CTGAN-augmented model predicts Kawasaki disease in a second-hospital validation cohort
Source check: Source checked. Not independent scientific confirmation.
Finding: The authors developed stacked disease-classifier models, including a CTGAN-augmented version, using clinical data from febrile children. In validation at a second Taiwanese hospital, CTGAN-DC reported 95% sensitivity and 95% specificity for Kawasaki disease; the comparator model cited from prior work performed poorly on that cohort. This retrospective cross-hospital assessment is a model validation within the study, not prospective clinical testing or proof of improved care.
Technology: Stacked disease classifier with conditional tabular GAN (CTGAN) data augmentation
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- CTGAN augmentation and a stacked disease classifier were used to address class imbalance and predict Kawasaki disease among febrile children.
- Test and result
- The study evaluated the CTGAN-DC model using a second-hospital validation cohort and reported 95% sensitivity and specificity there.
- Limitation / still unknown
- The evaluation is retrospective and does not establish prospective clinical utility or improved patient outcomes; comparator differences should not be interpreted as independent replication.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://doi.org/10.1371/journal.pone.0314995
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: Full primary publisher article, including Methods, Results, author affiliations and publication history.
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.