“
TL;DR
The model achieved high predictive accuracy, with nutritional and fat/muscle indices contributing to its performance.
Journal of Cachexia, Sarcopenia and Muscle · 8 authors, 5 centres
This summary was generated by AI from a single paper. It has not been reviewed by a clinician and is not clinical advice. Verify against the source before acting on it.
The model achieved high predictive accuracy, with nutritional and fat/muscle indices contributing to its performance.
Using data from 4025 patients, an ensemble multi-tree XGBoost algorithm was trained on 89 variables, including nutritional and body morphometry indices. A sensitivity analysis indicated that nutritional and fat/muscle indices contributed to prediction accuracy, with a 0.31% and 6.29% influence in the internal and external validation sets, respectively. The study suggests that incorporating pre- and postoperative nutritional and morphometric data can enhance long-term survival prediction after gastrectomy.