AI Science Tracker
EN

Deep-learning chest-radiograph pressure estimates associate with heart-failure prognosis

Nihon University School of Medicine · 29 Dec 2022 · medicine

Discovery

Deep-learning chest-radiograph pressure estimates associate with heart-failure prognosis

Source check: Source checked. Not independent scientific confirmation.

Finding: A post hoc analysis of 534 acute-heart-failure admissions applied a previously developed VGG16-based model to estimate pulmonary arterial wedge pressure from chest radiographs. Higher discharge estimates and smaller decreases during hospitalization were associated with death or heart-failure readmission after adjustment; 223 patients reached the composite endpoint. This single-centre observational analysis does not establish a causal treatment target or improved care. Imaging-view differences and exclusion of in-hospital deaths limit interpretation.

Technology: VGG16-based regression convolutional neural network

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

AI contribution
Previously trained VGG16 regression CNN with global-average-pooling output estimates pressure; model trained on 748 and tested on 188 data in earlier study. Present post hoc 534-patient registry links estimated pressure to outcomes; original prospective registry collection does not establish frozen-model prospective validation.
Test and result
Retrospective admission/discharge imaging and adjusted outcome association over median 289 days follow-up.
Limitation / still unknown
Prospective clinical utility, multicentre transportability and equivalence across radiograph views remain unestablished.
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/PMC10053257/

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: Primary full article Methods, Results and limitations plus Europe PMC bibliographic metadata; PMC full article..

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.