CT-scan vs. 3D surface scanning of a skull: first considerations regarding reproducibility issues

Three-dimensional surface scanning (3DSS) and multi-detector computed tomography (MDCT) are two techniques that are used in legal medicine for digitalizing objects, a body or body parts such as bones. While these techniques are more and more commonly employed, surprisingly little information is known about the quality rendering of digitalized three-dimensional (3D) models provided by each of them. This paper presents findings related to the measurement precision of 3D models obtained through observation of a study case, where a fractured skull reconstructed by an anthropologist was digitalized using both post-mortem imaging methods. Computed tomography (CT) scans were performed using an 8-row MDCT unit with two different slice thicknesses. The variability of 3D CT models superimposition allowed to assess the reproducibility and robustness of this digitalization technique. Furthermore, two 3D surface scans were done using a professional high resolution 3D digitizer. The comparison of 3D CT-scans with 3D surface scans by superimposition demonstrated several regions with significant differences in topology (average difference between +1.45 and -1.22 mm). When comparing the reproducibility between these two digitalizing techniques, it appeared that MDCT 3D models led in general to greater variability for measurement precision between scanned surfaces. Also, the reproducibility was better achieved with the 3D surface digitizer, showing 3D models with fewer and less pronounced differences (from +0.32 to -0.31 mm). These experiments suggest that MDCT provides less reproducible body models than 3D surface scanning. But further studies must be undertaken in order to corroborate this first impression, and possibly explain the reason for these findings.