Ultrasonic kinetocardiogram segmentation algorithm

Abstract

The invention relates to an ultrasonic kinetocardiogram segmentation algorithm. The ultrasonic kinetocardiogram segmentation algorithm comprises the steps of: firstly, designing a fuzzy clustering method, realizing a self-adaptive optimization process by combining with spatial information, avoiding an morphological operation in the middle, realizing automatic rough segmentation on an image, reducing influence of noise, and regarding a result thereof as a control parameter of a level set segmentation operation for controlling the evolution process of a level set; secondly, performing secondary segmentation on an ultrasonic image by adopting a level set method, further reducing the influence of noise and acquiring a relatively clear contour line; and finally, introducing priori knowledge of a ventriculus sinister shape to conduct evolution of the fitting calculation constraint segmentation shape on an ROI region since the problems of under-segmentation and over-segmentation caused by fuzzy boundary and nonuniformity of pixels are very common and difficult to avoid, so as to obtain an expected smooth closed contour. The ultrasonic kinetocardiogram segmentation algorithm significantly improves the precision and robustness of segmentation.

Description

technical field [0001] The invention relates to the field of digital image processing, in particular to an echocardiogram segmentation algorithm. Background technique [0002] Echocardiography is the medical imaging modality of choice for visualization of the left ventricle of the heart due to the low cost and portability of ultrasound imaging equipment. Typically, cardiologists analyze ultrasound images of the left ventricle and segment the endocardial border during the end-systolic and end-diastolic phases of the left ventricle, which is then used to provide a quantitative analysis of cardiac function to diagnose heart disease. In clinical practice, manual segmentation is often used, which cannot be used for automatic parameter collection. It takes a long time and the process is tedious, and the segmentation results will vary from person to person. In addition, the algorithm based on the active contour model is a method that has attracted the attention of researchers, and...

AI Technical Summary

Problems solved by technology

In clinical practice, manual segmentation is mostly used, which cannot be used for automatic parameter collection. It takes a long time and the process is tedious, and the segmentation results will vary from person to person.

Images