Evaluation of Anatomical Shape Priors in Deep Learning-Based Cardiac Multi-Compartment Segmentation
For researchers in medical image segmentation, this paper provides a negative result showing that simple handcrafted shape priors do not improve upon standard CNNs, indicating the need for more advanced methods.
The paper evaluates lightweight explicit shape priors for cardiac CT segmentation and finds that a standard 3D U-Net remains a strong baseline, with priors yielding marginal and inconsistent improvements. The results suggest that future gains require more expressive learned priors.
Whole-heart multi-compartment CT segmentation is clinically important, but standard CNNs do not explicitly enforce anatomical plausibility. Based on statistics derived from the training data, we evaluate whether lightweight explicit shape priors, implemented as shape-aware losses and spatial label distribution heatmap-guided U-Net variants, improve 3D cardiac segmentation on MM-WHS CT and WHS++. Across all experiments, a standard 3D U-Net surprisingly remained a very strong baseline, with handcrafted priors yielding at best marginal and inconsistent changes and often degrading performance. These results suggest that the baseline already captures substantial implicit anatomical regularities and that future gains will likely require more expressive learned priors rather than simple handcrafted anatomical shape constraints.