计算机科学 ›› 2020, Vol. 47 ›› Issue (10): 200-206.doi: 10.11896/jsjkx.190900073

所属专题: 医学图像

• 计算机图形学&多媒体 • 上一篇    下一篇

联合成对学习和图像聚类的无监督肺癌亚型识别

任雪婷1, 赵涓涓1, 强彦1, Saad Abdul RAUF1, 刘继华2   

  1. 1 太原理工大学信息与计算机学院 太原030024
    2 吕梁学院计算机科学与技术学院 山西 吕梁033000
  • 收稿日期:2019-09-10 修回日期:2019-11-15 出版日期:2020-10-15 发布日期:2020-10-16
  • 通讯作者: 赵涓涓(zhaojuanjuan@tyut.edu.cn)
  • 作者简介:964325345@qq.com
  • 基金资助:
    国家自然科学基金项目(61872261);北京航空航天大学虚拟现实技术与系统国家重点实验室项目(VRLAB2018A08);山西省重点研发计划国际科技合作项目(201803D421036)

Lung Cancer Subtype Recognition with Unsupervised Learning Combining Paired Learning and Image Clustering

REN Xue-ting1, ZHAO Juan-juan1, QIANG Yan1, Saad Abdul RAUF1, LIU Ji-hua2   

  1. 1 School of Information and Computer Science,Taiyuan University of Technology,Taiyuan 030024,China
    2 School of Computer Science and Technology,Lvliang University,Lvliang,Shanxi 033000,China
  • Received:2019-09-10 Revised:2019-11-15 Online:2020-10-15 Published:2020-10-16
  • About author:EN Xue-ting,born in 1994,postgra-duate,is a member of China Computer Federation.Her main research interests include medical image processing and deep learning.
    ZHAO Juan-juan,born in 1975,Ph.D,professor,Ph.D supervisor,is a member of China Computer Federation.Her main research interests include intelligent information processing and image recognition.
  • Supported by:
    National Natural Science Foundation of China (61872261),Open Funding Project of State Key Laboratory of Virtual Reality Technology and Systems, Beihang University (VRLAB2018A08) and International Science and Technology Cooperation Project of Shanxi Key R&D Program (201803D421036)

摘要: 基因诊断是近年来提高肺癌治愈率的一种新型且有效的方法,但这种方法存在基因检测时间长、费用高、侵入式取样损伤大的问题。文中提出了基于成对学习和图像聚类的无监督学习的肺癌亚型识别方法。首先,采用无监督卷积特征融合网络用于学习肺癌CT图像的深度表示,有效地捕捉被忽略的重要特征信息,并使用包含不同层次抽象信息的最终融合特征来表征肺癌亚型。然后,使用联合成对学习和图像聚类的分类学习框架进行建模,充分利用学习到的特征表示,确保有效的聚类学习,以取得更高的分类精度。最后,利用生存分析和基因分析对肺癌亚型进行多角度验证。在合作医院和TCGA-LUAD数据集上的实验结果表明,该方法通过可靠无创的影像分析和放射成像技术,发现了3种具有不同分子特征的肺癌影像亚型,在降低基因检测问题的同时可有效辅助医师进行精准诊断和个性化治疗,进而提高肺癌患者的治愈生存率。

关键词: 成对学习, 肺癌亚型识别, 深度表示, 图像聚类, 无监督学习

Abstract: In recent years,gene diagnosis has been one of the new and effective methods to improve the cure rate of lung cancer,but it has the problems of time-consuming,high cost and serious damage from invasive sampling.In this paper,an unsupervised learning method of Lung cancer subtype recognition based on paired learning and image clustering is proposed.Firstly,the unsupervised convolution feature fusion network is used to learn the deep representation of lung cancer CT images and effectively capture the important feature information that is ignored,and the final fusion features containing different levels of abstract information is used to represent lung cancer subtypes.Then,the classification learning framework of combined paired learning and image clustering is used for modeling,and the learnt feature representation is fully utilized to ensure effective clustering learning,so as to achieve higher classification accuracy.Finally,survival analysis and gene analysis are used to verify lung cancer subtypes from multiple perspectives.Experiments on the data sets of the cooperative hospital and TCGA-LUAD show that,through reliable and non-invasive image analysis and radiological imaging technology,three subtypes of lung cancer with different molecular characte-ristics have been found by this method.It can effectively assist doctors in accurate diagnosis and personalized treatment while reducing problems in gene detection,so as to improve the survival rate of lung cancer patients.

