冠状动脉CT血管成像联合人工智能对冠心病病变情况的诊断价值分析  

Diagnostic Value of Coronary Artery CT Angiography Combined with Artificial Intelligence in Coronary Heart Disease

作　　者：李余 唐育斌 杨连付 王莉莉 Li Yu;Tang Yubin;Yang Lianfu;Wang Lili(Jiayuguan Hospital of Traditional Chinese Medicine,Jiayuguan 735100,Gansu Province,China)

机构地区：[1]嘉峪关市中医医院,甘肃嘉峪关735100

出　　处：《中国社区医师》2024年第30期83-85,共3页Chinese Community Doctors

摘　　要：目的:分析冠状动脉CT血管成像联合人工智能对冠心病病变情况的诊断价值。方法:选取2022年1—12月嘉峪关市中医医院收治的56例冠心病患者作为研究对象,均进行冠状动脉CT血管成像检查,并将图像上传至工作站,分别由两位高年资医生进行手动分析,以及人工智能分析。记录两种图像分析方法处理和诊断时间;以冠状动脉造影诊断结果为“金标准”,比较两种图像分析方法对冠心病病变情况的诊断效能。结果:人工智能分析对图像的处理和诊断时间是(1.82±0.21)min,医师人工分析是(8.65±0.62)min,人工智能分析对图像的处理和诊断时间短于医师人工分析,差异有统计学意义(t=78.080,P<0.001)。56例冠心病患者共选择168个评价节段,冠状动脉造影检查显示,168个评价节段狭窄节段共110段,病变率为65.48%(110/168)。两种图像分析方法诊断冠状动脉狭窄的灵敏度、特异度、准确度对比,无统计学差异(P>0.05)。结论:冠状动脉CT血管成像联合人工智能对冠心病病变情况的诊断价值较高,能够缩短图像处理时间,可以作为医师诊断冠状动脉狭窄的有效辅助工具。Objective:To analyze the diagnostic value of coronary CT angiography combined with artificial intelligence in coronary heart disease.Methods:Fifty-six patients with coronary heart disease treated in Jiayuguan Hospital of Traditional Chinese Medicine from January 2022 to December 2022 were selected as the study objects,and coronary artery CT angiography was performed in all of them.The images were uploaded to a workstation for manual analysis by two senior doctors and analysis by artificial intelligence.The processing and diagnosis time of two image analysis methods were recorded.Taking the diagnosis result of coronary angiography as the"gold standard",the diagnostic efficiency of two image analysis methods for coronary heart disease was compared.Results:The time of image processing and diagnosis for artificial intelligence analysis was(1.82±0.21)min,and that for physician manual analysis was(8.65±0.62)min,the processing and diagnosis time of artificial intelligence analysis was shorter than that of physician manual analysis,and the difference was statistically significant(t=78.080,P<0.001);a total of 168 evaluation segments were selected in 56 patients with coronary heart disease;coronary angiography showed that there were 110 stenosis segments in 168 evaluated segments,and the lesion rate was 65.48%(110/168);there was no significant difference in sensitivity,specificity and accuracy of the two image analysis methods in diagnosing coronary artery stenosis(P>0.05).Conclusion:Coronary CT angiography combined with artificial intelligence has a high diagnostic value for coronary heart disease,can shorten the image processing time,and can be used as an effective auxiliary tool for doctors to diagnose coronary artery stenosis.

关 键 词：冠心病 冠状动脉造影 人工智能 CT血管成像 

分 类 号：R816.2[医药卫生—放射医学]