Whole myocardium tracking in 2D-echocardiography in multiple orientations using a motion constrained level-set

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Cardiovascular Imaging

Background:

• Echocardiographic image segmentation and tracking of the myocardium are crucial for diagnosing heart disease.
• Echographic images present challenges like low contrast, speckle noise, signal dropout, and shadows, complicating accurate analysis.

Purpose of the Study:

• To extend an existing level-set method for robust myocardium tracking in echocardiographic sequences.
• To improve temporal coherence and accuracy in cardiac image analysis.

Main Methods:

• An extended level-set method incorporating a novel motion prior energy term for temporal coherence.
• Spatially adaptive adjustment of hyperparameters using prior knowledge of echocardiographic regions.
• Evaluation against expert segmentations and a state-of-the-art method on a dataset of 15 sequences (approximately 900 images) across three views.

Main Results:

• The proposed method demonstrates results consistent with inter-observer variability.
• Outperforms a current state-of-the-art method in accuracy and robustness.
• Exhibits stability across various parameter settings, confirmed by a comprehensive parameter influence study.

Conclusions:

• The enhanced level-set method provides accurate and robust myocardium segmentation and tracking in echocardiographic sequences.
• The approach effectively addresses challenges in cardiac imaging, offering a valuable tool for heart disease diagnosis.