The influence of CT scan resolution on the size of intra-observational error in digitizing long bones cross-sections

The primary aim of this study is to analyse cross-sections of long bones using structural biomechanics methods and to determine the size of the intra-observational error of the two digitisations, which can be influenced by various factors such as the size of the resolution or the presence of soft tissue. The study compares two datasets: an archaeological sample from the Pohansko region, consisting of skeletal remains in two different resolutions (0.195 a 0.977 mm/pixel) , and a sample of modern endurance runners and sprinters from Spain in resolution 0.977 mm/pixel. The cross-sections were obtained using non-invasive CT scanning. For the Spanish athlete sample, individual long bones of each subject were segmented from full-body CT scans using the 3D Slicer software. A cross-section was then created at 50% of the biomechanical length of the bone, which was calculated and properly oriented. For both datasets, the images were digitized twice by estimating the periosteal and endosteal contours. From the values of the cross-sectional parameters obtained during this repeated digitization intra-observational errors were calculated, which show the error rate of digitization made by a single evaluator. Since the CT scans of the archaeological dataset were acquired using μCT technology and thus have...