Purpose: This study developed a deep learning-based artificial intelligence (AI) observer to address the shortage of skilled human observers and evaluated the impact of substituting human observers with AI. Methods: We used a CT system (Aquilion Prime SP; Canon Medical Systems, Tochigi) and modules CTP682 and CTP712 to scan the phantom (Catphan 700; Toyo Medic, Tokyo). The imaging conditions were set to a tube voltage of 120 kV and tube currents of 200, 160, 120, 80, 40, and 20 mA. Each condition was scanned twice, resulting in a total of 24 images. After the paired comparison experiment with 5 observers, deep learning models based on VGG19 and VGG16 were trained. We evaluated the variance, including both human and AI observers, and examined the impact of replacing humans with AI on the average degree of preference and statistical significance. These evaluations were conducted both when the training and assessments were from the same module and when they were from different modules. Results: Variance ranged from 0.085 to 0.177 (mean: 0.124). Despite using different modules for training and evaluation, the variance remained consistent, indicating that the results are independent of the training data. The average degree of preference and image rankings were nearly identical. Between 200 mA and 160 mA, AI results differed from human results in terms of statistical significance, though the difference was minimal. The discrepancy arose from differences in observations between humans and AI, yet it fell within the expected range of variation typically observed among human observers. Conclusion: Our results suggest that replacing human observers with AI has a minimal impact and may help alleviate observer shortages. The main limitation is the inability to modify evaluation criteria or stages with the trained models.

Purpose: Fast turbo spin echo (TSE) Dixon method, which acquires both in-phase and opposed-phase in a single echo train, resulting in a 50% reduction in imaging times compared to the TSE Dixon method. However, it has been observed that the Fast TSE Dixon method exhibits degraded resolution characteristics when scanned under the same parameters as the TSE Dixon method. This study aimed to reveal suitable scan parameters for the Fast TSE Dixon method by examining its resolution characteristics when altering the parameter of the receiver bandwidth (BW) and asymmetric echoes related to sampling time. Methods: The BW was adjusted to 400, 600, 800, and 1000 Hz/pixel for the TSE Dixon and Fast TSE Dixon methods. In addition, the parameter of asymmetric echoes was changed to on (weak), and off of the Fast TSE Dixon method. The modulation transfer function (MTF) of each sequence was assessed using the edge spread function in the frequency encoding direction and phase encoding direction. Results: The findings indicated that the MTFs of the TSE Dixon method remained constant despite alterations in the sampling time. Conversely, the MTFs of the Fast TSE Dixon method showed that resolution characteristics improved by shortening sampling time when the BW was expanded, and the asymmetric echoes were changed to on. Conclusion: We revealed suitable scan parameters for the Fast TSE Dixon method have been found to improve resolution characteristics by shortening sampling time when the BW was expanded and the asymmetric echoes were changed to on.

Purpose: To evaluate the accuracy of fixation devices during irradiation of our thoracic and lumbar vertebral stereotactic body radiation therapy (SBRT) and to suggest appropriate planning target volume margins. Methods: Nine patients (45 sessions) with spinal metastases treated with spinal SBRT were studied. A fixation system with a torso shell was used as a fixture for thoracic and lumbar spinal lesions. In the patient setup, radiographic imaging was performed using Exactrac (Brainlab, Munich, Germany), followed by cone beam computed tomography (CBCT) confirmation to ensure that the images were matched within 0.3 mm just before irradiation. Irradiation was started after the collation, and additional X-ray imaging using Exactrac was performed immediately before, during, and immediately after treatment, respectively, to evaluate the fixation accuracy during irradiation. Results: The mean ±1 standard deviation for each direction for movement during irradiation with the body shell was as follows: Lateral direction: 0.01±0.21 mm, head-tail direction: −0.01±0.18 mm, anteroposterior direction: 0.06±0.16 mm, roll: 0.02±0.19°, pitch: −0.02±0.29°, yaw: −0.03±0.23°. Conclusion: The image guided radiotherapy system using X-ray imaging with Exactrac allowed the evaluation of positioning accuracy during irradiation. The body shell fixation system was able to ensure fixation accuracy within 1 mm during irradiation. In addition, the positioning during irradiation using Exactrac is very useful in the evaluation of positioning accuracy during irradiation in SBRT of the vertebral body, where very high irradiation accuracy is required.

