目的:对比基于MRI与CT三维骨盆重建的特点及测量研究。方法:对25例在南方医科大学南方医院志愿行盆腔CT及MRI检查的发育正常、无病理性骨盆女性,应用Mimics软件分别构建出骨盆三维模型,测量三维骨盆模型的各径线、角度值并进行分析。结果:1基于MRI与CT数据集的三维骨盆模型,解剖结构清晰、形态逼真、表面光滑。2CT、MRI三维骨盆测量的入口后矢状径、入口前后径、中骨盆横径、中骨盆后矢状径、出口后矢状径、骶骨长度,差异有统计学意义(t=3.132、-4.021、-4.048、-3.468、-2.956、5.203,P<0.01),但CT与MRI三维骨盆测量差值平均值的绝对值为4 mm;入口横径、骶耻内径、中骨盆前后径、出口横径、出口前后径、耻骨联合高度、骶骨弯度,差异无统计学意义(t=-1.126、0.947、0.796、-1.916、0.192、2.172、-2.484,P>0.05)。3MRI与CT三维骨盆测量具有较好的一致性,骨盆入口横径、入口后矢状径、入口前后径、骶耻内径、中骨盆横径、中骨盆后矢状径、中骨盆前后径、骨盆出口横径、骨盆出口后矢状径、出口前后径、耻骨联合高度、骶骨长度、骶骨弯度的95%一致性界限分别为(-5.1,4.1)mm、(-2.9,5.6)mm、(-5.4,2.3)mm、(-5.9,2.8)mm、(-6.6,4.5)mm、(-6.2,3.3)mm、(-6.1,2.5)mm、(-3.6,4.4)mm、(-4.2,4.9)mm、(-5.2,5.4)mm、(-3.1,4.8)mm、(-1.8,6.2)mm和(-6.5,3.9)mm。结论:1基于MRI与CT数据集均可构建出清晰逼真的三维骨盆。2MRI与CT三维骨盆测量值比较具有较好一致性,虽有6条径线值存在统计学差异,但平均值的绝对值小于5 mm,无临床意义。MRI三维骨盆测量可应用于临床。 OBJECTIVE: To compare the characteristics and measurement of three-dimensional pelvic reconstruction based on MRI and CT. Methods: Twenty-five patients with normal and pathological pelvis who underwent pelvic CT and MRI in Nanfang Hospital of Southern Medical University were selected. Mivics software was used to construct three-dimensional model of pelvis and to measure the diameters and angles of three-dimensional pelvic model Analyze. Results: 1 The three-dimensional pelvic model based on MRI and CT dataset has clear anatomical structure, vivid morphology and smooth surface. There were significant differences between the 2CT and MRI three-dimensional pelvis in the measurement of posterior sagittal diameter, anteroposterior diameter, mid-pelvic diameter, mid-pelvic sagittal diameter, sagittal diameter and sacral length at exit (t = 3.132, -4.021 , -4.048, -3.468, -2.956, 5.203, P <0.01). However, the absolute value of the difference between the three-dimensional CT and MRI pelvis measurements was 4 mm. The diameter of the entrance, the sacral diameter of the sacrum, the anteroposterior diameter of the pelvis, Diameter, exit anteroposterior diameter, pubic symphysis height, sacral curvature, the difference was not statistically significant (t = -1.126,0.947,0.796, -1.916,0.192,2.172,2.484, P> 0.05). 3MRI and CT three-dimensional pelvic measurement has good consistency, pelvic entry diameter, sagittal entry, entrance anteroposterior diameter, sacral shame diameter, the pelvic diameter, the pelvic sagittal diameter, the pelvis anteroposterior diameter, the pelvis The 95% consistency limits of exit transverse diameter, sagittal diameter of pelvic exit, anteroposterior diameter of exit, symphysis pubis, sacral length and sacral curvature were (-5.1, 4.1) mm, (-2.9, 5.6) mm, (- 6.2, 2.3) mm, (- 6.1, 2.5) mm, (- 3.6, 4.4) mm, 4.9, mm, (-5.2,5.4) mm, (-3.1,4.8) mm, (-1.8,6.2) mm, and (-6.5,3.9) mm. Conclusions: A clear and vivid 3D pelvis can be constructed based on both MRI and CT datasets. 2MRI and CT three-dimensional pelvic measurements have good consistency, although the six radial values ​​were statistically different, but the average absolute value of less than 5 mm, no clinical significance. MRI three-dimensional pelvic measurement can be applied to clinical.