AI Science Tracker
EN

Fully Automatic Coronary Calcium Score Software Empowered by Artificial Intelligence Technology: Validation Study Using Three CT Cohorts.

Biomedical Engineering Research Center, Asan Institute for Life Sciences, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea · 26 Jul 2021 · multidisciplinary

Discovery

Fully Automatic Coronary Calcium Score Software Empowered by Artificial Intelligence Technology: Validation Study Using Three CT Cohorts.

Source check: Source checked. Not independent scientific confirmation.

Finding: Across 2,985 patients in three previously published CT cohorts, the system had 93.3% per-lesion sensitivity and 0.11 false-positive lesions per patient; Agatston-score ICC was 0.99 overall and weighted kappa for risk categories was 0.94. Direct comparison against manual reference scores in three retrospective cohorts and four CT models. It is single-center; only 116 participants had Agatston scores above 400, which the authors identify as insufficient for strong validation in that subgroup.

Technology: An atlas-based deep-learning system automatically detects coronary calcium and estimates Agatston scores from CT, compared with manually segmented scores.

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

AI contribution
An atlas-based deep-learning system automatically detects coronary calcium and estimates Agatston scores from CT, compared with manually segmented scores.
Test and result
Across 2,985 patients in three previously published CT cohorts, the system had 93.3% per-lesion sensitivity and 0.11 false-positive lesions per patient; Agatston-score ICC was 0.99 overall and weighted kappa for risk categories was 0.94.
Limitation / still unknown
Direct comparison against manual reference scores in three retrospective cohorts and four CT models. It is single-center; only 116 participants had Agatston scores above 400, which the authors identify as insufficient for strong validation in that subgroup.
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/PMC8546141/ · https://www.kjronline.org/DOIx.php?id=10.3348%2Fkjr.2021.0148

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: 22 Nov 2026. Review scope: The primary full text provides a specific multi-cohort clinical validation result for an AI calcium-scoring system. Exact DOI/title and result-level scans found no active or archived catalog match. The article identifies this as a single-center retrospective validation and flags the small high-calcium subgroup..

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 3/5 · Validation 2/5 · Novelty 3/5. 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 multidisciplinary discoveries.