implementation will ultimately depend upon rigorous multicenter validation studies, universally standardized reporting frameworks, practical workflow integration, and careful alignment with evolving healthcare regulatory and reimbursement structures.
Subject of Research: Artificial intelligence-enabled body composition analysis on routine computed tomography
Article Title: Body composition analysis on routine CT imaging: a predictive health opportunity
Article References: Behzad, S., Mohseni, G., Valizadeh, P., Gholamrezanezhad, A., & Akhlaghpoor, S. (2026). Body composition analysis on routine CT imaging: a predictive health opportunity. Journal of Medical Imaging and Interventional Radiology. https://doi.org/10.1007/s44326-026-00120-2
Image Credits: AI Generated
DOI: 10.1007/s44326-026-00120-2
Keywords: computed tomography, body composition analysis, artificial intelligence, deep learning, image segmentation, skeletal muscle, visceral adipose tissue, subcutaneous adipose tissue, ectopic fat, vertebral bone, opportunistic risk assessment, physiologic phenotyping, risk stratification, radiology reporting systems, electronic health records, clinical decision support, oncology, cardiovascular disease
Cite Scienmag News
Blake Davidson. (November 17, 2026). Multidimensional and Longitudinal Body Composition Phenotypes on CT Improve Risk Stratification. Scienmag. https://scienmag.com/multidimensional-and-longitudinal-body-composition-phenotypes-on-ct-improve-risk-stratification/
Blake Davidson. "Multidimensional and Longitudinal Body Composition Phenotypes on CT Improve Risk Stratification." Scienmag, 17 November 2026, https://scienmag.com/multidimensional-and-longitudinal-body-composition-phenotypes-on-ct-improve-risk-stratification/. Accessed 18 November 2026.
Blake Davidson. "Multidimensional and Longitudinal Body Composition Phenotypes on CT Improve Risk Stratification." Scienmag. November 17, 2026. https://scienmag.com/multidimensional-and-longitudinal-body-composition-phenotypes-on-ct-improve-risk-stratification/








