AI Science Tracker
EN

Random forests link CT features to catheter-measured endpoints in CTEPH

Technical University of Munich; Saarland University Medical Center; University Hospital Schleswig-Holstein; Robert Bosch Hospital · 24 Dec 2023 · medicine

Discovery

Random forests link CT features to catheter-measured endpoints in CTEPH

Source check: Source checked. Not independent scientific confirmation.

Finding: Random forests combined CT measurements and qualitative findings to classify catheter-measured hemodynamic endpoints in patients with operable chronic thromboembolic pulmonary hypertension. Retrospective internal testing reported AUC 0.82 for a pulmonary-pressure category and 0.74 for a pulmonary-artery oxygen-saturation category. Replacement of invasive testing or improved management was not demonstrated.

Technology: Random forest

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

AI contribution
Random forests learned combinations of CT-derived measurements and qualitative imaging findings to classify catheter-referenced hemodynamic categories; feature filtering characterized the model-relevant signals.
Test and result
Single-center retrospective cohort of 127 operable CTEPH patients from 2015–2020 with CT and preoperative right-heart catheterization. Pulmonary pressure was available in 127 patients and pulmonary-artery oxygen saturation in 55. Random forests used normalized quantitative and qualitative CT features, evaluated through a source-described threefold random-sampling procedure. Reported endpoint-classification AUCs were 0.82 and 0.74.
Limitation / still unknown
Small selected surgical cohort, including only 55 patients for the oxygen endpoint; no external model validation. Exact normalization/filtering nesting and hyperparameter selection were incompletely specified. CT-feature associations are observational, and feature importance is not causation. Benefits for patient management and replacement of catheterization remain untested.
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/PMC10950991/

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: Read primary cohort/reference, CT-feature assessment, normalization/random-forest evaluation Methods, endpoint Results and limitations. Actual source describes threefold random sampling; a more detailed nested validation design is not inferred. Exact primary identity/date and fresh primary PubMed correction/retraction relationships checked..

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 in the full catalog for filters, charts and related discoveries. Browse medicine discoveries.

Related catalog entries

Links use documented relationships where available, plus shared catalog context. A shared topic or model family does not establish a scientific dependency.