基于人工智能的血细胞形态分析仪白细胞分类性能的多中心研究  

A multi-center study on evaluation of leukocyte differential performance by an artificial intelligence-based Digital Cell Morphology Analyzer

作　　者：蒋浩琴 陈葳[2] 何军[3] 江虹[4] 刘丹丹[3] 刘敏[5] 李绵洋[6] 毛志刚[4] 潘玉玲[6] 屈晨雪[7] 曲林琳[8] 孙德华[9] 孙自镛[10] 王剑飚[11] 吴文婧 王学锋[11] 续薇[8] 邢莹[7] 张弛 郑磊[9] 张式鸿[5] 关明 Jiang Haoqin;Chen Wei;He Jun;Jiang Hong;Liu Dandan;Liu Min;Li Mianyang;Mao Zhigang;Pan Yuling;Xing Ying;Qu Chenxue;Qu Linlin;Sun Dehua;Sun Ziyong;Wang Jianbiao;Wu Wenjing;Wang Xuefeng;Xu Wei;Zhang Chi;Zheng Lei;Zhang Shihong;Guan Ming(Department of Clinical Laboratory,Huashan Hospital Fudan University,Shanghai 200040,China;Department of Clinical Laboratory,the First Affiliated Hospital of Xi′an Jiaotong University,Xi′an 710061,China;Department of Clinical Laboratory,the First Affiliated Hospital of Soochow University,Suzhou 215006,China;Department of Clinical Laboratory,West China Hospital of Sichuan University,Chengdu 610044,China;Department of Clinical Laboratory,the First Affiliated Hospital,Sun Yat-sen University,Guangzhou 510062,China;Department of Clinical Laboratory,Chinese PLA Ceneral Hospital,Beijing 100080,China;Department of Clinical Laboratory,Peking University First Hospital,Beijing 100034,China;Department of Clinical Laboratory,the First Bethune Hospital of Jilin University,Jilin 130061,China;Department of Clinical Laboratory,Nanfang Hospital,Guangzhou 516006,China;Department of Clinical Laboratory,Tongji Hospital,Tongji Medical College of Hust,Wuhan 430030,China;Department of Clinical Laboratory,Ruijin Hospital,Shanghai Jiaotong University School of Medicine,Shanghai 200025,China)

机构地区：[1]复旦大学附属华山医院检验医学科,上海200040 [2]西安交通大学附属第一医院检验医学科,西安710061 [3]苏州大学附属第一医院检验医学科,苏州215006 [4]四川大学华西医院检验医学科,成都610044 [5]中山大学附属第一医院检验医学科,广州510062 [6]解放军总医院检验医学科,北京100080 [7]北京大学第一医院检验医学科,北京100034 [8]吉林大学第一医院检验医学科,吉林130061 [9]南方医科大学南方医院检验医学科,广州516006 [10]华中科技大学同济医学院附属同济医院检验医学科,武汉430030 [11]上海交通大学医学院附属瑞金医院检验医学科,上海200025

出　　处：《中华检验医学杂志》2023年第3期265-273,共9页Chinese Journal of Laboratory Medicine

摘　　要：目的评估人工智能(AI)血细胞形态分析仪(AI阅片机)对外周血白细胞检测的性能。方法多中心研究。(1)从全国11家三级医院收集3010份静脉血标本,用AI阅片机进行14类白细胞分析,并且将预分类结果与资深形态学专家审核后结果进行比较,以此评估AI阅片机白细胞预分类的符合率、检出率、特异度和一致性。(2)从3010份血液标本中选取400份血液标本(预分类人工审核后含异常白细胞的标本不少于50%),由形态学专家进行人工镜检并分析人工审核后的WBC分类结果和专家显微镜镜检结果之间的相关性。(3)从3010份血液标本中找出诊断为淋巴瘤、急性淋巴细胞白血病、急性髓系白血病、骨髓增生异常综合征和骨髓增殖性肿瘤患者的标本结果,通过比较预分类和专家审核后结果,反映AI阅片机对这5种恶性血液病中具有重要临床意义的异常白细胞的预分类的能力。WBC预分类和专家审核结果的一致性分析采用Cohen′s Kappa检验,比对试验采用Passing-Bablok回归分析,并根据公式统计符合率、检出率、特异度、准确度。结果(1)AI阅片机可对14类白细胞及有核红细胞进行预分类,与形态学专家审核后结果对比,白细胞总的预分类符合率达97.97%,其中正常类型白细胞预分类符合率均>96%,总的异常白细胞预分类符合率>87%。(2)AI阅片机专家审核后结果与人工镜检对比,各种白细胞类型及有核红细胞均具有较好的相关性(中性粒细胞、淋巴细胞、单核细胞、嗜酸性粒细胞、嗜碱性粒细胞、各阶段幼稚粒细胞、原始细胞、有核红细胞、恶性细胞均r>0.90,反应性淋巴细胞r=0.85),参考国际血液学复检专家组定义的白细胞阳性血涂片标准,AI阅片机对白细胞异常标本具有与人工镜检同等的筛查能力。(3)对5种恶性血液病中有较大意义的白细胞与形态学专家审核后结果进行对比,AI阅片机对原始细胞预分类�Objective To evaluate the performance of an artificial intelligent(AI)-based automated digital cell morphology analyzer(hereinafter referred as AI morphology analyzer)in detecting peripheral white blood cells(WBCs).Methods A multi-center study.1.A total of 3010 venous blood samples were collected from 11 tertiary hospitals nationwide,and 14 types of WBCs were analyzed with the AI morphology analyzers.The pre-classification results were compared with the post-classification results reviewed by senior morphological experts in evaluate the accuracy,sensitivity,specificity,and agreement of the AI morphology analyzers on the WBC pre-classification.2.400 blood samples(no less than 50%of the samples with abnormal WBCs after pre-classification and manual review)were selected from 3010 samples,and the morphologists conducted manual microscopic examinations to differentiate different types of WBCs.The correlation between the post-classification and the manual microscopic examination results was analyzed.3.Blood samples of patients diagnosed with lymphoma,acute lymphoblastic leukemia,acute myeloid leukemia,myelodysplastic syndrome,or myeloproliferative neoplasms were selected from the 3010 blood samples.The performance of the AI morphology analyzers in these five hematological malignancies was evaluated by comparing the pre-classification and post-classification results.Cohen′s kappa test was used to analyze the consistency of WBC pre-classification and expert audit results,and Passing-Bablock regression analysis was used for comparison test,and accuracy,sensitivity,specificity,and agreement were calculated according to the formula.Results 1.AI morphology analyzers can pre-classify 14 types of WBCs and nucleated red blood cells.Compared with the post-classification results reviewed by senior morphological experts,the pre-classification accuracy of total WBCs reached 97.97%,of which the pre-classification accuracies of normal WBCs and abnormal WBCs were more than 96%and 87%,respectively.2.The post-classification results re

关 键 词：人工智能 血细胞形态分析仪 白细胞分类 多中心研究 

分 类 号：R446.11[医药卫生—诊断学]