Source check: Source checked. Not independent scientific confirmation.
Finding: The v1 preprint evaluates supervised models for classifying the PIMA dataset diabetes label from health indicators; a held-out 80/20 evaluation reports performance. This qualifies only as a narrow retrospective dataset classification result, not future-risk prediction, clinical screening, or patient benefit.
Technology: Methods train supervised Logistic Regression, KNN, Gaussian Naive Bayes, Random Forest and SVM models using the dataset outcome label and health features; abstract and Methods differ on algorithm count, so the Results section’s five-model table was followed.
Publication: Preprint · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Methods train supervised Logistic Regression, KNN, Gaussian Naive Bayes, Random Forest and SVM models using the dataset outcome label and health features; abstract and Methods differ on algorithm count, so the Results section’s five-model table was followed.
- Test and result
- Held-out table reports Random Forest accuracy 81.0% and ROC-AUC 85.4%; the text elsewhere selects Logistic Regression/Naive Bayes, so no single best-model claim is made.
- Limitation / still unknown
- PIMA dataset has 768 records, 80/20 stratified split and five-fold training cross-validation. The source notes the narrow demographic dataset, no real-world patient testing and no clinical validation; outcome is existing diabetes status, not prospective risk.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://www.preprints.org/manuscript/202505.2269
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.
Source check: Source checked. Not independent scientific confirmation.
Finding: Machine-learning models using clinical and laboratory features predicted SARS-CoV-2 severity and progression to critical illness; the preprint reports cohort-specific performance and predictors.
Technology: Machine-learning models use clinical and laboratory measurements to predict COVID-19 severity and progression to critical illness.
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Machine-learning models use clinical and laboratory measurements to predict COVID-19 severity and progression to critical illness.
- Test and result
- The preprint reports cohort-specific prediction of severity and critical illness from clinical and laboratory features.
- Limitation / still unknown
- The evidence is limited to the reported cohorts and predictor evaluation; prospective deployment benefit is not established.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://doi.org/10.21203/rs.3.rs-1074101/v1 · https://europepmc.org/article/PPR/PPR426469 · https://www.researchsquare.com/article/rs-1074101/v1.pdf
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.
Source check: Source checked. Not independent scientific confirmation.
Finding / tested result: Machine learning analyzes measured exhaled nitric oxide to screen for COVID-19, a specific rapid respiratory-diagnostic application.
Technology: Breath and control data from two Chinese provinces are used to evaluate classification against infection status and comparator measures.
Publication: Preprint · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Breath and control data from two Chinese provinces are used to evaluate classification against infection status and comparator measures.
- Limitation / still unknown
- Preliminary screening results require independent prospective confirmation and should not be read as definitive diagnosis.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://www.researchsquare.com/article/rs-56416/v1.pdf?c=1631870927000
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.