基于深度学习的人工智能模型自动识别颈动脉斑块  

Deep Learning-Based Artificial Intelligence Model for Automatic Carotid Plaque Identification

作　　者：赫兰 申锷[2] 杨泽堃 张颖[4] 王玉东 陈伟导 王一同 贺永明[5] HE Lan;SHEN E;YANG Zekun;ZHANG Ying;WANG Yudong;CHEN Weidao;WANG Yitong;HE Yongming(Department of Ultrasound Medicine,Shanghai Chest Hospital,Shanghai Jiao Tong University School of Medicine,Shanghai,200030;Department of Ultrasound Medicine,Songjiang Hospital Affiliated to Shanghai Jiao Tong University School of Medicine,Shanghai,201600;Beijing Infervision Technology Co.,Ltd.,Beijing,100020;Department of Ultrasound Medicine,Xinhua Hospital,Dalian University,Dalian,116021;Department of Cardiology,The First Affiliated Hospital of Soochow University,Suzhou,215006)

机构地区：[1]上海市胸科医院/上海交通大学医学院附属胸科医院超声科,上海市200030 [2]上海交通大学医学院附属松江医院超声科,上海市201600 [3]推想医疗科技股份有限公司,北京市100020 [4]大连大学附属新华医院超声医学科,大连市116021 [5]苏州大学附属第一医院心内科,苏州市215006

出　　处：《中国医疗器械杂志》2024年第4期361-366,共6页Chinese Journal of Medical Instrumentation

基　　金：上海市徐汇区智慧医疗专项研究项目(XHZH202108);上海市徐汇区人工智能医疗院地合作项目(23XHYD-22)。

摘　　要：该研究旨在构建一个用于颈动脉斑块超声图像的有无判定的数据集,由1 165例受检者的1 761张超声图像组成。研究采用了一种融合了双线性卷积神经网络与残差神经网络的深度学习架构,即单输入BCNN-ResNet模型,以辅助临床医生通过颈动脉超声图像进行斑块的诊断。该模型经过训练以及内部和外部验证后,在内部验证中,ROC AUC达到了0.99,其95%置信区间为(0.91, 0.84),在外部验证中ROC AUC为0.95,其95%置信区间为(0.96, 0.94),此表现优于ResNet-34网络模型在内部验证中0.98 AUC的95%置信区间(0.99,0.95)和外部验证中0.94 AUC的95%置信区间(0.95, 0.92)。因此,单输入BCNN-ResNet网络模型展示了优异的诊断性能,为颈动脉斑块的自动识别提供了一种创新的解决方案。This study aims at developing a dataset for determining the presence of carotid artery plaques in ultrasound images,composed of 1761 ultrasound images from 1165 participants.A deep learning architecture that combines bilinear convolutional neural networks with residual neural networks,known as the single-input BCNN-ResNet model,was utilized to aid clinical doctors in diagnosing plaques using carotid ultrasound images.Following training,internal validation,and external validation,the model yielded an ROC AUC of 0.99(95%confidence interval:0.91 to 0.84)in internal validation and 0.95(95%confidence interval:0.96 to 0.94)in external validation,surpassing the ResNet-34 network model,which achieved an AUC of 0.98(95%confidence interval:0.99 to 0.95)in internal validation and 0.94(95%confidence interval:0.95 to 0.92)in external validation.Consequently,the single-input BCNN-ResNet network model has shown remarkable diagnostic capabilities and offers an innovative solution for the automatic detection of carotid artery plaques.

关 键 词：单输入BCNN-ResNet网络模型 颈动脉超声 深度学习 

分 类 号：R445.1[医药卫生—影像医学与核医学]