计算机科学 ›› 2022, Vol. 49 ›› Issue (4): 203-208.doi: 10.11896/jsjkx.201000153
赵凯, 安卫超, 张晓宇, 王彬, 张杉, 相洁
ZHAO Kai, AN Wei-chao, ZHANG Xiao-yu, WANG Bin, ZHANG Shan, XIANG Jie
摘要: 非增强CT扫描是急诊室诊断疑似脑出血的首选方法,医疗人员通常借助CT图像对疑似急性脑出血患者病灶部位进行手动分割,进而根据临床经验进行分类,这种人工诊断的方式对医师的经验要求较高,主观性较强,将分割和分类任务分开执行,不能充分利用两个任务间相关联的特征信息,时间成本高,增大了基于CT图像快速进行脑出血病灶部位分割及分类的难度。针对上述问题,文中提出了一种共享浅层参数多任务学习的脑出血图像分割及分类模型,一方面,根据不同任务学习的难易程度对损失函数的权值进行优化,另一方面,在多任务学习网络的浅层实现公有信息共享,深层提取不同任务的私有信息,获取更具代表性的特征,从而快速、准确地对脑出血患者的CT图像进行分割及分类。实验结果表明,共享浅层参数多任务学习网络生成的分割标注与真实标注有较好的视觉一致性。在最优权值下所有被试的平均Dice系数(DSC)为0.828,敏感度为0.842,特异度为0.985,阳性预测值(PPV)为0.838。共享浅层参数多任务学习网络分类的准确率、敏感度、特异度和AUC值分别为95.00%,90.48%,100.00%和0.982。与单任务深度学习、Y-Net以及借助分类辅助的多任务学习相比,该方法更加有效地利用了相关任务信息,同时通过调节损失函数权值,提升了出血病灶区域的分割和分类精度。
中图分类号:
[1] BALAMI J S,BUCHAN A M.Complications of intracerebralhaemorrhage[J].Lancet Neurology,2012,11(1):101-118. [2] ROH D,SUN C,SCHMIDT J M,et al.Primary intracerebral hemorrhage:a closer look at hypertension and cerebral amyloid angiopathy[J].Neurocritical Care,2018,29(1):77-83. [3] FUKUDA K,MAJUMDAR M,MASOUD H,et al.Multicenter assessment of morbidity associated with cerebral arteriovenous malformation hemorrhages[J].Journal of NeuroInterventional Surgery,2017,9(7):664-668. [4] WANG Z F,YIN M L,LIU S,et al.Microsurgical treatment of cerebral arteriovenous malformation in children[J].Journal of Clinical Neurosurgery,2020,17(2):171-175,181. [5] DING Z J,ZHANG X.The role of CT angiography in preoperative diagnosis and surgical evaluation of cerebral arteriovenous malformations and intracerebral hematoma[J].Chinese Clinical Doctor,2020,48(5):582-585. [6] LONG J,SHELHAMER E,DARRELL T.Fully convolutional networks for semantic segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,39(4):640-651. [7] RONNEBERGER O,FISCHER P,BROX T,et al.U-Net:con-volutional networks for biomedical image segmentation[C]//Medical Image Computing and Computer-Assisted Intervention.2015:234-241. [8] DHAR R,FALCONE G J,CHEN Y,et al.Deep learning forautomated measurement of hemorrhage and perihematomal edema in supratentorial intracerebral hemorrhage[J].Stroke,2020,51(2):648-651. [9] IRONSIDE N,CHEN C,MUTASA S,et al.Fully automatedsegmentation algorithm for perihematomal edema volumetry after spontaneous intracerebral hemorrhage[J].Stroke,2020,51(3):815-823. [10] YE Z Y,WU M L.Choroidal Neovascularization Segmentation Combining Temporal Supervision and Attention Mechanism[J].Computer Science,2021,48(8):118-124. [11] WANG T,LEI Y,TIAN S,et al.Learning-based automatic segmentation of arteriovenous malformations on contrast CT images in brain stereotactic radiosurgery[J].Medical Physics,2019,46(7):3133-3141. [12] CICEK O,ABDULKADIR A,LIENKAMP S S,et al.3DU-Net:learning dense volumetric segmentation from sparse annotation[J].arXiv:Computer Vision and Pattern Recognition,2016. [13] HAMIDIAN S,SAHINER B,PETRICK N,et al.3D convolu-tional neural network for automatic detection of lung nodules in chest CT[J].Proceedings of SPIE,2017,10134:1013409. [14] GAO X W,HUI R,TIAN Z,et al.Classification of CT brainimages based on deep learning networks[J].Computer Methods and Programs in Biomedicine,2017,138:49-56. [15] SILVA G L,VALENTE T L,SILVA A C,et al.Convolutional neural network-based PSO for lung nodule false positive reduction on CT images[J].Computer Methods and Programs in Bio-medicine,2018:109-118. [16] HE T,HU J,SONG Y,et al.Multi-task learning for the segmentation of organs at risk with label dependence[J].Medical Image Analysis,2020. [17] MEHTA S,MERCAN E,BARTLETT J,et al.Y-Net:joint seg-mentation and classification for diagnosis of breast biopsy images[J].arXiv:1806.01313. [18] ZHANG Y,ZHANG B,LIANG F,et al.Radiomics features on non-contrast-enhanced CT scan can precisely classify AVM-related hematomas from other spontaneous intraparenchymal hematoma types[J].European