Neural Network Performance Evaluation of Simulated and Genuine Head-and-Neck Computed Tomography Images to Reduce Metal Artifacts.

Background: This study evaluated the performances of neural networks in terms of denoizing metal artifacts in computed tomography (CT) images to improve diagnosis based on the CT images of patients. Methods: First, head-and-neck phantoms were simulated (with and without dental implants), and CT images of the phantoms were captured. Six types of neural networks were evaluated for their abilities to reduce the number of metal artifacts. In addition, 40 CT patients' images with head-and-neck cancer (with and without teeth artifacts) were captured, and mouth slides were segmented. Finally, simulated noisy and noise-free patient images were generated to provide more input numbers (for training and validating the generative adversarial neural network [GAN]). Results: Results showed that the proposed GAN network was successful in denoizing artifacts caused by dental implants, whereas more than 84% improvement was achieved for images with two dental implants after metal artifact reduction (MAR) in patient images. Conclusion: The quality of images was affected by the positions and numbers of dental implants. The image quality metrics of all GANs were improved following MAR comparison with other networks.

Investigating dose homogeneity in radiotherapy of oral cancers in the presence of a dental implant system: an in vitro phantom study.

BACKGROUND: Materials with high atomic numbers are part of the composition of dental implant systems. In radiotherapy of oral cavity cancers, an implant can cause dose perturbations that affect target definition, dose calculation, and dose distribution. In consequence, this may result in poor tumor control and higher complications. In this study, we evaluated dose homogeneity when a dental implant replaced a normal tooth. We also aimed to evaluate the concordance of dose calculations with dose measurements. MATERIALS AND METHODS: In this study, 2 sets of planning CT scans of a phantom with a normal tooth and the same phantom with the tooth replaced by a Z1 TBR dental implant system were used. The implant system was composed of a porcelain-fused-to-metal crown and titanium with a zirconium collar. Three radiotherapy plans were designed when the density of the implant material was corrected to match their elements, or when all were set to the density of water, or when using the default density conversion. Gafchromic EBT-3 films at the level of isocenter and crowns were used for measurements. RESULTS: At the level of crowns, upstream and downstream dose calculations were reduced when metal kernels were applied (M-plan). Moreover, relatively measured dose distribution patterns were most similar to M-plan. At this level, relative to the non-implanted phantom, mean doses values were higher with the implant (215.93 vs. 192.25), also, new high-dose areas appeared around a low-dose streak forward to the implant (119% vs. 95%). CONCLUSIONS: Implants can cause a high dose to the oral cavity in radiotherapy because of extra scattered radiation. Knowledge of the implant dimensions and defining their material enhances the accuracy of calculations.

Design of band-pass filters in fluoroscopy by experimental and simulation methods for the 40 keV energy X-ray.

This study was carried out empirical computational and to design filters that while eliminating low-energy radiations according to the conventional methods attenuate high-energy beams that do not fundamentally affect images quality improvement and the absorbed patient dose reduction as well. In this regard, the impacts of thickness and filter material were examined on the contrast, resolution, absorbed patient dose, and image quality. We found that the use of filters increases the resolution, image quality and reduces the output dose intensity greatly, and the 0.1 mm thickness tin element was selected as the most suitable element for the filter.