What is the clinical evidence from this study?

Study design: Other. Population: Diverse cardiac abnormalities (n=19). Intervention: Active Shape Model (ASM) algorithm vs. Manual expert-delineated ground truth segmentations. Primary outcome: Dice Similarity Coefficient (DSC) and Hausdorff Distance (HD).

What question did this study set out to answer?

The aim is to evaluate the efficacy of Active Shape Models for automated ventricular segmentation in patients with cardiac abnormalities.

March 22, 2026Open Access

Evaluation of ASM for Ventricular Segmentation in Patients with Diverse Cardiac Abnormalities

Key Result

The Active Shape Model framework achieved mean Dice Similarity Coefficient scores ranging from 0.50 to 0.64 for automated ventricular segmentation in patients with diverse cardiac abnormalities.

Key Points

The aim is to evaluate the efficacy of Active Shape Models for automated ventricular segmentation in patients with cardiac abnormalities.
Constructed a statistical shape model using 19 CCTA scans
Employed principal component analysis to capture morphological variability
Utilized strict point correspondence for topological consistency
Validated using leave-one-out cross-validation
Mean Dice scores ranged from 0.50 to 0.64, with some cases achieving 0.84
Quantitative performance reflects the complexity of pathological morphology
The ASM framework accommodates high morphometric variance effectively

Structured PICO

Does the Active Shape Model (ASM) framework accurately segment ventricles in CCTA scans of patients with diverse cardiac abnormalities compared to manual expert segmentation?

Population

19 Coronary Computed Tomography Angiography (CCTA) scans derived from patients with diverse cardiac abnormalities

Intervention

Active Shape Model (ASM) framework for automated ventricular segmentation

Comparator

Manual expert-delineated ground truths

Outcome

Segmentation accuracy measured by Dice Similarity Coefficient (DSC) and Hausdorff Distance (HD)surrogate

The ASM framework provides a resilient, interpretable foundation for automated ventricular segmentation in complex clinical geometries where unconstrained deep learning models may fail.

Limitations

Sparse training dataset limited to 19 scans
Manual expert deformation was required for every case to correct misalignments during preprocessing
The Dice Similarity Coefficient metric heavily penalized minor spatial boundary shifts in the thin-walled right ventricle

Abstract

The efficacy of Active Shape Models (ASM) for automated ventricular segmentation was evaluated to address the computational demands of manual segmentation and the interpretability limitations of deep learning. A statistical shape model was constructed using a limited cohort of 19 Coronary Computed Tomography Angiography (CCTA) scans derived from patients with diverse cardiac abnormalities. Principal Component Analysis (PCA) was employed to encapsulate morphological variability, and strict point correspondence was enforced to maintain topological consistency. Validation was conducted via leave-one-out cross-validation, benchmarking automated segmentations against expert-delineated ground truths using the Dice Similarity Coefficient (DSC) and Hausdorff Distance (HD). It was found that mean Dice scores ranged from 0.50 to 0.64, with individual high-fidelity cases achieving scores up to 0.84. These results indicated that while quantitative performance reflected the complexity of pathological morphology, the methodology successfully accommodated high morphometric variance. It can be concluded that the ASM framework provides a resilient, interpretable foundation for managing complex clinical geometry where unconstrained models may fail.

Bookmark

View Full Paper