Leveraging large language models to mimic domain expert labeling in unstructured text-based electronic healthcare records in non-english languages
Source check: Source checked. Not independent scientific confirmation.
Finding / tested result: The open-access primary article fine-tunes GPT-3 on Turkish respiratory-infection examples to label poorly structured pediatric EHR complaints; four pediatric specialists provide ground truth and the fine-tuned model reaches 99.88% accuracy versus 78.54% pretrained. The result is limited to the reported study population and evaluation; the source does not establish transfer to other populations or improved clinical outcomes.
Technology: Task-fine-tuned GPT-3 for Turkish EHR complaint classification
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Task-fine-tuned GPT-3 for Turkish EHR complaint classification
- Limitation / still unknown
- The study does not establish improved patient outcomes or generalization beyond the evaluated setting.
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/s12911-025-02871-6 · https://doi.org/10.1186/s12911-025-02871-6 · https://pubmed.ncbi.nlm.nih.gov/40165165/
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: linked separately.
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.