|
|||||
|
|
| 自适应脑组织影像分割 | |
| 引用本文： | 贾迪,杨金柱,张一飞,赵大哲,于戈.自适应脑组织影像分割[J].吉林大学学报(信息科学版),2012,42(1):161-165. |
| 作者姓名： | 贾迪 杨金柱 张一飞 赵大哲 于戈 |
| 作者单位： | 1. 东北大学 医学影像计算教育部重点实验室,沈阳 110179; 2. 辽宁工程技术大学 电子与信息工程学院, 辽宁 葫芦岛 125105 |
| 基金项目： | 国家自然科学基金项目(61001047). |
| 摘 要： | 提出一种支持二维及三维MR脑影像的脑组织分割方法。首先采用改进的C-V模型去除脑脊液对灰质分割准确性的干扰。其次通过采用结合C-V模型的带标记区域增长算法,去除脑壳并提取脑室。最后结合覆盖背景的方法提取灰质及白质,从而实现了脑组织的自动分割。对该算法进行了仿真与实验验证,结果表明,该算法具备良好的准确性、通用性与实用性。 |
| 关 键 词： | 计算机应用 水平集 C-V模型 区域增长 脑组织分割 |
| 收稿时间： | 2010-09-12 |
Self-adapting segmentation for brain tissue |
|
| Institution: | 1. Key Laboratory of Medical Image Computing of Ministry of Education, Northeast University,Shenyang 110179,China; 2. School of Electronics and Information Engineering,Liaoning Technical University,Huludao 125105,China |
| Abstract: | A method for 2D and 3D brain tissue segmentation was presented. First, the noise of spinal fluid affecting the segmentation accuracy was eliminated using an improved C-V model. Then, the ventricle extraction and skull stripping were performed by C-V model and regions merging with tags. Finally, the white matter and grey matter were extracted through covered background method and the brain tissue was segmented. Simulation data was used for theoretical analysis, and the results were verified by real data. The accuracy, universality and practicality were validated by experiment results. |
| Keywords: | computer application level set C-V model regions merging brain tissue segmentation |
| 点击此处可从《吉林大学学报(信息科学版)》浏览原始摘要信息 | |
| 点击此处可从《吉林大学学报(信息科学版)》下载免费的PDF全文 | |