EchodiST improves heart muscle motion estimates from ultrasound images
EchoDiST: Self-distillation-based joint learning for diffusion-conditioned echocardiographic myocardial motion estimation
Computer Vision and Pattern Recognition
Summary
Tracking heart muscle movement in ultrasound images is hard because of unclear images and not having exact data to train on. The authors created EchoDiST, a method that learns to estimate heart motion without needing exact movement data by teaching itself and using image patterns called diffusion. EchoDiST works well even with only a little anatomical information available, and it gives more accurate measurements for heart function compared to other methods. This helps doctors better understand how the heart is working using ultrasound.
What this means in practice
- •For cardiology imaging teams: Enhance heart motion measurements during ultrasound exams when anatomical labels are sparse to improve diagnosis of cardiac conditions.
- •For medical device developers: Integrate EchoDiST into echocardiography software for improved myocardial motion tracking without requiring extensive manual annotations.$Commercial implications: Provides a more accurate and annotation-efficient imaging tool that can be sold to hospitals improving cardiac ultrasound systems.
Authors
Feiyue Qi, Xingyue Wei, Jianwen Luo
Abstract
Motion estimation in echocardiography is essential for quantitative assessment of cardiac function and myocardial mechanics, but remains challenging due to image artifacts, limited image information, speckle decorrelation, and the scarcity of ground-truth displacement fields. Anatomy-guided approaches can provide structural information, yet often rely on expert-labeled myocardial segmentations. We propose EchoDiST, a framework for unsupervised echocardiographic myocardial motion estimation that integrates self-distillation-based joint learning with a diffusion-conditioned motion estimation network. Here, unsupervised motion estimation refers to learning without ground-truth displacement fields. The self-distillation strategy jointly optimizes anatomical segmentation and myocardial motion estimation under limited anatomical annotations. Diffusion-based conditioning is used during training with stochastic perturbations, while inference requires only a single deterministic forward pass without iterative reverse-diffusion sampling. EchoDiST was evaluated on three echocardiographic datasets, including two external test datasets under cross-view and cross-dataset settings. Compared with seven representative learning-based methods, EchoDiST consistently improved anatomical alignment, myocardial strain assessment, and motion-derived functional and cardiac-phase assessment. These gains were statistically significant across the evaluated tasks and datasets. Overall, EchoDiST provides an effective approach for reliable myocardial motion estimation under limited anatomical supervision and supports downstream quantitative assessment of cardiac function.