影像脑功能区自动提取的研究.pdfVIP

PET 影像脑功能区自动提取的研究 The Study of Automatic Functional Area of Brain Extraction in PET Images Jian-Ge Zhang 张剑戈，Wan-Hua Guo 郭万华，Cheng Wang 王成，Xu-Feng Jiang 江旭峰 张剑戈，王成，上海第二医科大学生物医学工程教研室，上海市 200025 Jian-Ge Zhang,Cheng-Wang, Department of Biomedical Engineering, Shanghai Second medical University, Shanghai 200025, China 郭万华， 江旭峰，瑞金医院核医学科，上海市 200025 Department of nuclear medicine, Ruijin Hospital, Shanghai 200025，China 作者简介：张剑戈，男，1972 年生，安徽省芜胡人，汉族，2001 年上海第二医 科大学硕士毕业，讲师，主要从事医学图像处理、神经影像学的研究。 Biography: Jian-Ge Zhang(1972-), male, Han nationality, born in Wuhu, Anhui province, lecture, Master of Shanghai Second Medical University. Research direction: Medical image processing, neruoimage. Telephone:+86-21776425(ext) E-mail: zhangjg@shsmu.edu.cn 【摘要】目的 建立自动提取脑部功能区的方法。方法 通过非线性的空 间变换将PET 图像经过图像配准和图像形变两个步骤，归一化到Talairach 坐 标系，根据图谱中脑功能区的坐标，在3D 图像中提取相应的功能区；为验证 功能区提取的有效性，对志愿者带眼罩和不带眼罩各 14 例，加光刺激，提取 视区并染色，计算经小脑归一化后的平均密度值。结果 两组视区平均密度值 显著性差异(P0.01)；视区位置符合解剖学定义；同时邻近功能区平均密度值 无显著性差异。结论 该自动从 PET 图像中提取功能区的基于图谱的方法是 有效的。 【主题词】视区; 配准; 空间归一化; PET 【Abstract 】 Aim To establish a method that can automatically extract the brain functional area. Methods 18F-FDG PET images were spatial normalized to the Talairach Atlas space through two steps: image registration and image deformation. The functional areas were extracted from 3D PET images based on the coordinates gotten from Atlas. 14 Volunteers with eye patch and 14 volunteers without eye patch were stimulated by light; Brodmann 17 area was extracted and rendered; the mean of the grey value of the whole area, which was normalized with cerebellum, was calculated. Results The grey mean between two groups had significant difference (P0.01). The position of the Brodmann 17 a