摘要
目的探讨人工智能辅助CT血管成像(CTA)在西藏地区颅内动脉瘤诊治中的应用价值。方法回顾性分析2021年8月至2022年4月拉萨市人民医院神经外科收治的26例颅内动脉瘤患者。所有患者均于24 h内行CTA检查,数据分别使用人工智能辅助和人工方法进行脑血管三维重建,比较2种方法的重建时间、诊断结果、图像质量。采用独立样本t检验比较人工智能重建组和人工重建组重建时间和图像评分的差异,采用χ^(2)检验比较疾病诊断准确性的差异。结果人工智能重建组动脉瘤位置诊断准确性为92.3%(24/26),人工重建组准确性为96.2%(25/26),2组差异无统计学意义(P>0.05)。人工智能重建组CTA重建所需时间显著低于人工重建组[(24.2±11.8)min vs(94.7±42.0)min],差异具有统计学意义(t=-8.82,P<0.001)。人工智能重建组图像评分高于人工重建组[(4.53±0.58)分vs(3.46±0.94)分],差异具有统计学意义(t=4.24,P<0.001)。结论人工智能辅助CTA脑血管重建成像技术较人工重建更快速,显示动脉瘤情况更满意,适合在基层医院应用。
Objective To investigate the application value of artificial intelligence(AI)-assisted CT angiography(CTA)in diagnosing and treating intracranial aneurysms in Tibet.Methods The author retrospectively reviewed 26 patients diagnosed with intracranial aneurysms admitted to the Department of Neurosurgery of Lhasa People's Hospital from October 2021 to April 2022.All patients underwent CTA examination within 24 hours after admission.The data were reconstructed using both AI-assisted and manual methods for 3D reconstruction of cerebral blood vessels,respectively.The variables of reconstruction time,diagnostic accuracy,and image quality between two groups were compared.Independent samples t-test was used for continuous variables,and chi-square test for categorical samples.Results The diagnostic accuracy of the aneurysm was 92.3%(24/26)in the AI group and 96.2%(25/26)in the manual group,reaching no significant difference between two groups(P>0.05).The time interval required for CTA reconstruction was 24.2±11.8 minutes in the AI-assisted group,which was significantly lower than 94.7±42.0 minutes in the manual group(t=-8.82,P<0.001).The reconstruction quality score of the AI group was 4.53±0.58,and that of the manual group was 3.46±0.94.The AI group was significantly better than the artificial group(t=4.24,P<0.001).Conclusion AI-assisted CTA cerebral vascular reconstruction imaging technology is faster than manual reconstruction,and shows more satisfactory aneurysm condition,which is suitable for promotion in primary hospitals.
作者
付永鹏
拉巴索朗
马强
陈群超
郑裕峰
吴蕻
郑圆杰
胡婧
于洮
张东
Yongpeng Fu;La ba suo lang;Qiang Ma;Qunchao Chen;Yufeng Zheng;Hong Wu;Yuanjie Zheng;Jing Hu;Tao Yu;Dong Zhang(Department of Neurosurgery,People's Hospital of Lhasa,Lhasa 850000,China;Beijing Telezx Technology Co.,Ltd,Beijing 100085,China;Shukun(Beijing)Technology Co.,Ltd,Beijing 102299,China;Beijing Tiantan Hospital affiliated to Capital Medical University,Beijing Institute of Neurosurgery,100050 Beijing,China)
出处
《中华脑血管病杂志(电子版)》
2023年第1期26-30,共5页
Chinese Journal of Cerebrovascular Diseases(Electronic Edition)
基金
国家重点研发计划(2021YFC2500500)
首都卫生发展科研专项(2022-2-1075)
西藏自治区自然科学基金组团式医学援藏项目(XZ2022ZR-ZY19(Z))。
关键词
颅内动脉瘤
蛛网膜下腔出血
人工智能
CT血管成像
Intracranial aneurysm
Subarachnoid hemorrhage
Artificial intelligence
CT vascular imaging