Fan Yang, Yaoqin Xie, Jia Gu, Lei Wang, and Huailing Zhang
image segmentation, kidney segmentation, level set method, contrast enhancement
Kidney segmentation is a key step in developing any ultrasound-based computer-aided diagnosis (CAD) systems for percutaneous renal intervention. In this paper, we propose a novel fast framework for kidney segmentation in ultrasound (US) images combined with nonlocal total variation (NLTV) image denoising and distance regularized level set evolution (DRLSE). We firstly use NLTV image denoising to get a denoised US image with the kidney region in almost homogenous intensity gray scale. Secondly, DRLSE is applied in the segmentation of kidney to get the final results. The effectiveness of this method is demonstrated through experimental results on both synthetic and US data compared with other image segmentation method. The sensitivity (SN), specificity (SP) and positive predictive value (PPV) of our segmentation result can reach 97%, 88% and 78% respectively; DRLSE method have corresponding values of 99%, 94%, 89% respectively.
Important Links:
Go Back
