Self-Supervised Temporal Regularization for Landmark-Based Cardiac Segmentation with Automatic AHA Regional Mapping
For clinicians analyzing cardiac ultrasound sequences, this method improves temporal consistency of segmentations and enables standardized regional assessment without additional annotations.
This work introduces self-supervised temporal regularization to enforce temporally consistent cardiac segmentation and motion estimation in graph-based models, enabling automatic AHA 17-segment regional mapping. Validation on the CAMUS dataset demonstrates clinical utility.
Graph-based cardiac segmentation with implicit anatomical correspondences provides topological guarantees and population-level analysis capabilities, but models trained on independent frames of image sequences exhibit temporal discontinuities that affect reliable clinical measurements, particularly in cardiac ultrasound. In this work, we introduce self-supervised temporal regularization as a post-training refinement stage that exploits the temporal coherence in image sequences to enforce consistent cardiac segmentation and motion estimation over time, without requiring per-frame annotations. By penalizing velocity and acceleration discontinuities across consecutive frames, our method achieves temporally consistent segmentations while maintaining the learned anatomical correspondences. We further leverage these correspondences to automatically map landmarks to the AHA 17-segment clinical standard, enabling standardized regional assessment and detection of pathological myocardial motion patterns. Validation on CAMUS dataset demonstrates the clinical utility of combining temporal consistency with automatic regional mapping. The code is publicly available at https://github.com/david-montalvoo/MaskHybridGNet-TempReg