Geometric Active Model for Lesion Segmentation on Breast Ultrasound Images
Myungeun Lee, Yanjuan Chen, Soo-Hyung Kim, Kwang Gi Kim · 2011
Ultrasound image segmentation is a difficult task due to the low contrast, blurry and signal/noise ratio. Hence, in this paper, we propose a segmentation method for ultrasound image using the level set method. In particular, the level set method utilizes a new derived speed function to improve the segmentation performance. This speed function provides a general form that incorporates an alignment term as a part of the driving force for the proper edge direction of active contour, a probability term derived from the region partition scheme, and a smoothing term for regularization. Next, we test the accuracy and robustness of the proposed method for mass segmentation of various breast ultrasound images. The experimental results show that our method is relatively more excellent on effectiveness and performance than traditional approaches. Consequently, the results show the potential of our methodology to extract breast mass which could help to reduce false positives in subsequent computer aided lesion segmentation.