BackgroundDeep learning has been successfully applied to low‐dose CT (LDCT) denoising. But the training of the model is very dependent on an appropriate loss function. Existing denoising models often use per‐pixel loss, including mean abs error (MAE) and mean square error (MSE). This ignores the difference in denoising difficulty between different regions of the CT images and leads to the loss of large texture information in the generated image.PurposeIn this paper, we propose a new hybrid loss function that adapts to the noise in different regions of CT images to balance the denoising difficulty and preserve texture details, thus acquiring CT images with high‐quality diagnostic value using LDCT images, providing strong support for condition diagnosis.MethodsWe propose a hybrid loss function consisting of weighted patch loss (WPLoss) and high‐frequency information loss (HFLoss). To enhance the model's denoising ability of the local areas which are difficult to denoise, we improve the MAE to obtain WPLoss. After the generated image and the target image are divided into several patches, the loss weight of each patch is adaptively and dynamically adjusted according to its loss ratio. In addition, considering that texture details are contained in the high‐frequency information of the image, we use HFLoss to calculate the difference between CT images in the high‐frequency information part.ResultsOur hybrid loss function improves the denoising performance of several models in the experiment, and obtains a higher peak signal‐to‐noise ratio (PSNR) and structural similarity index (SSIM). Moreover, through visual inspection of the generated results of the comparison experiment, the proposed hybrid function can effectively suppress noise and retain image details.ConclusionsWe propose a hybrid loss function for LDCT image denoising, which has good interpretation properties and can improve the denoising performance of existing models. And the validation results of multiple models using different datasets show that it has good generalization ability. By using this loss function, high‐quality CT images with low radiation are achieved, which can avoid the hazards caused by radiation and ensure the disease diagnosis for patients.
Segmented Gamma Scanning (SGS) is a kind of nondestructive testing (NDT) techniques, widely used in online detection for nuclear waste drums to record the types and contents of radionuclides in the nuclear wastes. It is convenient to classify and dispose of nuclear waste according to the test results, and it also avoids radiation damage to inspectors from the destructive analysis. A new software of the radioactive analysis and monitoring system for nuclear waste barrels with SGS was developed in this work, which is mainly composed of two parts including control upper computer software and radionuclide analysis software, in which control software contained motion control and Multi-channel Analyzer (MCA) control. The controlling of the mechanical platform realized the rotation of the waste drums and the synchronous lifting and lowering of the transmission source and the detector, so as to facilitate the layered scanning of the drums. The motion control is an indispensable part of the detection system, whose precision of the movement is directly affected the accuracy of the detection results. In the radioactive measurement, a high purity germanium (HPGe) gamma-ray spectrometer was used to obtain the gamma-ray spectrum, in which the MCA was responsible for the control of spectrometer and gamma-ray spectrum data record. Therefore, the MCA control part must adjust the high voltage of the HPGe and the parameters of measurement. The gamma-ray spectrum contained the radioactive information of the nuclear wastes in the drums. The analysis of the radioactive data is the core of the software, including spectrum data resolving and the radioactive reconstructed of the radionuclides in the drums. Finally, the information such as the type and activity of the radionuclides in the barrels was provided to the user. The software was written with the C# programming language, which realized the accurate control and operation of the mechanical device and the orderly performed of motion detection. To establish communication, the software used Ethernet’s TCP/IP as the control network, in which the manual mode and auto mode were alternatives. In conclusion, the software promotes the coordination and integration of motion control, MCA control and gamma-ray spectrum data analysis in the process of automatic detection of barreled nuclear waste with SGS.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.