人工智能在肺结节筛查和肺癌诊断中的应用  

Applications of artificial intelligence in lung nodule detection and lung cancer diagnosis

作　　者：王新宇 赵静文 刘翔[1] 石蕴玉[1] 佘云浪 WANG Xinyu;ZHAO Jingwen;LIU Xiang;SHI Yunyu;SHE Yunlang(School of Electronic and Electrical Engineering,Shanghai University of Engineering Science,Shanghai 201620,China;Department of Thoracic Surgery,Shanghai Pulmonary Hospital,Tongji University,Shanghai 200433,China)

机构地区：[1]上海工程技术大学电子电气工程学院,上海201620 [2]同济大学附属上海市肺科医院胸外科,上海200433

出　　处：《中国医学物理学杂志》2023年第9期1182-1188,共7页Chinese Journal of Medical Physics

基　　金：上海市自然科学基金(19ZR1421500)。

摘　　要：肺癌是影响人类健康和寿命最大的恶性肿瘤之一,肺结节的筛查与肺癌的早期诊断可以帮助患者尽早开始治疗。随着癌症进入精准治疗的时代,人工智能在其中扮演重要的角色,可以实现对肺结节和肺癌的检测、分割以及性质判断,极大地提高医生诊断的效率,优化医疗资源配置,因此基于医学影像的人工智能方法已经广泛应用于肺结节和肺癌的筛查、早期诊断、分级与预后。本研究基于CT、PET、PET-CT、3D-CT、MRI、病理图像6种成像方式综述了人工智能在肺结节筛查和肺癌诊断中的应用与发展,并指出其临床应用价值以及未来可能的研究方向。Lung tumor is one of the largest malignant tumors threatening human health and life.The detection of lung nodules and the early diagnosis of lung cancer can help patients start treatment as soon as possible.As the development of precision treatment,artificial intelligence is playing an increasingly important role for it can realize the detection,segmentation and property determination of lung nodules and lung cancer,greatly improve the diagnostic efficacy,and optimize the allocation of medical resources.Therefore,artificial intelligence methods based on medical images have been widely used in the detection,early diagnosis,grading and prognosis of lung nodules and lung cancer.Herein the application and development of artificial intelligence in the detection of lung nodules and the early diagnosis of lung cancer based on CT,PET,PET-CT,3DCT,MR and pathological images are reviewed,and the potential challenges and future research are further put forward.

关 键 词：肺癌 肺结节 人工智能 深度学习 影像学 综述 

分 类 号：R318[医药卫生—生物医学工程] R563[医药卫生—基础医学]