赣南医科大学学报2024，Vol.44Issue(10)：1030-1033,4.DOI:10.3969/j.issn.1001-5779.2024.10.010
人工智能在县级基层医院恶性肺结节鉴别诊断中的辅助价值研究
Research on the auxiliary value of artificial intelligence in the differential diagnosis of malignant pulmonary nodules in county-level grassroots hospitals
钟祖春 1蔡龙 1李第芬1
作者信息
- 1. 赣县区人民医院CT-MR室,江西 赣州 341000
- 折叠
摘要
Abstract
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.关键词
人工智能/恶性肺结节/阅片耗时/鉴别诊断Key words
Artificial intelligence/Malignant pulmonary nodules/Reading time/Differential diagnosis分类
医药卫生引用本文复制引用
钟祖春,蔡龙,李第芬..人工智能在县级基层医院恶性肺结节鉴别诊断中的辅助价值研究[J].赣南医科大学学报,2024,44(10):1030-1033,4.
