MR空间精度测试体模与图像校正方法

目的 观察自行设计的MR空间精度测试体模及图像校正方法在临床MRI定位精度方面的应用价值。方法 根据MR质量控制标准设计并校验体模，对试验获取的MRI进行预处理、特征提取、检测与校正，提取空间各控制点的失真参数，计算几何畸变情况并校正失真。比较美国放射学会（ACR）与自制体模在相同试验条件下几何畸变率的一致性，分析MR设备几何畸变情况及校正方法的有效性。结果 相同试验条件下，自制体模与ACR体模的几何畸变率一致性良好（ICC>0.96）。校正前试验控制点位置与已标定控制点位置差异有统计学意义（P<0.05）。将体模置于MR设备成像时，所得MR图像存在一定程度几何畸变。校正后试验控制点位置与已标定控制点位置差异无统计学意义（P>0.05）。结论 自制体模能满足临床MR质量监控需求。通过检测与校正几何失真，可在一定程度上保障MRI定位精度，为临床诊断疾病及放射治疗计划提供辅助依据。

MR spatial precision test phantom and image correction method

Objective To explore the application value of self-designed MR spatial precision test phantom and image correction method in clinical MRI localization accuracy. Methods The phantom was designed and validated according to the MR quality control standards. The pre-processing, feature extraction, detection and correction of MRI obtained by the experiment were performed. The distortion parameters of each control point in the 3D space were extracted, and the geometric distortion was calculated. Meanwhile, the distortion was corrected. The consistency of the geometric distortion rate of American College of Radiology (ACR) and self-designed phantom under the same test conditions was compared, and the geometric distortion of MR equipment and the validity of the correction method were also analyzed. Results The consistency of the geometric distortion rate of ACR and self-designed phantom under the same test conditions was good (ICC>0.96). Significant difference in spatial position between the experimental before calibration and calibrated control points was found (P<0.05). When the phantom was placed in MR device for imaging, the MR equipment exerted a certain degree of geometrical distortion. There was no significant difference of spatial position between the experimental after calibration and calibrated control points (P>0.05). Conclusion The self-designed phantom can meet the quality control requirements of clinical MR equipment. The positional accuracy of MRI can be guaranteed to a certain extent through detection and correction of geometric distortion, therefore providing subsidiary basis for clinical diagnosis and radiation treatment planning.