摘要
目的探究CT血管造影(CTA)人工智能(AI)后处理技术在诊断颅内动脉瘤中的临床意义。方法回顾性分析2020年10月至2022年10月因临床疑诊颅内动脉瘤在南京市江宁医院行颅脑CTA及数字减影血管造影(DSA)检查的131例患者临床及影像资料,分别通过数坤CerebralDoc头颈CT智能辅助诊断系统(AI组)与影像人工后处理法(人工组)对2组在工作效率、图像质量以及诊断效能方面进行统计学比较。结果AI组和人工组在平均完成时间[(176.74±17.49)s vs(822.05±103.11)s]、图像质量主观评分[(3.88±0.33)vs(3.64±0.48)]方面差异有统计学意义(P<0.05);以DSA检查结果为金标准,AI组、人工组对颅内动脉瘤的检出率分别为91.43%、92.38%,差异无统计学意义(P>0.05);AI组与人工组诊断颅内动脉瘤的灵敏度、特异度、准确率、阳性预测值及阴性预测值分别为91.43%vs 92.38%、88.46%vs 80.67%、90.84%vs 90.08%、96.97%vs 95.10%、71.88%vs 72.14%,组间比较差异均无统计学意义(P>0.05)。结论采用AI技术能够获得更佳的后处理图像及显著提高影像医师的工作效率,并在颅内动脉瘤诊断方面具有精准可靠的辅助作用。
Objective To explore the clinical significance of computed tomography angiography(CTA)artificial intelligence(AI)post⁃processing technology in the diagnosis of intracranial aneurysm.Methods The clinical and imaging data of 131 patients who suspectedclinical diagnosis of intracranial aneurysm and received CTA and DSA examination in our hospital from October 2020 to October 2022 were retrospectively analyzed.The data of working efficiency,image quality and diagnostic efficacy of the two groups were⁃compared statistically by ShuKun CerebralDoc head and neck CT intelligent auxiliary diagnostic system(AI group)and artificial image post⁃processing(artificial group),respectively.Results The AIgroup was significantly better than the artificial group in terms of av⁃erage completion time[(176.74±17.49)s vs(822.05±103.11)s]and subjective score of image quality[(3.88±0.33)vs(3.64±0.48)](P<0.05).The DSA results were considered the gold standard.The detection rates of intracranial aneurysms in the AI group and artificial group were 91.43%and 92.38%.respectively(P>0.05).Thesensitivity,specificity,accuracy,positive predictive value and negative predictive value of AI group and artificial group in diagnosing intracranial aneurysm were 91.43%vs 92.38%,88.46%vs 80.67%,90.84%vs 90.08%,96.97%vs 95.10%,71.88%vs 72.14%,respectively(P>0.05).Conclusion AI technology can obtain better post⁃processing images and improve the work efficiency of radiologistssignificantly,and get a precise and reliable auxiliary role in the diagnosis of intracranial aneurysm.
作者
何柳
邢滔
许定虎
余玉盛
张宏
HE Liu;XING Tao;XU Dinghu;YU Yusheng;ZHANG Hong(Department of Radiology,Nanjing Jiangning Hospital,Nanjing 211100,Jiangsu,China)
出处
《东南国防医药》
2023年第2期156-159,共4页
Military Medical Journal of Southeast China
关键词
人工智能
颅内动脉瘤
CTA
图像后处理
artificial intelligence
intracranial aneurysms
computed tomography angiography
image post-process
