魏子伊,汤奕,滕泽,李宏锋,彭芸,操江峰,高天姿,张恒,韩鸿宾.胸部X线片人工智能联邦学习系统用于病原学诊断儿童社区获得性肺炎[J].中国介入影像与治疗学,2024,21(6):368-373
胸部X线片人工智能联邦学习系统用于病原学诊断儿童社区获得性肺炎
Artificial intelligence federated learning system based on chest X-ray films for pathogen diagnosis of community-acquired pneumonia in children
投稿时间:2024-05-17  修订日期:2024-05-31
DOI:10.13929/j.issn.1672-8475.2024.06.012
中文关键词:  肺炎  儿童  X线  人工智能
英文关键词:pneumonia  child  X-rays  artificial intelligence
基金项目:国家自然科学基金(62301615)。
作者单位E-mail
魏子伊 北京大学医学部医学技术研究院, 北京 100191  
汤奕 北京大学医学部医学技术研究院, 北京 100191  
滕泽 国家癌症中心/国家肿瘤临床医学研究中心/中国医学科学院北京协和医学院肿瘤医院影像诊断科, 北京 100021  
李宏锋 北京大学医学部医学技术研究院, 北京 100191  
彭芸 国家儿童医学中心 首都医科大学附属北京儿童医院影像中心, 北京 100045  
操江峰 中国科学院信息工程研究所, 北京 100093  
高天姿 北京大学医学部医学技术研究院, 北京 100191  
张恒 承德医学院生物医学工程系, 河北 承德 067000  
韩鸿宾 北京大学医学部医学技术研究院, 北京 100191
北京大学第三医院放射科, 北京 100191
磁共振成像设备与技术北京市重点实验室, 北京 100191 
hanhongbin@bjmu.edu.cn 
摘要点击次数: 92
全文下载次数: 107
中文摘要:
      目的 观察基于胸部X线片建立的人工智能联邦学习系统用于病原学诊断儿童社区获得性肺炎(CAP)的价值。方法 回顾性选取2所医院共900例CAP患儿,包括细菌性、病毒性及支原体CAP各300例,对每例选取1幅胸部正位片。收集公开数据集GWCMCx中的5 856幅儿童胸部正位片,分别来自4 273例CAP患儿和1 583例胸部无明显异常患儿。按8∶2比例将全部6 756幅胸片分为训练集(n=5 359)与验证集(n=1 397)。建立基于注意力机制的病原学诊断儿童CAP模型,设计二分类及三分类诊断算法并进行联邦部署训练;与DenseNet模型对比,观察所获学习系统用于病原学诊断儿童CAP的效能。结果 人工智能联邦学习系统模型针对全部数据诊断CAP的准确率为97.00%,曲线下面积(AUC)为0.990。基于来自医院的数据,本系统根据单一影像学数据及临床-影像学数据实现病原学诊断儿童CAP的AUC分别为 0.858及0.836,均高于DenseNet模型的0.740(P均<0.05)。结论 基于胸部X线片的人工智能联邦学习系统可用于病原学诊断儿童CAP。
英文摘要:
      Objective To explore the value of artificial intelligence federated learning system based on chest X-ray films for pathogen diagnosis of community-acquired pneumonia (CAP) in children. Methods Totally 900 cases of CAP children from 2 hospitals were retrospectively enrolled, including bacterial, viral and mycoplasma CAP (each n=300), and chest posteroanterior X-ray films were collected. Meanwhile, chest posteroanterior X-ray films of 5 856 children from the publicly available dataset GWCMCx were collected, including 4 273 CAP images and 1 583 healthy chest images. All above 6 756 images were divided into training set (n=5 359) and validation set (n=1 397) at the ratio of 8∶2. Then a pathogen diagnosis model of children CAP was established based on attention mechanism. Binary and ternary diagnostic algorithms were designed, and federated deployment training was performed. The efficacy of this system for pathogen diagnosis of children CAP was analyzed and compared with DenseNet model. Results Based on all data, the accuracy of the obtained artificial intelligence federated learning system model for diagnosing children CAP was 97.00%, with the area under the curve (AUC) of 0.990. Based on hospital data, the AUC of this system using single imaging data and clinical-imaging data for pathogen diagnosis of children CAP was 0.858 and 0.836, respectively, both better than that of DenseNet model (0.740, both P<0.05). Conclusion The artificial intelligence federated learning system based on chest X-ray films could be used for pathogen diagnosis of children CAP.
查看全文  查看/发表评论  下载PDF阅读器
关闭