摘要
目的:探讨人工智能(AI)在县级基层医院恶性肺结节鉴别诊断中的辅助价值。方法:选取2022年7月—2023年6月接受计算机断层扫描(CT)且经病理诊断为恶性肺结节的100例患者,分成A组和B组各50例,A组病灶数量为59个,B组病灶数量为57个。A组先由阅片小组独立阅片(人工阅片),再由此阅片小组在AI系统辅助下阅片(AI辅助阅片)。B组先由阅片小组在AI系统辅助下阅片,再由此阅片小组独立阅片。比较2组不同方法(人工阅片、AI辅助阅片、AI独立阅片)对恶性肺结节的诊断情况以及阅片耗时的差异。结果:AI辅助阅片、AI独立阅片对恶性肺结节的诊断准确率均高于人工阅片,差异有统计学意义(P<0.05);AI辅助阅片对恶性肺结节的诊断准确率均高于AI独立阅片,差异有统计学意义(P<0.05);2组间人工阅片、AI辅助阅片、AI独立阅片对恶性肺结节的诊断准确率比较,差异无统计学意义(P>0.05);2组AI辅助阅片平均耗时、AI独立阅片平均耗时均短于人工阅片平均耗时,差异有统计学意义(P<0.05);2组AI独立阅片平均耗时均短于AI辅助阅片平均耗时,差异有统计学意义(P<0.05);2组间人工阅片平均耗时、AI辅助阅片平均耗时、AI独立阅片平均耗时比较,差异无统计学意义(P>0.05)。结论:AI系统辅助人工阅片能够提高对恶性肺结节的鉴别诊断,提高阅片效率,值得临床推广。
Objective:To explore the auxiliary value of artificial intelligence(AI)in the differential diagnosis of malignant pulmonary nodules in county-level grassroots hospitals.Methods:A study was conducted on 100 patients who underwent computed tomography(CT)scans in our hospital from July 2022 to June 2023 and were pathologically confirmed to have malignant pulmonary nodules.They were divided into Group A(59 lesions)and Group B(57 lesions),with 50 cases in each group.Group A was independently reviewed by the film review team(manual),and then the film review team will review the film with the assistance of the AI system.Group B was first reviewed by the review team with the assistance of the AI system,and then independently reviewed by this review team.The diagnostic performance of two different methods for malignant pulmonary nodules and the difference in film reading time were compared.Results:The diagnostic accuracy of AI assisted film reading and AI independent film reading for malignant pulmonary nodules was higher than that of manual film reading,with statistical significance(P<0.05).The diagnostic accuracy of AI assisted film reading for malignant pulmonary nodules was higher than that of AI independent film reading,with statistical significance(P<0.05).There was no significant difference(P>0.05)in the diagnostic accuracy of malignant pulmonary nodules between two groups of manual film reading,AI assisted film reading,and AI independent film reading.The average time spent on AI assisted film reading and AI independent film reading were both shorter than the average time spent on manual film reading,with statistical significance(P<0.05).The average time required for independent AI film reading in both groups was shorter than that for AI assisted film reading,with statistical significance(P<0.05).There was no significant difference in statistical results between the two groups in terms of average time spent on manual film reading,average time spent on AI assisted film reading,and average time spent on AI independent film reading(P>0.05).Conclusion:AI assisted manual film reading can improve the differential diagnosis of malignant pulmonary nodules and improve the efficiency of film reading,which is worthy of clinical recommendation.
作者
钟祖春
蔡龙
李第芬
ZHONG Zu-chun;CAI Long;LI Di-fen(CT-MR of People's Hospital of Ganxian District,Ganzhou,Jiangxi 341000)
出处
《赣南医科大学学报》
2024年第10期1030-1033,共4页
JOURNAL OF GANNAN MEDICAL UNIVERSITY
关键词
人工智能
恶性肺结节
阅片耗时
鉴别诊断
Artificial intelligence
Malignant pulmonary nodules
Reading time
Differential diagnosis
