The accuracy of three-dimensional reconstructions of the ovine knee: dissectional validation

A single, complete, ovine knee was imaged using conventional MRI then dissected allowing comprehensive dimensional measurement of the intra-articular structures. A three-dimensional (3-D) computer model of the knee was generated from the MR sections. Thirty-two individual structural measures were recorded from the image output for which there was an available surgical measure for direct comparison. The results of Pearson's correlation testing show a rounded score of 1.00, suggesting an exceptional linear correlation between direct anatomical measurement and the 3-D image output. Further analysis of the data revealed an average error of measurement of 0.2mm across the 32 measures. The findings of this preliminary study suggest that 3-D reconstruction from MR data may be an appropriate, and accurate, means for making dimensional measurements of the bony and soft tissue structures of the ovine knee. It is unlikely that the measurement error would be of any great clinical significance. There is evidence in the literature to suggest that an ovine knee may be considered an acceptable model for substitution for the human knee in diagnostic assessment studies. Therefore, such findings may be considered clinically relevant in the field of human knee assessment.