AI Science Tracker
EN

Early cardiac MRI images identify pleural effusions in retrospective testing

Imperial College London; Imperial College Healthcare NHS Trust · 29 Oct 2020 · medicine

Discovery

Early cardiac MRI images identify pleural effusions in retrospective testing

Source check: Source checked. Not independent scientific confirmation.

Finding: An adapted HRNet segmented early anatomical cardiac MRI images and quantified pleural fluid. In a 200-scan test set containing 42 report-positive effusions, fluid quantification distinguished report-positive scans with an AUC of 0.906. Adaptive scanning and patient benefits were proposed rather than tested.

Technology: HRNet convolutional neural network

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

AI contribution
An adapted HRNet learned pixel labels from clinician annotations; segmented slices were reconstructed into a 3D fluid volume used to distinguish report-positive pleural effusions.
Test and result
Retrospective 575-scan study across two hospitals and two scanner manufacturers: 375 scans for training/validation and 200 for testing. Clinicians labelled 1,500 development slices. HRNet segmentation and 3D fluid quantification yielded pleural-effusion AUC 0.906 against final clinical reports; 42 of 200 test scans had reported effusions.
Limitation / still unknown
Report-based reference rather than an independently adjudicated effusion standard; no prospective workflow or patient-benefit test. Generalization beyond these centers remains uncertain. Separate chamber-size claims require manufacturer-specific regression calibration whose fitting partition is unclear, and are not represented here. Threshold-specific accuracy is omitted because optimal cut-offs were evaluated in the study dataset.
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/PMC7969571/

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 full-text cohort, HRNet training, reconstruction/calibration and statistical Methods, pleural-effusion Results and limitations. Author cardiac-MRI page checked; it provides a high-level description, not additional calibration documentation. Supplementary ROC figure not independently inspected. 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.