The improved auto-snake algorithm for segmentation of B-mode ultrasonic breast mass
Mantun Gao · Beijing shengwu yixue gongcheng · 2008
The obvious difference between the shapes of benign and malignant breast mass is a very vital judgment to the clinical doctor.However the widely used conventional segmentation methods of the ultrasonic imaging systems in clinical are the segmentation of threshold-based or the manual segmentation of doctors.And the results of these methods are not very good for most of the ultrasonic images with a low contrast.A method of image segmentation based on active profile model is adopted in this paper.It is demonstrated that the algorithm is very adaptive after some improvement of the auto-snake,thus the interactive and half-automatic segmentation of the breast mass region is better realized.