宫颈细胞学人工智能分析系统在室间质量评价中的应用  

Application of artificial intelligence analysis system in cervical cytology in external quality assessment

作　　者：刘毓玲[1] 袁伟 狄媛媛 Liu Yuling;Yuan Wei;Di Yuanyuan(Department of Pathology,Wuhan Hospital of Traditional Chinese Medicine,Wuhan 430000,Hubei,China;Department of Pathology,Xiangyang No.1 People's Hospital,Xiangyang 441100,Hubei,China;Wuhan Zhongji Pathological Diagnosis Center,Wuhan 430000,Hubei,China)

机构地区：[1]武汉市中医医院病理科,湖北武汉430000 [2]襄阳市第一人民医院病理科,湖北襄阳441100 [3]武汉中纪病理诊断中心,湖北武汉430000

出　　处：《实用检验医师杂志》2023年第1期103-106,共4页Chinese Journal of Clinical Pathologist

摘　　要：目的 探讨宫颈细胞学人工智能(AI)分析系统应用于宫颈细胞学室间质量评价(质评)中的效果。方法 根据湖北省病理质控中心要求,送检实验室非典型鳞状上皮细胞(ASC)、低度鳞状上皮内病变(LSIL)、高度鳞状上皮内病变(HSIL)、非典型腺细胞病变(AGC)样本的细胞学玻片各1例,参加湖北省病理质控中心定期举办的室间质评活动。通过质控中心反馈意见,改进实验室细胞学检测标准操作程序(SOP)文件。在改进的SOP文件指导下,完成新一轮质量控制评价活动。结果 2020年送检4张宫颈细胞学样本,质评结果为:制片质量评价“涂片背景不够清晰,细胞染色红、橘黄及苏木素对比不分明。不通过。”其中LSIL细胞学判读应为ASC,诊断结果相差1个级别。结合质控中心专家反馈意见,实验室作出相应改进,2021年再次参加宫颈细胞学室间质评,结果技术制片和细胞学判读均合格,通过质评。结论 运用宫颈细胞学AI分析系统参加宫颈细胞学室间质评,并持续改进SOP文件,能保证宫颈细胞学筛查结果的准确性,提高诊断效率。Objective To explore the effect of the application of artificial intelligence(AI)analysis system in cervical cytology for inter room quality evaluation.Methods According to the requirements of Hubei Provincial Pathological Quality Control Center,one sample of atypical squamous cell carcinoma(ASC),one sample of low-grade squamous intraepithelial lesion(LSIL),one sample of high grade squamous intraepithelial lesion(HSIL)and one sample of atypical glandular cell lesion(AGC)were sent to the laboratory for examination.They participated in the regular inter room quality assessment activities organized by Hubei Provincial Pathological Quality Control Center.The standard operating procedure(SOP)document was improved for laboratory cytology testing through feedback from the quality control center.Under the guidance of the improved SOP document,a new round of quality control evaluation activities was completed.Results In 2020,four cervical cytology samples were submitted for examination.The quality evaluation results showed that the smear background was not clear enough,the comparison of cell staining was red,orange yellow,and hematoxylin was not clear.The LSIL cytology interpretation should be ASC,with a difference of one level in diagnostic results.Based on feedback from experts from the Quality Control Center,the laboratory has made corresponding improvements.In 2021,it participated in the cervical cytology inter room quality assessment again,and the results showed that both the technical production and cytological interpretation were qualified,passing the quality assessment.Conclusion The use of cervical cytology AI analysis system to participate in cervical cytology inter room quality assessment and continuous improvement of SOP documents could ensure the accuracy of cervical cytology screening results and improve diagnostic efficiency.

关 键 词：人工智能 宫颈细胞学 室间质评 

分 类 号：R737.33[医药卫生—肿瘤]