Cardiac Chamber Contour Extraction: Performance Evaluation of an Algorithm and Physicians

Segmentation of echocardiography images presents a great challenge since such images contain strong speckle noise and artifacts. Most ultrasound segmentation methods are semi-automatic, requiring initial contour to be manually identified in the images. In this chapter, a level set algorithm based on the phase symmetry approach and on a new logarithmic-based stopping function is used to extract simultaneously the four heart cavities in a fully automatic way. The idea is to evaluate the algorithm potential for the clinical practice as an additional tool helping the physician ́s decision. Thus, the extracted contours are compared with the ones sketched by four physicians using for that several metrics, namely distance error, maximum distance, pratt function, similarity angle, similarity region, hausdorff distance, accuracy, overlap, sensitivity, and specificity. The authors show that the proposed algorithm performs well, producing contours very similar to the physicians’ ones. The experimental work was based on echocardiography images of children. DOI: 10.4018/978-1-4666-3667-5.ch018