Abstract: Objective　To evaluate the differences among the 3D measurement results of the cadaver pelvis, its digital 3D reconstruction model and 3D printing models. 　Methods 　A middle-aged male pelvic specimen was selected and a total of 14 feature points were selected and fixed on the surface of the pelvic specimen according to the physiological structure characteristics of the pelvis. The 1mm CT scan was performed on this pelvic specimen with fixed feature points. The Delta Medical Studio (DMS) was used to reconstruct the scanned image and record the 3D coordinates of the feature points. The 3D reconstruction model was printed by the 3D printing equipment (Melt deposiition modeling, FDM and Stereo Lithography Appearance, SLA), and the 3D coordinates of the recorded feature points were measured by a CMM. The distances and the angles among the feature points of the cadaver specimen, the digital model and the 3D printing solid models were calculated through the recorded 3D coordinates. The errors of the three groups of data were analyzed from the perspectives of maximum error, average error and t-value verification.　Results   The mean distance and angle measurement errors of the digital 3D reconstruction pelvic model were about 0.5mm and 0.35 degree, respectively. The mean distance error of the 3D printing models was about 0.8 ~ 1.1mm, and the mean angle error was about 0.4~0.5 degree.　Conclusions　The 3D reconstruction model and the 3D printing models are reliable for the preoperative reference of the pelvis in terms of the accuracy of feature measurement. The 3D reconstruction model or the 3D printing model can be selected as the reference object for the preoperative planning of the pelvis according to the actual needs.

Key words: Pelvis, 　3D reconstruction, 　3D printing, 　Distance measurement, 　Angle measurement, 　The error analysis