NeuralCMF, a novel self-supervised method for 3D myocardial motion tracking, achieved a median tracking error of 2.55 mm and 90% accuracy in classifying coronary heart disease.
NeuralCMF enables continuous, self-supervised 3D myocardial motion tracking from 2D and 3D echocardiograms without needing paired training datasets, potentially improving the detection of myocardial dysfunction.
Absolute Event Rate: 2.55% vs 3.72%
Myocardial motion tracking stands as an essential clinical tool in the prevention and detection of cardiovascular diseases (CVDs), the foremost cause of death globally. However, current techniques suffer from incomplete and inaccurate motion estimation of the myocardium in both spatial and temporal dimensions, hindering the early identification of myocardial dysfunction. To address these challenges, this paper introduces the Neural Cardiac Motion Field (NeuralCMF). NeuralCMF leverages implicit neural representation (INR) to model the 3D structure and the comprehensive 6D forward/backward motion of the heart. This method surpasses pixel-wise limitations by offering the capability to continuously query the precise shape and motion of the myocardium at any specific point throughout the cardiac cycle, enhancing the detailed analysis of cardiac dynamics beyond traditional speckle tracking. Notably, NeuralCMF operates without the need for paired datasets, and its optimization is self-supervised through the physics knowledge priors in both space and time dimensions, ensuring compatibility with both 2D and 3D echocardiogram video inputs. Experimental validations across three representative datasets support the robustness and innovative nature of the NeuralCMF, marking significant advantages over existing state-of-the-art methods in cardiac imaging and motion tracking. Code is available at: https://njuvision.github.io/NeuralCMF.
Shen et al. (Thu,) conducted a other in Healthy and Coronary Heart Disease (CHD) (n=127). Neural Cardiac Motion Field (NeuralCMF) vs. VoxelMorph and Co-AttentionSTN was evaluated on Median tracking error (mm) in LV Area on STRAUS dataset. NeuralCMF, a novel self-supervised method for 3D myocardial motion tracking, achieved a median tracking error of 2.55 mm and 90% accuracy in classifying coronary heart disease.