AI Science Tracker
EN

University of Dundee

1 discovery from 2022

Discoveries

Multi-omics learning distinguishes endocrine hypertension subtypes retrospectively

Source check: Source checked. Not independent scientific confirmation.

Finding: A random-forest model using 57 selected multi-omics features distinguished four hypertension subtypes in an internal held-out test, with approximately 92% balanced accuracy and AUC .95. The best single-omics comparator had approximately 81% balanced accuracy. The study supports a sample-specific diagnostic-reference validation, without establishing prospective screening benefit or biological mechanisms.

Technology: Random forest; Boruta feature selection; multi-omics classification

Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.

AI contribution
Learned forests combine selected measurements to classify established clinical subtypes; the scoped result is the observed held-out diagnostic-reference performance rather than PCA visualization or conventional correlations.
Test and result
The main-table complete-data cohort comprises 307 hypertensive participants (100 primary aldosteronism, 69 pheochromocytoma/paraganglioma, 30 Cushing syndrome and 108 primary hypertension); 101 normotensive volunteers were excluded from model fitting/testing. Data were initially split 80:20, with 100 repeated development splits and a final test set.
Limitation / still unknown
The abstract disease counts differ from the post-quality-control table; the represented cohort uses the latter rather than mixing denominators. Feature frequency cutoffs and outlier/class-balancing choices were developed empirically; their complete nesting requires supplemental detail not inspected here. This is an internal random holdout from expert referral centres, with small minority classes and no untouched external or prospective clinical evaluation. Comparing the best selected mono-omics and multi-omics models is study-specific, not a general guarantee of diagnostic improvement.
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/PMC9520210/

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: 21 Nov 2026. Review scope: Exact primary front and scoped computational/clinical model Methods and corresponding Results; no operational laboratory protocols evaluated.

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.

Open the interactive profile for the timeline, or browse all organizations and model families.