Predicting Short-term Survival after Liver Transplantation using Machine Learning
Source check: Source checked. Not independent scientific confirmation.
Finding: Magnesium appeared among the leading selected features alongside INR, lymphocytes, platelets, WBC, sodium, age and BMI. The random-forest model achieved temporal-validation AUC 0.771, specificity 0.815 and sensitivity 0.50, outperforming the compared models by AUC. Derivation: 480 patients from 2004–2012 with 10-fold cross-validation. Temporal validation: 58 patients from 2013, including four deaths. This is existing-data temporal validation, not external validation; the small event count and lack of external testing limit generalizability. The article states external validation remains future work.
Technology: Random forest selected candidate predictors and produced the 30-day postoperative survival model from preoperative measurements; the source reports feature importance and compares the model with XGBoost, decision tree and logistic regression.
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Random forest selected candidate predictors and produced the 30-day postoperative survival model from preoperative measurements; the source reports feature importance and compares the model with XGBoost, decision tree and logistic regression.
- Test and result
- Magnesium appeared among the leading selected features alongside INR, lymphocytes, platelets, WBC, sodium, age and BMI. The random-forest model achieved temporal-validation AUC 0.771, specificity 0.815 and sensitivity 0.50, outperforming the compared models by AUC.
- Limitation / still unknown
- Derivation: 480 patients from 2004–2012 with 10-fold cross-validation. Temporal validation: 58 patients from 2013, including four deaths. This is existing-data temporal validation, not external validation; the small event count and lack of external testing limit generalizability. The article states external validation remains future work.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://www.nature.com/articles/s41598-020-62387-z · https://pmc.ncbi.nlm.nih.gov/articles/PMC7101323/
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 source full text Methods and Results; version/date checked against bibliographic record..
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 3/5 · 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.