512层螺旋CT高分辨扫描结合人工智能系统对肺内不典型结核球诊断价值分析  

Analysis of the diagnostic value of 512-slice spiral CT high-resolution scanning combined with artificial intelligence system for atypical tuberculosis in the lung

作　　者：杨宏[1] YANG Hong(People's Hospital of Ganzhou City,Ganzhou 341000,China)

机构地区：[1]赣州市人民医院,江西赣州341000

出　　处：《临床医药实践》2022年第8期592-594,共3页Proceeding of Clinical Medicine

基　　金：赣州市指导性科技计划项目(项目编号:GZ20207SF178)。

摘　　要：目的:探讨512层螺旋CT高分辨扫描结合人工智能系统对肺内不典型结核球的诊断价值。方法:选择2019年8月—2021年3月就诊的肺内不典型结核球患者120例,随机分为观察组(n=60)和对照组(n=60)。观察组采用512层螺旋CT高分辨扫描联合人工智能系统分析CT图像,对照组采用常规扫描联合影像科医师分析CT图像,比较两组诊断的准确率及诊断效能。结果:总检出率观察组为96.67%,对照组为91.67%,均低于病理结果,差异有统计学意义(P<0.05);两组扫描结果比较,差异无统计学意义(χ^(2)=0.833,P=0.975)。与对照组相比,观察组CT值改变幅度较大,差异有统计学意义(P<0.05)。结论:人工智能阅片在肺内不典型结核球诊断中准确度与灵敏度均较医师阅片高,有助于降低肺内不典型结核球漏诊率与误诊率,但两种方法诊断价值无明显差异,临床可采用人工智能识别与医师阅片联合诊断方式。Objective:To explore the diagnostic value of 512-slice spiral CT high-resolution scanning combined with artificial intelligence system in the diagnosis of atypical tuberculosis in the lung.Methods:A total of 120 patients with atypical tuberculosis in the lung who were treated in our hospital from August 2019 to March 2021 were selected as the research objects,and they were randomly divided into observation group(n=60)and control group(n=60).The observation group used 512-slice spiral CT high-resolution scanning combined with artificial intelligence system to analyze CT images,and the control group used conventional scanning combined with imaging physicians to analyze CT images to compare the diagnostic accuracy and diagnostic efficiency of the two groups.Results:The total detection rate in the observation group was 96.67%,and the detection rate in the control group was 91.67%,both lower than the pathological results(P<0.05).There was no statistically significant difference in the comparison of the two groups of scans analyzed(χ^(2)=0.833,P=0.975).Compared with the control group,the observation group had a larger change in CT value,there was significient difference(P<0.05).Conclusion:The accuracy and sensitivity of artificial intelligence reading in the diagnosis of atypical tuberculosis in the lung are higher than that of physicians,which helps to reduce the missed and misdiagnosed rate of atypical tuberculosis in the lung,but there is no significant difference in the diagnostic value of the two methods.The clinical use of artificial intelligence recognition and doctors reading pictures combined diagnosis method.

关 键 词：512层螺旋CT 人工智能 肺内不典型结核球 诊断 

分 类 号：R521[医药卫生—内科学]