The absolute reproducibility errors did not signi fi cantly differ in most structures between the two cases. Prior to registration, the CT and radiograph were cropped to a consistent region of support and the collimated regions of the radiograph were cropped. Only the test-session was considered, the results from the retest session were similar. This short period between the test and retest sessions was chosen to minimize biological changes that could affect the reliability of the measures and to mimic the variability expected from separate sessions, as measured in longitudinal studies. Since the ground-truth registration (including the source position) is not available, and the anatomical landmarks for target definition also involved a degree of uncertainty in the patient study, it is not straightforward to draw a definitive conclusion from those two cases. The GC was calculated on a GPU by treating one image as a column vector and using matrix multiplication in CUBLAS (nVidia, Santa Clara, CA) to simultaneously evaluate multiple moving images against a single fixed image.
The coordinate frame of the CT volume was defined at the center of the volume and its position and orientation with respect the world coordinate frame was represented as a 6-element vector of translations and rotations (x, y, z, η, θ, ϕ)obj using the ZYX Euler angle. The unmetabolized parent fraction was determined as the ratio of the radioactivity corresponding to the parent (retention time of ~ 11 min) to the total amount of radioactivity collected, fitted with an inverted gamma function. Med Phys. 2010;37:1424–39. [PubMed]Zheng G, Zhang X, Steppacher SD, Murphy SB, Siebenrock KA, Tannast M. HipMatch: an object-oriented cross-platform program for accurate determination of cup orientation using 2D–3D registration of single standard x-ray radiograph and a CT volume. The LG stream gave a signi fi cant reduction of the reproducibility error in the entorhinal cortex relative to the CS analysis (Wilcoxon test, p b 0.01). For all other evaluated cortical structures there were no signi fi cant reproducibility differences between the LG and CS analyses. Antipsychotic dosages were determined through chart review and ingested dose confirmed through interviews with the patients and monitoring of antipsychotic blood levels. The across-session test – retest spatial overlaps for both analysis streams are reported in Table 5, which shows that for each site, structure and analysis the mean Dice coef fi cient of spatial overlap averaged across subjects, across the 4 test – retest scans and across hemispheres.
Target points were manually defined in the CT volume at the approximate centroid of each vertebra. For example, one of the failure modes includes a vertical shift by exactly one level of vertebra due to the similarities between the neighboring vertebrae. The signed difference means were not signi fi cantly different from zero, indicating no biases between the across-session measures. Ann Biomed Eng. 2013;41:2077–87. [PubMed]Siddon RL. Fast calculation of the exact radiological path for a 3D CT array. However, no group difference in cortical thickness within the superior frontal gyrus was observed, and Figure 4 reveals a large degree of overlap between groups.