基于人工智能的PET图像增强方法的研究进展  

Research Progress of PET Image Enhancement Methods Based on Artificial Intelligence

作　　者：郭嘉杰 张志浩 项磊 宫恩浩 王浩[3] 唐桥虹 GUO Jia-jie;ZHANG Zhi-hao;XIANG Lei;GONG En-hao;WANG Hao;TANG Qiao-hong(Center for Certification and Evaluation,Guangdong Food and Drug Administration,Guangdong Guangzhou 510080;Subtle Medical,Shanghai 200062;Medical Device Inspection and Testing Institute,National Institutes for Food and Drug Control,Beijing 100050)

机构地区：[1]广东省药品监督管理局审评认证中心,广东广州510080 [2]深透医疗,上海200062 [3]中国食品药品检定研究院医疗器械检定所,北京100050

出　　处：《中国医疗器械信息》2023年第13期25-34,共10页China Medical Device Information

基　　金：国家重点研发计划资助课题(课题名称:人工智能医学信息系统软件审评指导体系构建,课题编号:2019YFC0118801)。

摘　　要：正电子发射体层成像利用示踪剂可视化人体中分子的代谢活动,在肿瘤的早期诊断中发挥重要作用。然而,探测器扫描时间以及示踪剂注射剂量的约束导致PET图像普遍存在低分辨率和低信噪比的问题。目前,基于人工智能的PET图像增强方法已取得优异成果。为此,重点介绍PET影像检查的应用现状与发展趋势,基于人工智能的PET图像增强方法,以及已上市的PET图像增强软件产品,旨在为研究开发人员和监管部门提供借鉴。Positron Emission Tomography(PET) uses tracers to visualize the metabolic activities of molecules in the human body and plays an important role in the early diagnosis of tumors.However,the constraints of detector scan time and tracer injection dose lead to low resolution and low signal-to-noise ratio of PET images.PET image enhancement methods based on artificial intelligence have achieved excellent results.To this end,this paper focuses on the current application status and development trend of PET imaging examination,artificial intelligence-based PET image enhancement methods,and marketed PET image enhancement software products,aiming to provide references for research developers and regulatory authorities.

关 键 词：正电子发射体层成像 人工智能 图像增强 肿瘤 

分 类 号：TH774[机械工程—仪器科学与技术]