Key words: Deep representation, Image clustering, Lung cancer subtype recognition, Paired learning, Unsupervised learning

中图分类号: 

  • TP391.41
[1]SHAO L,SONG Z B,ZHANG Y P,et al.Advances of Molecular Subtype and Targeted Therapy of Lung Cancer[J].Chinese Journal of Lung Cancer,2012,15(9):545-552.
[2]YOUNG J D,CAI C,LU X.Unsupervised deep learning reveals prognostically relevant subtypes of glioblastoma[J].BMC Bioinformatics,2017,18(S11):381.
[3]MAZUROWSKI M A,ZHANG J,GRIMM L J,et al.Radio-genomic Analysis of Breast Cancer:Luminal B Molecular Subtype Is Associated with Enhancement Dynamics at MR Imaging[J].Radiology,2014,273(2):365-372.
[4]WILSON R,DEVARAJ A.Radiomics of pulmonary nodules and lung cancer[J].Translational Lung Cancer Research,2017,6(1):86.
[5]ZHANG Y,OIKONOMOU A,WONG A,et al.Radiomics-basedprognosis analysis for non-small cell lung cancer[J].Scientific Reports,2017,7:46349.
[6]THAWANI R,MCLANE M,BEIG N,et al.Radiomics and radiogenomics in lung cancer:a review for the clinician[J].Lung Cancer,2018,115:34-41.
[7]WU J,CUI Y,SUN X L,et al.Unsupervised Clustering ofQuantitative Image Phenotypes Reveals Breast Cancer Subtypes with Distinct Prognoses and Molecular Pathways[J].Clinical Cancer Research,2017,23(13):3334-3342.
[8]ITAKURA H,ACHROL A S,MITCHELL L A,et al.Magnetic resonance image features identify glioblastoma phenotypic subtypes with distinct molecular pathway activities[J].Science Translational Medicine,2015,7(303):303ra138-303ra138.
[9]YANG J,PARIKH D,BATRA D.Joint unsupervised learning of deep representations and image clusters [C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:5147-5156.
[10]HAO R,QIANG Y,LIAO X,et al.An automatic detectionmethod for lung nodules based on multi-scale enhancement filters and 3D shape features[J].KSII Transactions on Internet & Information Systems,2019,13(1).
[11]WEST L,VIDWANS S J,CAMPBELL N P,et al.A novel classification of lung cancer into molecular subtypes[J].PloS one,2012,7(2):e31906.
[12]WU M,MA J.Association Between Imaging Characteristics and Different Molecular Subtypes of Breast Cancer[J].Academic Radiology,2016,24(4):426-434.
[13]RATHORE S,AKBARI H,ROZYCKI M,et al.Radiomic MRI signature reveals three distinct subtypes of glioblastoma with different clinical and molecular characteristics,offering prognostic value beyond IDH1[J].Scientific Reports,2018,8(1):5087-5098.
[14]LAMBIN P,RIOS-VELAZQUEZ E,LEIJENAAR R,et al.Radiomics:extracting more information from medical images using advanced feature analysis[J].European Journal of Cancer,2012,48(4):441-446.
[1] 宋杰, 梁美玉, 薛哲, 杜军平, 寇菲菲.
基于无监督集群级的科技论文异质图节点表示学习方法
Scientific Paper Heterogeneous Graph Node Representation Learning Method Based onUnsupervised Clustering Level
计算机科学, 2022, 49(9): 64-69. https://doi.org/10.11896/jsjkx.220500196
