Cardiac CT and MRI radiomics: systematic review of the literature and radiomics quality score assessment

“Radiomics” — mining medical images for thousands of quantitative features, often with AI — is a hot research frontier. This systematic review delivers a sobering audit of how good that research actually is in cardiac CT and MRI.

Using a standardized Radiomics Quality Score, the authors assessed the published cardiac studies and found them wanting: a median score around 19% of the maximum, with consistent gaps in prospective design, cost-effectiveness analysis, and open-science practices like sharing data and code. The work is promising but methodologically immature.

The practical takeaway is caution: cardiac radiomics is not ready for clinical use, and clinicians should treat current claims skeptically until the methodology tightens. The resistance, if any, is from enthusiasts overselling early results.

We rate the evidence moderate: a rigorous, conflict-free systematic review with a clear scoring method, though by nature it assesses research quality rather than patient outcomes. Its clinical significance is low for now — it doesn’t change patient care directly, but it sets the standards the field must meet before radiomics can earn a clinical role. A valuable reality check.