https://scholars.lib.ntu.edu.tw/handle/123456789/641833
標題: | Image synthesis for low-count PET acquisitions: lower dose, shorter time | 作者: | TZE-HSIANG CHEN Zaharchuk, Greg |
關鍵字: | ALARA | Low-count PET | Low-dose PET | Short-time PET | 公開日期: | 1-一月-2022 | 來源出版物: | Biomedical Image Synthesis and Simulation: Methods and Applications | 摘要: | In positron emission tomography (PET) imaging, there is a tradeoff between radiation exposure of the subjects and the quality of the reconstructed image. However, to improve the image quality, factors such as scan time, radiotracer dose, and cost will all affect the scalability of this image modality. While it may be possible to directly interpret the short-time or low-dose PET images, methods have been proposed to enhance the quality of these images. A survey of these techniques is introduced in this chapter, from the acquisition and simulation of low-count images to their enhancement using machine learning, deep learning, and advanced reconstruction algorithms. Methods and metrics for evaluating the value of these algorithms are also introduced, as the validation with actual studies can greatly increase the utility of PET, significantly impacting the dose received, economics, and logistics of scanning. Modifications to current best practices as well as hardware and software issues must be considered as these low-count techniques are being translated into the clinic. |
URI: | https://scholars.lib.ntu.edu.tw/handle/123456789/641833 | ISBN: | 9780128243497 | DOI: | 10.1016/B978-0-12-824349-7.00025-6 |
顯示於: | 醫學工程學研究所 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。