基于弥散张量配准和皮层分割的白质纤维束追踪技术

DTI Tractography Based on Tensor Registration and Cortical Segmentation

目的实现弥散张量图像与标准模板的非线性对齐,并以自动分割的皮层和皮层下区域为感兴趣区域进行纤维追踪。资料与方法使用1.5T磁共振扫描仪,对1例脑梗死恢复期患者进行DTI、DWI和T1WI扫描,对DTI图像进行预处理,应用ART矫正DTI图像变形、使用dtitk将张量图像与标准模板进行非线性对齐,随后将对齐结果应用到T1WI中;应用FreeSurfer对标准化的T1WI进行皮层和皮层下脑区分割,在MedINRIA软件包中以分割后的脑区作为感兴趣区域,进行白质纤维追踪,经过在CINCH软件包中修剪后,在三维背景中显示白质纤维。结果在标准脑空间中可清晰显示全脑表面、运动皮层、脑干以及锥体束。结论利用张量图像配准和自动皮层分割技术,提供三维背景下较准确的白质纤维图像,可为临床诊断、判断预后提供依据。

Purpose To perform fiber tractography with normalized tensor and automated segmentation of brain ROIs.Materials and Methods The DWI data were preprocessed and the distortion was fixed using ART.Dtitk was employed in normalizing the tensor nonlinearly to the template.The normalization result was applied to T1WI which was segmented by free surfer.The labeled brain areas were used as ROI in tractography utilizing MedINRIA.After pruned in CINCH,the white matter paths were shown on 3D background.Results All the segmented cortex along with the motor area,brainstem and pyramidal tract could be shown in the normalized space.Conclusion Tractography based on tensor registration and brain automated segmentation makes a useful tool for clinical diagnosis and prognosis.