Beyond Segmentation: Coronary Artery Calcium Scoring with Frozen Self-Supervised CT Features
Year: 2026
Type: article
Abstract: 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.
Source: Journal of Imaging
Language: English
Subfield: Radiology, Nuclear Medicine and Imaging
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $2,172
OpenAlex ID: W7234808148
