MRI kidney segmentation based on thresholding and active contour model.
- Author:
Xiaowei WU
1
;
Chuanqing ZHOU
;
Qiu-Shi REN
Author Information
1. Department of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China.
- Publication Type:Journal Article
- MeSH:
Algorithms;
Humans;
Image Interpretation, Computer-Assisted;
methods;
Image Processing, Computer-Assisted;
methods;
Kidney;
Magnetic Resonance Imaging;
methods
- From:
Chinese Journal of Medical Instrumentation
2011;35(3):161-163
- CountryChina
- Language:Chinese
-
Abstract:
The proposed hybrid image segmentation method, based on MRI renal images, aims at extracting the cortical and medullary regions of the kidney. It incorporates the thresholding segmentation and the active contour model. To begin with, the binary mask was obtained by preprocessing and thresholding. Then, the manually set starting contour deformed like a balloon under the framework of a simplified active contour model, thus getting the initial contour of the medulla or the cortex. Finally, the initial contour deformed according to the traditional active contour model and was attracted by the actual boundary of the tissues, defined by the gradient image.