Purpose: Beam profiles of enhanced dynamic wedge (EDW), a non-physical wedge, were obtained using EPID and 2D array detector, and the effectiveness of the QA/QC method for EDW using EPID was investigated. Methods: Using a radiotherapy unit (Clinac iX; Varian Medical Systems, Palo Alto, CA, USA), the EDW beam profile was measured 10 times with an EPID (aS1000; Varian Medical Systems) and a 2D array detector (Profiler2; Sun Nuclear, Melbourne, FL, USA) to evaluate the reproducibility of the set-up and the EDW beam profiles. The beam profiles of the physical wedge were also obtained with EPID and Profiler2 as a comparison for the EDW beam profiles. In addition, EDW irradiation logs were obtained to analyze output fluctuations during EDW irradiation. Results: Comparing the EPID and Profiler2 reproducibility of beam profiles, the EPID showed better setup position reproducibility, but the Profiler2 showed better EDW reproducibility of beam profiles. The coefficient of variation for the physical wedge reproducibility of beam profiles was equal or smaller for EPID, and for the EDW irradiation log, the variation was more significant for larger EDW angles. Conclusion: The effectiveness of the QA/QC method for EDW by EPID is high because EPID is considered to capture the EDW variation in detail and the installation accuracy is also excellent.

Purpose: Stent enhancement processing (SV) is an image processing technique that improves the visibility of coronary artery stents, and its usefulness in percutaneous coronary intervention (PCI) has been reported. However, the re-acquisition of image reconstruction errors increases patient exposure dose. This study aims to retrospectively evaluate cases of SV and identify the effect of improving image reconstruction errors using the region of interest (ROI) function of SV. Methods: The evaluation targets were 584 PCI cases where SV was used from January 2016 to December 2020. SV error was defined as the image reconstruction error in SV imaging. We evaluated the improvement of SV errors by using the ROI function of SV for images that exhibited SV errors. Significant differences in the improvement effect on SV error by using the ROI function of SV were determined using a paired t-test. A chi-squared test was performed to determine the significance of the relationship between each evaluation item and the SV error. Results: There were 53 cases of SV errors when the ROI function of SV was not used. The ROI function of SV improved SV errors in 52 cases (p<0.05). In addition, there was no significant relationship between SV errors when using the ROI function of SV and each evaluation item. Conclusion: The ROI function of SV improved the SV error in 98.11% of the cases where the SV error occurred. The ROI function of SV is useful for reducing SV errors.

Purpose: The purpose of this study was to verify the clinical applicability of PYP accumulation rates obtained with a cadmium zinc telluride (CZT) cardiac camera in the diagnosis of cardiac amyloidosis. Methods: We performed ^99mTc-PYP/²⁰¹Tl-Cl dual-isotope imaging in 84 patients with suspected cardiac amyloidosis using both a CZT cardiac camera and a conventional Anger camera. We obtained the PYP accumulation rate from CZT SPECT images and visual score and H/CL ratio from planar images. We evaluated PYP accumulation rates in 14 ATTR and 9 non-ATTR patients diagnosed by biopsy. We also evaluated the correlation between the PYP accumulation rate and both the visual assessment grade and H/CL ratio in 84 patients. Results: The PYP accumulation rate was 18.5±10.6% in the non-ATTR group and 72.4±28.7% in the ATTR group, which was significantly higher in the ATTR group (p<0.01). The correlation coefficient between the PYP accumulation rate and visual score was 0.77, and that of the H/CL ratio was 0.76, both showing strong positive correlations. Conclusion: The PYP accumulation rate obtained with a CZT cardiac camera showed significant differences between ATTR and non-ATTR groups and strong correlations with conventional assessment indices, suggesting its potential applicability in the diagnosis of cardiac amyloidosis.