扩散张量磁共振成像数据分析的TBSS方法评价

发布时间：2018-05-30 10:13

本文选题：基于体素分析 + 基于纤维束示踪的空间统计方法　；参考：《科学技术与工程》2017年23期

【摘要】：针对流行的磁共振扩散张量成像数据分析工具TBSS,提出了一种定量评估的方法。首先构建模拟数据,由31个正常人的DTI-FA图像人工模拟得到多组白质不同程度损伤程度的阿兹海默症病人组数据。将模拟得到的多组病人数据分别与正常组进行基于体素的TBSS分析。通过计算分析结果中有显著性差异的体素数量和真实变化体素数量之间的比例关系,得到该分析方法的敏感性和特异性指标。结果显示TBSS分析敏感性增大的同时伴随着特异性的减小。当模拟病人组FA值减小幅度比较大时,敏感性可达到90%以上,但特异性会不足50%,甚至更低。因此在使用TBSS进行相关疾病研究时,分析结果的判读要考虑到可能的假阳性结果,应结合其他的医学实验和研究结果进行综合分析。
[Abstract]:A quantitative evaluation method is proposed for TBS, a popular tool for data analysis of diffusive magnetic resonance (Zhang Liang) imaging. Firstly, the simulated data were constructed, and the data of Alzheimer's disease patients with different degrees of white matter damage were obtained by artificial simulation of 31 normal subjects' DTI-FA images. The simulated multigroup patient data were analyzed by voxel based TBSS with normal group. The sensitivity and specificity of the method were obtained by calculating the proportional relationship between the quantity of voxel and the quantity of real variable voxel. The results showed that the sensitivity of TBSS was increased with the decrease of specificity. When the FA value of the simulated patients decreases by a large margin, the sensitivity can reach more than 90%, but the specificity will be less than 50%, or even lower. Therefore, when using TBSS to study related diseases, the interpretation of analysis results should take into account the possible false positive results, and should be combined with other medical experiments and research results for comprehensive analysis.
【作者单位】：南方医科大学生物医学工程学院;南方医科大学南方医院;中山大学附属第三医院;
【基金】：国家自然科学基金(61372063)资助
【分类号】：R445.2

【相似文献】