AI Science Tracker
EN

101 machine learning algorithms for predicting mortality in critically ill patients with spinal cord injury and sarcopenia: a multi-database study.

Sichuan Provincial Rehabilitation Hospital; Yunnan University; Kunming Medical University; Yunnan University of Chinese Medicine · 27 Feb 2026 · medicine

Discovery

101 machine learning algorithms for predicting mortality in critically ill patients with spinal cord injury and sarcopenia: a multi-database study.

Source check: Source checked. Not independent scientific confirmation.

Finding: A random survival forest selected from 101 candidate models stratified retrospective mortality risk in spinal-cord-injury and high-sarcopenia ICU cohorts; C-index ranges were 0.88–0.91 in training and 0.62–0.70 externally.

Technology: Ten machine-learning algorithms produced 101 candidate mortality models; a random survival forest was selected by ten-fold cross-validation and presented as risk scores and nomograms.

Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.

AI contribution
Ten machine-learning algorithms produced 101 candidate mortality models; a random survival forest was selected by ten-fold cross-validation and presented as risk scores and nomograms.
Test and result
Retrospective MIMIC-IV data (7,904 patients) supported training and eICU-CRD supported external validation. Across spinal-cord-injury and high-sarcopenia cohorts, training C-indices were 0.88–0.91 and external C-indices 0.62–0.70; higher predicted risk separated survival groups.
Limitation / still unknown
External discrimination was lower than training performance. Retrospective database validation does not establish prospective utility or improved survival from use of the risk score.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://link.springer.com/article/10.1186/s12911-026-03408-1

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.

Review date: 22 Nov 2026. Review scope: Primary result-bearing material for the stated method, result and evaluation, with bibliographic version information..

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 2/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open in the full catalog for filters, charts and related discoveries. Browse medicine discoveries.