A Novel Machine Learning Approach for Predicting Prognosis of SFTS Patients in the Early Stages of Disease
Source check: Source checked. Not independent scientific confirmation.
Finding: In a retrospective study of 233 patients with severe fever with thrombocytopenia syndrome, a logistic-regression model using five early clinical and laboratory indicators predicted survival. It achieved AUC 0.913 in the internal test set and 0.923 in a separate 73-patient external cohort. These results do not establish prospective clinical benefit.
Technology: Logistic regression selected after comparison with XGBoost, LightGBM and random forest
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Machine-learning models compared early clinical and laboratory indicators; the selected logistic-regression model estimated survival outcome from five indicators.
- Test and result
- The study compared logistic regression, XGBoost, LightGBM and random forest. The selected logistic model used age, LDH, albumin, APTT and platelet count; AUC was 0.913 in the 31-patient internal test set and 0.923 in the external cohort.
- Limitation / still unknown
- The study is retrospective; the internal test set and external cohort are small, and no prospective effect on clinical decisions or patient outcomes was measured.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://pmc.ncbi.nlm.nih.gov/articles/PMC13316448/
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 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.