摘要
目的探讨人工智能(artificial intelligence,AI)技术联合CT血管造影(CT angiography,CTA)对冠状动脉(简称冠脉)疾病的诊断价值。方法选取冠脉病变患者70例,采用AI技术联合CTA进行检查。对比人工与AI辅助诊断时间,并比较人工、AI辅助及二者联合诊断在容积再现(volume rendering,VR)中的图像质量及造成图像缺陷的相关因素;以冠脉造影(coronary angiography,CAG)检查结果为“金标准”,分析人工、AI辅助及二者联合应用对冠脉狭窄的诊断效能,并评估不同方法对冠脉狭窄的诊断结果与CAG检查结果的一致性。结果人工后处理与出报告总耗时长于AI辅助诊断行图像后处理并生成报告总耗时[(25.37±6.74)min vs.(1.53±0.28)s],且差异有统计学意义(P<0.05)。人工、AI辅助及联合诊断在VR图像质量评级比较中,差异无统计学意义(P>0.05),且3种诊断方法在不同图像缺陷因素对比中,差异亦无统计学意义(P>0.05)。根据“金标准”,70例患者中明确冠脉狭窄38例,无狭窄32例。人工联合AI辅助诊断在冠脉狭窄中的诊断价值高于人工诊断和AI辅助诊断,且与CAG检查结果一致性最强,诊断效能差异有统计学意义(P<0.05)。结论AI技术联合CTA检查在冠脉疾病诊断中优势明显,可有效缩短冠脉病变诊断耗时,提高诊断准确率,且与CAG诊断结果一致性较强,可将其作为临床鉴别诊断冠脉疾病的有效方法,值得推广应用。
Objective To investigate the diagnostic value of artificial intelligence(AI)technology combined with CT angiography(CTA)for coronary artery disease.Methods Seventy patients with coronary lesions underwent CTA examination combined with AI technology.The time of manual and AI-assisted diagnosis were compared,along with image quality in volume rendering(VR)and factors contributing to image defects separately by manual,AI-assisted,and combined diagnostic approaches.Using coronary angiography(CAG)as the“gold standard”,the diagnostic efficiency for coronary stenosis was analyzed among the three methods,and their consistency with CAG results was evaluated.Results The total time for manual post-processing and report generation was longer than that of AI-assisted diagnosis[(25.37±6.74)min vs.(1.53±0.28)s],with statistically significant difference(P<0.05).No statistically significant differences were found in VR image quality ratings or image defect factors among the three diagnostic approaches(all P>0.05).Among the 70 patients,38 were confirmed with coronary stenosis and 32 without stenosis by the gold standard.The combined manual and AIassisted diagnosis demonstrated superior value for coronary stenosis compared to either method alone,showing the strongest consistency with CAG results(P<0.05).Conclusion The combination of AI technology and CTA examination demonstrates significant advantages in diagnosing coronary artery disease,effectively reducing diagnostic time and improving diagnostic accuracy.Its strong consistency with CAG results supports its use as an effective clinical method for differentiating coronary artery disease,which deserves widespread clinical application.
作者
刘静静
洪念全
LIU Jingjing;HONG Nianquan(Department of Medical Imaging,Puyang Huimin Hospital,Puyang Henan 457000,China)
出处
《实用心电与临床诊疗》
2025年第4期531-535,共5页
PRACTICAL ELECTROCARDIOLOGY AND CLINICAL TREATMENT
关键词
冠状动脉
人工智能技术
CT血管造影
冠心病
coronary artery
artificial intelligence technology
CT angiography
coronary heart disease
作者简介
刘静静,主要从事医学影像学相关研究,E-mail:459036796@qq.com。
