Beyond Segmentation: Coronary Artery Calcium Scoring with Frozen Self-Supervised CT Features
Coronary artery calcium (CAC) scoring on computed tomography (CT) acquired for other indications could provide calcium-burden information without a dedicated scoring examination. We investigate direct, segmentation-free Agatston score prediction from frozen CT patch representations using only volume-level supervision for the scoring head. A lightweight additive head maps attention-selected features to non-negative contributions, combines them within slices, and sums slice scores across each volume. In developmental evaluation, applying the nonlinear mapping before aggregation improved linearly weighted κ by 0.27–0.38 over a pooled multilayer perceptron across six encoders. Using spatially adaptive label-guided temperature (SALT) features, held-out κ was 0.930 on gated CT and 0.844 on contrast-enhanced coronary CT angiography (CCTA), with separately fitted heads. On external non-gated CT, a head fitted on 165 patients from the same cohort achieved κ=0.727 on 40 held-out patients; patient-level five-fold cross-validation over all 205 patients yielded κ=0.639, indicating preliminary local-adaptation performance. Unchanged-head external gated transfer achieved κ=0.854 (n=443). Gated intraclass correlation coefficient ICC(A,1) on log(1+score) was 0.984. Study-level CAC scores served as CCTA reference targets; non-contrast images were not model inputs. SALT uses gated calcium masks during pretraining, not downstream scoring. Its highest-attention patch overlapped calcium in 89.5% of calcium-bearing slices, but localization rankings did not directly track scoring rankings. These results support direct, spatially decomposable CAC prediction from reusable features; broader validation is needed before clinical use.
Authors
- C. Seth Parker (ORCID: https://orcid.org/0000-0002-3887-1237)
- Vincent L. Sorrell (ORCID: https://orcid.org/0000-0003-2740-2993)
- Caroline Leach
- Steve Leung (ORCID: https://orcid.org/0000-0003-2832-2258)
- Stephen Hobbs (ORCID: https://orcid.org/0000-0002-2712-5730)
- Cody Bumgardner (ORCID: https://orcid.org/0000-0001-9588-3501)
- M. Elizabeth Oates (ORCID: https://orcid.org/0000-0002-6060-5322)
- Mahmut Selman Gokmen
- Jie Zhang (ORCID: https://orcid.org/0000-0003-2006-9964)
- Moneera Haque
- W. Brent Seales
- Evan Damron
Institutions
- University of Kentucky (US)
Publication Details
- Journal
- Journal of Imaging
- Published
- 2026-11-16
- DOI
- https://doi.org/10.3390/jimaging12110579
- Primary Topic
- Cardiac Imaging and Diagnostics
- Type
- article
- Field-Weighted Citation Impact
- 0.00