⚠ Evidence note: An aging clock is a statistical model, not a measurement of aging itself; predictive validity varies widely between clocks and populations.
CT L3 Radiomics and Deep Learning for Sarcopenia
Current understanding
To be populated as evidence accumulates.
Evidence log
- 2026-09-28 — Opportunistic CT-Based Quantification of Myosteatosis: A Potent Universal Imaging Biomarker for Frailty and Mortality.: cross-connection with ct-myosteatosis: L3 radiomics/deep-learning sarcopenia models and CT myosteatosis quantification share the opportunistic-CT substrate; myosteatosis adds muscle-quality (fat infiltration) signal beyond cross-sectional muscle area. (novelty: 3)
- 2026-10-01 — Integrated Assessment of Sarcopenia in Patients with Gastric Cancer Using Deep Learning and Radiomics.: Multi-center retrospective study (n=1067) of a transformer multimodal model fusing clinical variables, L3 CT radiomics, and 2.5D ResNet50 deep-learning features reported external-test AUCs of 0.89 for sarcopenia and 0.84 for severe sarcopenia in gastric cancer patients, outperforming single-modality radiomics and 2D/2.5D DL baselines; retrospective design, no prospective or outcome validation. [human-observational] (novelty: 3)
Open questions
Related
Contributing findings
Integrated Assessment of Sarcopenia in Patients with Gastric Cancer Using Deep Learning and Radiomics.
01-oct-2026
novelty 3
per-area 3
strengthens