摘要
目的探讨人工智能(artificial intelligence,AI)在放射科专业基地住院医师规范化培训中的应用价值。方法以冠状动脉CT血管造影(CT angiography,CTA)报告书写培训为试点,将放射科专业基地住院医师随机分为利用AI自学(观察组7人)和由传统教学法(对照组6人)培训后书写冠脉CTA报告。每位住院医师完成学习后独立书写10份冠脉CTA报告,由指导医师评分并进行组间比较。结果两组住院医师的培训结果都令人满意(P>0.05)。观察组在冠脉的解剖命名、斑块的检出及冠脉管腔狭窄程度的判定方面表现更优(P<0.01);而对照组则在判读图像质量及冠脉外病变的检出方面表现更好(P<0.01)。结论AI自学法或传统教学法各有优劣,两者的有机结合、彼此借鉴可能是实现培训效果最优化的最佳途径。
Objective To evaluate the application of artificial intelligence(AI)in standardized training of imaging residents.Methods Taking the coronary computed tomography angiography(CTA)report as the pilot,the imaging residents were randomly divided into observation group(studied on their own with AI system)and control group(taught by traditional teaching methods).After learning,ten coronary CTA reports of every trained resident were evaluated.Comparison and analysis was conducted of the scores of these ten reports between the two groups of residents.Results Learning effect was satisfactory in both two learning groups(P>0.05).The observation group performed better than the control group in anatomical identification of coronary artery,detection of plaque and discrimination of the degree of coronary lumen stenosis(P<0.01).The control group performed better in image quality control and detection of extracoronary lesions(P<0.01).Conclusion Both observation and control groups have their own advantages and disadvantages.Effective combination and learning from each other might be the best way to optimize teaching effect.
作者
冯琦
公绪华
殷焱
吴连明
周滟
Feng Qi;Gong Xuhua;Yin Yan;Wu Lianming;Zhou Yan(Department of Radiology,Renji Hospital Affiliated to Shanghai Jiao Tong University School of Medicine,Shanghai 200127,China)
出处
《中国毕业后医学教育》
2023年第5期408-410,423,共4页
Chinese Journal of Graduate Medical Education
基金
中华医学会医学教育分会和中国高等教育学会医学教育专业委员会2018年医学教育研究立项课题(2018B-N05015)
上海交通大学医学院毕业后医学教育研究课题(BYH20180102)
上海交通大学医学院住院医师/专科医师规范化培训课程建设课题(BYH20180204)
2020年上海交通大学仁济临床医学院医学教育研究课题。
关键词
放射科
住院医师规范化培训
人工智能
冠脉CTA
radiology
standardized residency training
artificial intelligence
coronary CT angiography
