人工智能在提高放射科住培医师诊断肺栓塞效能中的应用

Application of artificial intelligence in improving the efficacy of radiology resident physicians in diagnosing pulmonary embolism

目的:探讨人工智能在提高放射科住培医师诊断肺栓塞效能中的应用价值。方法:选取2022年10月—2023年2月在梅州市人民医院行肺动脉CTA检查的62例患者资料。放射科住培医师采用独立阅片法与AI辅助阅片法判读CTA原始图像及后处理重建图像,两种阅片方法之间间隔两周的洗脱期。以主治医师和主任医师的CTA诊断结果为参考标准,比较住培医师独立阅片法与AI辅助阅片法诊断效能及诊断时间的差异性。结果:住培医师采用AI辅助阅片法诊断肺栓塞的灵敏度、特异度高于独立阅片法,但差异均无统计学意义(P>0.05)。住培医师采用AI辅助阅片法的诊断时间均短于独立阅片法,差异有统计学意义(P<0.05)。结论:人工智能在一定程度上有助于提高放射科住培医师诊断肺栓塞的灵敏度和特异度,并显著缩短诊断时间。

Objective To explore the application value of artificial intelligence in improving the efficacy of radiology resident physicians in diagnosing pulmonary embolism.Methods Selected data of 62 patients who underwent pulmonary artery CTA examination in Meizhou People's Hospital from October 2022 to February 2023.The radiology resident physicians used independent reading method and AI-assisted reading method to interpret CTA raw images and post-processed reconstructed images,with a two-week elution period between the two reading methods.Using CTA diagnosis results by the attending physician and chief physician as the reference standard,the differences in diagnostic efficacy and diagnostic time between the independent reading method of resident physicians and the AI assisted reading method were compared.Results The sensitivity and specificity of AI-assisted reading method used by resident physicians in diagnosing pulmonary embolism were higher than those of independent reading method,but the difference were not statistically significant(P>0.05).The diagnostic time of the AI-assisted method were shorter than that of the independent method for all resident physicians,the difference were statistically significant(P<0.05).Conclusion AI may contribute to a certain extent to improve the sensitivity and specificity of radiology resident physicians in diagnosing pulmonary embolism,and to significantly reduce the time to diagnosis.