Automatic Segmentation of Heart Cavity in Echocardiography Images: Two & Four-Chamber View Using Iterative Process Method

The heart is a vital part of the human organ which is responsible for receiving and pumping blood throughout the body. According to statistical data held by the World Health Organization noted that heart disease accounts for 31% of human deaths worldwide. Examination of heart health conditions usually uses an echocardiography device with sound waves. In echocardiography, images have speckle noise and need a repair process. Noise speckle is the fine-grained texture-like pattern seen in echocardiography, and indeed in all modalities of clinical ultrasound. This study proposed the two and four-chamber cavity segmentation method in the heart wall cavity. Segmentation process to obtain contour the of heart cavity, required several stages. Improve the image using a median filter to reduce speckle noise, high boost filters and morphology is used to clarify information on the part of the heart cavity. Segmentation cavity using adaptive thresholding and canny edge detection filter to the image. The results of edge detection need to be improved using the region area and collinear. Contour results obtained will be reviewed for the open and closed condition of heart cavity. An iterative process is performed, with increasing high boost filter parameter values that may affect the results of the closed contour. Based on these combinations, the accuracy contour area is 89,409%, and the results of calculation the contour similarity level approach the value of 0. The results of the research have conducted contour searches on the heart wall cavity, and future work focused on the tracking process and measurement of heart contour movements using optical flow.