摘要
目的:研究基于三维卷积神经网络3DU-net神经网络的深度学习方法,在临床常规颞骨CT影像中,对中耳手术相关的迷路、面神经和听骨进行全自动分割的可行性。方法:随机调取门诊常规行颞骨CT检查患者的薄层CT扫描数据30例为正常结构组,既往耳蜗、面神经和听骨形态或走行变异者各1例为异常结构组。所有数据由两位临床医生在Mimics 20.0软件中,对面神经、迷路和听骨3个结构进行手工初分割和精细分割,同时利用3DU-Net对上述数据进行深度学习。分别对正常结构组中测试集5例和异常结构组中的迷路、听骨和面神经进行手工分割与自动分割的Dice相似指数(DSC)比较。结果:利用3DU-net网络结构对常规颞骨CT中迷路、听骨和面神经进行自动分割,其DSC分别为0.79±0.03、0.64±0.05和0.49±0.09;对异常的迷路、听骨和面神经的识别,其DSC也可达到0.71、0.54和0.40。结论:根据颞骨解剖特点,采用3DU-net神经网络结构,可以实现对迷路、听骨和面神经的全自动化分割,并获得接近手工分割的精度,该方法可行、快捷、准确度高。
Objective:To study the feasibility of fully automatic segmentation of labyrinth,facial nerve and ossicles in clinical routine temporal bone CT images based on 3DU-net neural network.Method:Clinical data were divided into two groups:①Normal group:data were randomly assigned from 30patients for routine temporal bone CT examination;②Abnormal group:cochlear,ossicles and facial nerve morphology variation of 1case each.The structures of facial nerve,labyrinth and ossicles were manually initial segmented and fine segmented by 2clinicians with Mimics 20.0.Three-dimensional convolutional neural network(3D U-Net)was selected to conduct deep learning on the same data.The dice similarity coefficient(DSC)was used as the evaluation index.Result:The 3DU-net neural network was used to automatically segment the labyrinth,ossicles and facial nerve in the routine temporal bone CT.In the normal group,the DSC of labyrinth,ossicles and facial nerve were 0.79±0.03,0.64±0.05and 0.49±0.09,respectively.In the abnormal group,the DSC of these structures were 0.71,0.54and 0.40.Conclusion:According to the anatomical characteristics of the temporal bone,the labyrinth,ossicles and the facial nerve can be totally automatic segmented by 3DU-net neural network,and the accuracy was closed to that of manual segmentation.This method is feasible,fast and accurate.
作者
柯嘉
吕弈
杜雅丽
王君臣
王江
孙世龙
马芙蓉
KE Jia;LV Yi;DU Yali;WANG Junchen;WANG Jiang;SUN Shilong;MA Furong(Department of Otorhinolaryngology Head and Neck Surgery,Third Hospital,Peking University,Beijing,100191,China;School of Mechanical Engineering and Automation,Beihang University)
出处
《临床耳鼻咽喉头颈外科杂志》
CAS
北大核心
2020年第10期870-873,共4页
Journal of Clinical Otorhinolaryngology Head And Neck Surgery
基金
国家自然科学基金(No:61701014,61911540075)
首都卫生发展科研专项项目资助(首发-2016-2-4094)
北京大学第三医院院临床重点项目(No:BYSY2017025)。
关键词
深度学习
卷积神经网络
中耳手术
面神经
自动分割
deep learning
convolutional neural network
middle ear surgery
facial nerve
automatic segmentation
