Somasundaram Karuppanagounder, Ezhilarasan Kamalanathan
From the Image Processing Laboratory, Department of Computer Science and Applications, Gandhigram Rural Institute, Gandhigram, Dindigul, Tamil Nadu, India.
J Comput Assist Tomogr. 2015 Jul-Aug;39(4):552-8. doi: 10.1097/RCT.0000000000000249.
To develop an automatic skull stripping method for magnetic resonance imaging (MRI) of human head scans.
The proposed method is based on gray scale transformation and morphological operations.
The proposed method has been tested with 20 volumes of normal T1-weighted images taken from Internet Brain Segmentation Repository. Experimental results show that the proposed method gives better results than the popular skull stripping methods Brain Extraction Tool and Brain Surface Extractor. The average value of Jaccard and Dice coefficients are 0.93 and 0.962 respectively.
In this article, we have proposed a novel skull stripping method using intensity transformation and morphological operations. This is a low computational complexity method but gives competitive or better results than that of the popular skull stripping methods Brain Surface Extractor and Brain Extraction Tool.
开发一种用于人类头部扫描磁共振成像（MRI）的自动颅骨剥离方法。
所提出的方法基于灰度变换和形态学运算。
所提出的方法已使用从互联网大脑分割库获取的20组正常T1加权图像进行测试。实验结果表明，所提出的方法比流行的颅骨剥离方法脑提取工具（Brain Extraction Tool）和脑表面提取器（Brain Surface Extractor）产生更好的结果。杰卡德（Jaccard）系数和骰子（Dice）系数的平均值分别为0.93和0.962。
在本文中，我们提出了一种使用强度变换和形态学运算的新型颅骨剥离方法。这是一种计算复杂度低的方法，但比流行的颅骨剥离方法脑表面提取器和脑提取工具产生具有竞争力或更好的结果。
