A New Level Set Method for Image Segmentation and Its Application to Spatio-Temporal Image Correlation

摘要

Spatio-temporal image correlation (STIC) is a new approach for clinical assessment of the fetal heart. It allows us to acquire three-dimensional (3D) cine sequences of the fetal heart's ultrasound images, which can be played real-time and then saved as a volume data with spatial information in it. Experts can review all the images in a looped cine sequence at anytime or anywhere with the help of the Internet. In this paper, a new distance regularized level set evolution level set (DRLSE) method is applied to the segmentation of the fetal heart ultrasound images to calculate the detected fetal heartbeat rate; by taking the gray level into the level set evolution, the time to acquire accurate contours is greatly reduced. With the help of the level set method, a new algorithm is introduced to calculate heart rate. In this paper, experiments have been implemented to prove the methods.