Radiology,2018,29(4):2157-2165. [19] NAJM M,KUANG H,FEDERICO A,et al.Automated brain extraction from head CT and CTA images using convex optimization with shape propagation[J].Computer Methods and Programs in Biomedicine,2019,176:1-8. [20] QIU W,YUAN J,UKWATTA E,et al.Prostate segmentation:an efficient convex optimization approach with axial symmetry using 3-D TRUS and MR images[J].IEEE Transactions on Medical Imaging,2014,33(4):947-960. [21] HE K,ZHANG X,REN S,et al.Delving deep into rectifiers:surpassing human-level performance on ImageNet classification[C]//Proceedings of the 2015 IEEE International Conference on Computer Vision (ICCV).2015:1206-1034. [22] LIU M,LI F,YAN H,et al.A multi-model deep convolutional neural network for automatic hippocampus segmentation and classification in Alzheimer’s disease[J].NeuroImage,2019,208:116459. |
[1] | 姜洋洋, 宋丽华, 邢长友, 张国敏, 曾庆伟. 蜜罐博弈中信念驱动的攻防策略优化机制 Belief Driven Attack and Defense Policy Optimization Mechanism in Honeypot Game 计算机科学, 2022, 49(9): 333-339. https://doi.org/10.11896/jsjkx.220400011 |
[2] | 周乐员, 张剑华, 袁甜甜, 陈胜勇. 多层注意力机制融合的序列到序列中国连续手语识别和翻译 Sequence-to-Sequence Chinese Continuous Sign Language Recognition and Translation with Multi- layer Attention Mechanism Fusion 计算机科学, 2022, 49(9): 155-161. https://doi.org/10.11896/jsjkx.210800026 |
[3] | 李宗民, 张玉鹏, 刘玉杰, 李华. 基于可变形图卷积的点云表征学习 Deformable Graph Convolutional Networks Based Point Cloud Representation Learning 计算机科学, 2022, 49(8): 273-278. https://doi.org/10.11896/jsjkx.210900023 |
[4] | 陈泳全, 姜瑛. 基于卷积神经网络的APP用户行为分析方法 Analysis Method of APP User Behavior Based on Convolutional Neural Network 计算机科学, 2022, 49(8): 78-85. https://doi.org/10.11896/jsjkx.210700121 |
[5] | 朱承璋, 黄嘉儿, 肖亚龙, 王晗, 邹北骥. 基于注意力机制的医学影像深度哈希检索算法 Deep Hash Retrieval Algorithm for Medical Images Based on Attention Mechanism 计算机科学, 2022, 49(8): 113-119. https://doi.org/10.11896/jsjkx.210700153 |
[6] | 檀莹莹, 王俊丽, 张超波. 基于图卷积神经网络的文本分类方法研究综述 Review of Text Classification Methods Based on Graph Convolutional Network 计算机科学, 2022, 49(8): 205-216. https://doi.org/10.11896/jsjkx.210800064 |
[7] | 金方焱, 王秀利. 融合RACNN和BiLSTM的金融领域事件隐式因果关系抽取 Implicit Causality Extraction of Financial Events Integrating RACNN and BiLSTM 计算机科学, 2022, 49(7): 179-186. https://doi.org/10.11896/jsjkx.210500190 |
[8] | 张颖涛, 张杰, 张睿, 张文强. 全局信息引导的真实图像风格迁移 Photorealistic Style Transfer Guided by Global Information 计算机科学, 2022, 49(7): 100-105. https://doi.org/10.11896/jsjkx.210600036 |
[9] | 戴朝霞, 李锦欣, 张向东, 徐旭, 梅林, 张亮. 基于DNGAN的磁共振图像超分辨率重建算法 Super-resolution Reconstruction of MRI Based on DNGAN 计算机科学, 2022, 49(7): 113-119. https://doi.org/10.11896/jsjkx.210600105 |
[10] | 刘月红, 牛少华, 神显豪. 基于卷积神经网络的虚拟现实视频帧内预测编码 Virtual Reality Video Intraframe Prediction Coding Based on Convolutional Neural Network 计算机科学, 2022, 49(7): 127-131. https://doi.org/10.11896/jsjkx.211100179 |
[11] | 徐鸣珂, 张帆. Head Fusion:一种提高语音情绪识别的准确性和鲁棒性的方法 Head Fusion:A Method to Improve Accuracy and Robustness of Speech Emotion Recognition 计算机科学, 2022, 49(7): 132-141. https://doi.org/10.11896/jsjkx.210100085 |
[12] | 孙福权, 崔志清, 邹彭, 张琨. 基于多尺度特征的脑肿瘤分割算法 Brain Tumor Segmentation Algorithm Based on Multi-scale Features 计算机科学, 2022, 49(6A): 12-16. https://doi.org/10.11896/jsjkx.210700217 |
[13] | 杜丽君, 唐玺璐, 周娇, 陈玉兰, 程建. 基于注意力机制和多任务学习的阿尔茨海默症分类 Alzheimer's Disease Classification Method Based on Attention Mechanism and Multi-task Learning 计算机科学, 2022, 49(6A): 60-65. https://doi.org/10.11896/jsjkx.201200072 |
[14] | 吴子斌, 闫巧. 基于动量的映射式梯度下降算法 Projected Gradient Descent Algorithm with Momentum 计算机科学, 2022, 49(6A): 178-183. https://doi.org/10.11896/jsjkx.210500039 |
[15] | 杨涵, 万游, 蔡洁萱, 方铭宇, 吴卓超, 金扬, 钱伟行. 基于步态分类辅助的虚拟IMU的行人导航方法 Pedestrian Navigation Method Based on Virtual Inertial Measurement Unit Assisted by GaitClassification 计算机科学, 2022, 49(6A): 759-763. https://doi.org/10.11896/jsjkx.211200148 |
|