Self-Supervised Temporal Regularization for Landmark-Based Cardiac Segmentation with Automatic AHA Regional Mapping

2026-06-30

AI Summary

This work addresses the temporal inconsistency in landmark-based cardiac segmentation models caused by frame-wise independent training in cine images. To resolve this, the authors propose a self-supervised temporal regularization method that leverages temporal coherence across sequences in a post-training phase. By enforcing continuity in velocity and acceleration between adjacent frames, the approach achieves temporally consistent cardiac segmentation and motion estimation while automatically mapping anatomical landmarks to the standard AHA 17-segment model. This is the first method to jointly integrate self-supervised temporal regularization with anatomical correspondence, enabling temporally coherent segmentation and standardized regional assessment without requiring per-frame annotations. Experiments on the CAMUS dataset demonstrate that the proposed method significantly improves temporal consistency in segmentation and effectively facilitates the detection of abnormal myocardial motion. The code is publicly available.

Abstract

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

Problem

• temporal discontinuities
• cardiac segmentation
• AHA regional mapping
• clinical measurements
• myocardial motion

Innovation

• self-supervised temporal regularization
• cardiac segmentation
• anatomical correspondences
• AHA regional mapping
• temporal consistency