Inflammation-type dysbiosis of the oral microbiome associates with COVID-19 symptom duration and long COVID
Source check: Source checked. Not independent scientific confirmation.
Finding: In a prospective cohort, Boruta-selected random-forest models with 5-fold cross-validation associated oral microbiome features with symptom duration among 27 followed patients. For long COVID, the best species-plus-clinical model had F1 0.615 and the pathway-only model had F1 0.689. This small observational analysis supports association, not causation or prospective clinical utility; the journal page gives first publication as 17 August 2021 and issue date as 22 October 2021.
Technology: Random-forest classification with Boruta feature selection and 5-fold cross-validation
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Random-forest classifiers use oral microbiome and clinical features to estimate symptom-duration outcomes; Boruta selects features before 5-fold cross-validation.
- Test and result
- Of 84 PCR-positive patients, 27 completed 4- and 10-week follow-up. Boruta feature selection and random-forest classification with 5-fold cross-validation gave F1 0.615 for a clinical-data-plus-microbiome-species long-COVID model and 0.689 for a pathway-only model; models without microbiome inputs showed no long-COVID signal.
- Limitation / still unknown
- The small cohort does not establish causality, external replication, or performance of a frozen model in prospective clinical use.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://insight.jci.org/articles/view/152346
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.