基于区域建议网络的肺结节检测及其去假阳性  

Pulmonary Nodules Detection and Its Removing False Positive Based on Region Proposal Network

作　　者：杨柳 杨勇[1] 叶宏伟 王小状 YANG Liu;YANG Yong;YE Hong-wei;WANG Xiao-zhuang(Department of Automation,Hangzhou Dianzi University,Hangzhou 310018,China;Zhejiang MinFound Intelligent Healthcare Technology Co.,Ltd.,Hangzhou 310018,China)

机构地区：[1]杭州电子科技大学自动化学院,杭州310018 [2]浙江明峰智能医疗科技有限公司,杭州310018

出　　处：《科学技术与工程》2022年第25期11105-11112,共8页Science Technology and Engineering

基　　金：国家自然科学基金(81671038,81171251,81871071);浙江省属高校基本科研业务费专项资金(GK209907299001-005)。

摘　　要：基于深度学习的肺结节检测技术不断发展,在辅助医生进行肺结节检查的任务中极大提升了肺结节的检出率和诊断的准确率。采用深度学习技术,提出了一种基于区域建议网络(region proposal network,RPN)结构的肺结节检测方法。针对肺结节的去假阳性阶段,将多个分类网络进行了性能对比。在快速区域卷积神经网络(faster region-based convolutional neural network,Faster R-CNN)上进行改进,使用SE(squeeze-and-excitation)结构以及ResNeXt的残差块构成特征提取模块,再结合UNet++网络结构,输出多个尺度的结果。最后将多尺度结果应用在3D RPN候选检测网络和R-CNN网络上,得到了灵敏度较高、假阳率更低的候选结节检测网络。在去假阳性结节网络阶段,用三维深度卷积神经网络(3D deep convolutional neural network,3D DCNN)网络对候选肺结节进行假阳性的筛除,有效去除了部分假阳性肺结节,提升了多个FP/scan检查点的灵敏度。最终得出灵敏度98.8%(8 FP/scan),竞争性指标(competition performance metric,CPM)达到0.879。在去假阳性结节方面,验证了3D DCNN网络相较于其他网络能够取得最好的效果,达到了15.6%的去假阳率。总的来说,所提网络进一步提升了检测的灵敏度,网络模型达到了较好的检测效果。在去假阳性网络方面,得出3D DCNN作为去假阳性网络具有比其他一些网络模型更好的效果。The detection technology of pulmonary nodules based on the deep learning method is constantly developing,improving the detection rate and diagnostic accuracy of pulmonary nodules,assisting doctors in the task of pulmonary nodules examination.A deep learning method based on region proposal network(RPN)structure was proposed to detect pulmonary nodules,as well as an improved faster region-based convolutional neural network(Faster R-CNN)was investigated.The performance of multiple classification networks was compared for the false-positive phase of pulmonary nodules.The squeeze-and-excitation(SE)structure and ResNeXt residual block were used to form the feature extraction module.Combined with the UNet++network structure,the multi-scale results were output.Finally,the multi-scale results were applied to the 3 D RPN candidate detection network and R-CNN network,then the candidate nodule detection network with higher sensitivity and lower false positive rate was obtained.In the stage of the false-positive nodules task,a 3 D deep convolutional neural network(3 D DCNN)was used to remove false positive pulmonary nodules for candidate pulmonary nodules,effectively removing some false positive pulmonary nodules and improving the sensitivity of multiple FPs/scan checkpoints.The results show that the sensitivity of the network was 98.8%(8 FP/scan),and the competition performance metric(CPM)was 0.879.In terms of the removal of false-positive nodules,the 3 D DCNN network could achieve the best effect in several networks,reaching a false-positive removal rate of 15.6%.It is concluded that the candidate nodule network further improved the detection sensitivity,and the network model achieved an outstanding detection effect.In false-positive network removal,as a false positive network,3 D DCNN had a better effect than some other network models.

关 键 词：多尺度网络 肺结节检测 Faster R-CNN网络 去假阳性网络 深度卷积神经网络 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]