[2] 韩洁, 陈俊芬, 李艳, 湛泽聪.
基于自注意力的自监督深度聚类算法
Self-supervised Deep Clustering Algorithm Based on Self-attention
计算机科学, 2022, 49(3): 134-143. https://doi.org/10.11896/jsjkx.210100001
[3] 侯宏旭, 孙硕, 乌尼尔.
蒙汉神经机器翻译研究综述
Survey of Mongolian-Chinese Neural Machine Translation
计算机科学, 2022, 49(1): 31-40. https://doi.org/10.11896/jsjkx.210900006
[4] 陈扬, 王金亮, 夏炜, 杨颢, 朱润, 奚雪峰.
基于特征自动提取的足迹图像聚类方法
Footprint Image Clustering Method Based on Automatic Feature Extraction
计算机科学, 2021, 48(6A): 255-259. https://doi.org/10.11896/jsjkx.200900033
[5] 李向利, 贾梦雪.
基于预处理的超图非负矩阵分解算法
Nonnegative Matrix Factorization Algorithm with Hypergraph Based on Per-treatments
计算机科学, 2020, 47(7): 71-77. https://doi.org/10.11896/jsjkx.200200106
[6] 李金霞, 赵志刚, 李强, 吕慧显, 李明生.
改进的局部和相似性保持特征选择算法
Improved Locality and Similarity Preserving Feature Selection Algorithm
计算机科学, 2020, 47(6A): 480-484. https://doi.org/10.11896/JsJkx.20190800095
[7] 王成章, 白晓明, 杜金栗.
图像的扩散界面无监督聚类算法
Diffuse Interface Based Unsupervised Images Clustering Algorithm
计算机科学, 2020, 47(5): 149-153. https://doi.org/10.11896/jsjkx.190300125
[8] 罗月,童卞,景帅,张蒙,饶永明,闫峰.
基于卷积去噪自编码器的芯片表面弱缺陷检测方法
Detection Method of Chip Surface Weak Defect Based on Convolution Denoising Auto-encoders
计算机科学, 2020, 47(2): 118-125. https://doi.org/10.11896/jsjkx.190100141
[9] 周昌, 李向利, 李俏霖, 朱丹丹, 陈世莲, 蒋丽榕.
基于余弦相似度的稀疏非负矩阵分解算法
Sparse Non-negative Matrix Factorization Algorithm Based on Cosine Similarity
计算机科学, 2020, 47(10): 108-113. https://doi.org/10.11896/jsjkx.190700112
[10] 陈深进, 薛洋.
基于改进卷积神经网络的短时公交客流预测
Short-term Bus Passenger Flow Prediction Based on Improved Convolutional Neural Network
计算机科学, 2019, 46(5): 175-184. https://doi.org/10.11896/j.issn.1002-137X.2019.05.027
[11] 曾凡智, 周燕, 余家豪, 罗粤, 邱腾达, 钱杰昌.
基于无监督学习的二维工程CAD模型端到端检索算法
End-to-End Retrieval Algorithm of Two-dimensional Engineering CAD Model Based on Unsupervised Learning
计算机科学, 2019, 46(12): 298-305. https://doi.org/10.11896/jsjkx.190900003
[12] 李锋,谢嗣弘.
基于无监督学习的移动心电信号异常诊断研究
Study on Abnormal Diagnosis of Moving ECG Signals Based on Unsupervised Learning
计算机科学, 2017, 44(Z11): 68-71. https://doi.org/10.11896/j.issn.1002-137X.2017.11A.013
[13] 吴镜锋,金炜东,唐鹏.
数据异常的监测技术综述
Survey on Monitoring Techniques for Data Abnormalities
计算机科学, 2017, 44(Z11): 24-28. https://doi.org/10.11896/j.issn.1002-137X.2017.11A.004
[14] 刘旋,王国胤,罗小波.
基于高斯云变换的遥感图像多粒度聚类
Multi-granularity Clustering of Remote Sensing Image Based on Gaussian Cloud Transformation
计算机科学, 2017, 44(9): 23-27. https://doi.org/10.11896/j.issn.1002-137X.2017.09.004
[15] 李丽萍,赵传荣,孔德仁,王芳.
基于图论的无监督区域遥感图像检索算法研究
Research on Unsupervised Regional Remote Sensing Image Retrieval Algorithm Based on Graph Theory
计算机科学, 2017, 44(7): 315-317. https://doi.org/10.11896/j.issn.1002-137X.2017.07.057
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!