[New Method of Paired Comparison for Improved Observer Shortage Using Deep Learning Models].

PURPOSE: The aim of this study was to validate the potential of substituting an observer in a paired comparison with a deep-learning observer. METHODS: Phantom images were obtained using computed tomography. Imaging conditions included a standard setting of 120 kVp and 200 mA, with tube current variations ranging from 160 mA, 120 mA, 80 mA, 40 mA, and 20 mA, resulting in six different imaging conditions. Fourteen radiologic technologists with >10 years of experience conducted pairwise comparisons using Ura's method. After training, VGG16 and VGG19 models were combined to form deep learning models, which were then evaluated for accuracy, recall, precision, specificity, and F1value. The validation results were used as the standard, and the results of the average degree of preference and significance tests between images were compared to the standard if the results of deep learning were incorporated. RESULTS: The average accuracy of the deep learning model was 82%, with a maximum difference of 0.13 from the standard regarding the average degree of preference, a minimum difference of 0, and an average difference of 0.05. Significant differences were observed in the test results when replacing human observers with AI counterparts for image pairs with tube currents of 160 mA vs. 120 mA and 200 mA vs. 160 mA. CONCLUSION: In paired comparisons with a limited phantom (7-point noise evaluation), the potential use of deep learning was suggested as one of the observers.

[Effects of Subject Position on Metal Artifact Reduction of a Reverse Shoulder Prosthesis Using Computed Tomography].

PURPOSE: To validate the effects of subject position on single energy metal artifact reduction (SEMAR) of a reverse shoulder prosthesis using computed tomography (CT). METHODS: A water phantom with a reverse shoulder prosthesis was scanned at four positions on the XY plane of the CT gantry (on-center, 50 mm, 100 mm, and 150 mm from on-center in the negative direction of the X axis, respectively). We obtained images with and without SEMAR. The artifact index (AI) was measured via physical assessment. Scheffé's (Ura) paired comparison methods were performed with the amount of metal artifact by ten radiological technologists via visual assessment. RESULTS: The AI was significantly reduced when using SEMAR. As the phantom moved away from the on-center position, the AI increased, and metal artifacts increased in Scheffé's methods. CONCLUSION: SEMAR reduces metal artifacts of a reverse shoulder prosthesis, but metal artifacts may increase as the subject position moves away from the on-center position.

[Effects of Changing the Source-to-image Receptor Distance Parameter of Scattered X-ray Correction after Exposure on Image Quality in Portable Chest Radiography].

PURPOSE: To validate the effects of changing the source-to-image receptor distance (SID) parameter of scattered X-ray correction after exposure on the image quality in portable chest radiography. METHODS: The actual SID and tube current-time product (mAs) were varied such that the direct X-ray dose to a flat panel detector (FPD) remained constant. We created two groups as follows: Group A (with the SID parameter unchanged) and Group B (with the SID parameter changed to the actual SID after a phantom chest exposure). The image contrast ratio and standard deviation (SD) were measured on the chest radiographs for physical assessment. Observer studies were performed by seven radiological technologists. Scheffé's (Ura) paired comparison methods were performed with image contrast, noise, and overall assessment as the assessment items. Receiver operating characteristic (ROC) analysis for lung nodules was performed. RESULTS: The image contrast ratio and SD in Group A changed, whereas the changes in Group B were less than those in Group A for both these properties. The observer study with Scheffé's methods showed a statistically significant difference (p<0.05) for all assessment items in Group A but not in Group B. The ROC analysis did not indicate any statistically significant differences in either group. CONCLUSION: Changing the SID parameter of scattered X-ray correction after exposure can possibly maintain image contrast and noise in portable chest radiography if the actual SID changes.