[Database of Radiotherapy Plan Image for Deformable Image Registration].

Guidelines require commissioning for deformable image registration (DIR) software before clinical use. The accuracy of DIR software depends upon data used. If common datasets for the DIR commissioning are available, the DIR results using the common datasets would be useful as an accuracy benchmark. Thus, the DIR-database (DIR-DB) was developed for DIR accuracy check and was open to access, which included radiotherapy plan data. This study was approved by Institutional Review Board (IRB). The DIR-DB recorded radiotherapy plans which had been finished on June 2017 and which at least two radiotherapy plans were built for a case in a treatment course. Cone-beam computed tomography (CBCT) images for patient setup were also collected and recorded in the DIR-DB, if it is available. All recorded data were anonymized and were allowed to access by users in Japan with the IRB approval. The accuracy metrics of DIR; Hausdorff distance, mean distance to agreement, Dice similarity coefficient, Jaccard were put up on the DIR-DB web site. The number of recorded cases were 11 cases for head and neck, 16 cases for thorax, 7 cases for abdomen, 8 cases for pelvis and 6 cases for prostate treated with brachytherapy. The number of case for CBCT was 17 cases. It was meaningful for DIR accuracy check in Japan that the DIR-DB and DIR results using the data in the database were released.

[Improvement Prediction on Contour Deformation Accuracy Using Deformable Image Registration Results Compared to Rigid Image Registration Results].

PURPOSE: The aim of this study was to analyze improvement prediction on contour deformation accuracy using deformable image registration (DIR) results compared to rigid image registration (RIR) results. METHOD: Radiotherapy plans for 31 cases (seven head and neck cases, 10 chest cases, six abdomen cases and eight female pelvis cases) from the privately open database for DIR were used. These cases used at least two radiotherapy plans, and registration was performed using two plans, not only for one case but also for different cases. The DIR and RIR were performed using the DIR software MIM Maestro (MIM software Inc., Cleveland, USA). The registration results for the following organs were analyzed: eye balls, optic nerves, brain stem, spinal cord and right and left parotid glands for head and neck; right and left lungs for chest; liver and right and left kidneys for abdomen; and rectum and bladder for pelvis. Dice similarity coefficient (DSC) for the organs was calculated from the results of RIR and DIR. The improvement in the DSC was observed. RESULTS AND DISCUSSION: DIR improved the DSC values by more than 0.2 for simple shapes, well-defined boundaries and large volumes such as eye balls, brain stem, lungs and liver. The minimum DSC for these organs was approximately 0.7. The improvement in DSC for the organs eye balls, brain stem, lungs and liver had ceiling values 0.95, 0.90, 1.0 and 1.0, respectively. DSC for the spinal cord, parotid gland, bladder and kidney also improved by DIR compared to RIR; however, DIR could not improve the DSC value for rectum compared to RIR because of a large difference in the position, shape and size due to stool and gas.

Acquisition count dependence of the specific binding ratio in 123I-FP-CIT SPECT.

OBJECTIVE: In the [123I]FP-CIT single-photon emission computed tomography (SPECT) examination, the specific binding ratio (SBR), calculated from the ratio of the striatal specific to extra-striatal background non-specific binding in the brain, is now commonly used as a quantitative index of parkinsonian syndrome. The purpose of this study was to examine the influence of count reduction on the SBR and to clarify the reliability of SBR values in patients with shorter scan times. METHODS: A striatum phantom was used in a phantom study, with the radioactivity concentration adjusted so that the right striatum:left striatum:brain parenchyma ratio was 8:4:1. Changes in SBR values and image quality, expressed as the % coefficient of variation (%CV) and normalized mean squared error (NMSE), with decreasing acquisition counts were evaluated. In the clinical study, 106 patients (73.1 ± 9.6 years) with suspected parkinsonian syndrome underwent [123I]FP-CIT SPECT, and SBR values from normal 30 min acquisitions (fullSBR) and half-count acquisitions (halfSBR) were compared. SBR values were calculated using the Tossici-Bolt (SBRTB) and a fully automatic count-based (SBRcb) methods. RESULTS: In the phantom study, image quality decreased with a reduction of acquisition counts. The %CV and NMSE decreased by up to 52.5% and 81.5%, respectively. SBR values decreased slightly as acquisition counts decreased. In the clinical study, the mean values of halfSBR were lower than those of fullSBR, and they were significantly different except for SBRTB without attenuation correction. halfSBR and fullSBR values correlated well, with halfSBR values 1-8% lower than fullSBR. The accuracy of diagnosis did not decrease even after acquisition counts were reduced by half. CONCLUSION: This study demonstrated that SBR values decrease as a function of reduced acquisition counts. Since halfSBR and fullSBR showed excellent correlation, it is suggested that fullSBR can be estimated from halfSBR using a calibration formula when scan times are reduced.