计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (6A): 230400043-7.doi: 10.11896/jsjkx.230400043

• 图像处理&多媒体技术 • 上一篇    下一篇

基于多距离测度异质集成学习的结肠病理图像细粒度分类研究

梁美彦1, 范莹莹1, 王琳2,3   

  1. 1 山西大学物理电子工程学院 太原 030006
    2 山西白求恩医院(山西医学科学院 同济山西医院),山西医科大学第三医院 太原 030032
    3 华中科技大学同济医学院附属同济医院 武汉 430030
  • 发布日期:2024-06-06
  • 通讯作者: 梁美彦(meiyanliang@sxu.edu.cn)
  • 基金资助:
    国家自然科学基金(11804209);山西省自然科学基金(201901D211173,202103021223411);山西省高等学校科技创新资助项目(2019L0064)

Fine-grained Colon Pathology Images Classification Based on Heterogeneous Ensemble Learningwith Multi-distance Measures

LIANG Meiyan1, FAN Yingying1, WANG Lin2,3   

  1. 1 Shanxi University School of Physics and Electronic Engineering,Taiyuan 030006,China
    2 Shanxi Bethune Hospital,Shanxi Academy of Medical Sciences,Tongji Shanxi Hospital,Third Hospital of Shanxi Medical University,Taiyuan 030032,China
    3 Tongji Hospital Tongji Medical College,Huazhong University of Science and Technology,Wuhan 430030,China
  • Published:2024-06-06
  • About author:LIANG Meiyan,born in 1984,Ph.D,associate professor.Her main research interests include machine learning,deep learning and medical image processing.
  • Supported by:
    National Natural Science Foundation of China(11804209),Natural Science Foundation of Shanxi Province,China(201901D211173,202103021223411) and Shanxi Higher Education Institution Science and TechnologyInnovation Grants Program,China(2019L0064).

PDF (PC)

268

摘要： 结肠病理学图像的细粒度分类对癌症治疗和预后评估都具有重要意义。然而,结肠病理学图像尤其是其组织学亚型图像在形态上极为相似,通过人工的方法进行高精度识别面临着巨大的挑战。而基于单个模型的计算机辅助诊断方法容易产生预测偏差。为此,提出了多距离测度异质集成学习的细粒度分类方法对结肠病理学微卫星状态进行分型预测。该方法分别通过余弦距离、曼哈顿距离与欧氏距离在潜在空间上度量每个基学习器输出的置信分数与理想解的差距,来集成不同基学习器的预测,再通过融合这些距离来提高模型的整体决策性能。实验结果表明,该方法在结肠病理学图像细粒度分类任务上,分类准确率、精确率、召回率与F-1分值都达到了94%以上,为病理学图像的亚型分类提供了新的视角。

关键词: 集成学习, 细粒度分类, 距离测度, 病理图像, 微卫星分型

Abstract: Fine-grained classification of colon pathology images is of great significance for both symptomatic treatment and prognosis assessment.However,the histopathological subtyping images of colon are extremely similar in morphology.It is a challenging task for manual methods to obtain high-precision predictions.Computer-aided diagnosis methods based on a single model also suffer from predictive bias in histological subtyping.Therefore,the fine-grained classification algorithm based on heterogeneous ensemble learning with multi-distance measures is proposed to predict the microsatellite state of colon pathology images.This method ensembles the predictions of the base learners by measuring the distance between the output confidence scores and the labels in latent space using Cosine distance,Manhattan distance,and Euclidean distance,respectively.Then,these distances are used to improve the overall decision performance of the model.The results show that the classification accuracy,precision,recall and F-1 score can reach 94% in the fine-grained classification,which provides a new perspective for subtype classification of pathological images.

Key words: Ensemble learning, Fine-grained classification, Distance measure, Pathology images, Microsatellite subtyping

中图分类号: 

  • TP391
[1]ZHENG R,ZHANG S,ZENG H,et al.Cancer incidence andmortality in China,2016[J].Journal of the National Cancer Center,2022,2(1):1-9.
[2]SUNG H,FERLAY J,SIEGEL R L,et al.Global cancer statistics 2020:GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J].CA:A Cancer Journal for Clinicians,2021,71(3):209-249.
[3]XIA C,DONG X,LI H,et al.Cancer statistics in China andUnited States,2022:profiles,trends,and determinants[J].Chinese Medical Journal,2022,135(5):584-590.
[4]YUAN Y.Chinese expert consensus on detection of microsatellite instability in colorectal cancer and other related solid tumors[J].Journal of Practical Oncology,2019(5):381-389.
[5]LI X,LI W,TAO R.Staged detection-identification frameworkfor cell nuclei in histopathology images[J].IEEE Transactions on Instrumentation and Measurement,2019,69(1):183-193.
[6]MALLICK P K,RYU S H,SATAPATHY S K,et al.BrainMRI image classification for cancer detection using deep wavelet autoencoder-based deep neural network[J].IEEE Access,2019,7:46278-46287.
[7]WANG T,LU W,YANG F,et al.Microsatellite instability prediction of uterine corpus endometrial carcinoma based on H&E histology whole-slide imaging[C]//2020 IEEE 17th International Symposium on Biomedical Imaging(ISBI).IEEE,2020:1289-1292
[8]SONG Z,YU C,ZOU S,et al.Automatic deep learning-basedcolorectal adenoma detection system and its similarities with pathologists[J].BMJ Open,2020,10(9):e036423.
[9]BYCHKOY D,LINDER N,TURKKI R,et al.Deep learningbased tissue analysis predicts outcome in colorectal cancer[J].Scientific Reports,2018,8(1):1-11.
[10]HAN Z,WEI B,ZHENG Y,et al.Breast cancer multi-classification from histopathological images with structured deep learning model[J].Scientific Reports,2017,7(1):1-10.
[11]GOUR M,JAIN S,SUNIL KUMAR T.Residual learning based CNN for breast cancer histopathological image classification[J].International Journal of Imaging Systems and Technology,2020,30(3):621-635.
[12]LI L,PAN X,YANG H,et al.Multi-task deep learning for fine-grained classification and grading in breast cancer histopathological images[J].Multimedia Tools and Applications,2020,79(21):14509-14528.
[13]CAO R,YANG F,MA S C,et al.Development and interpretation of a pathomics-based model for the prediction of microsatellite instability in Colorectal Cancer[J].Theranostics,2020,10(24):11080.
[14]KATHER J N,PEARSON A T,HALAMA N,et al.Deep learningcan predict microsatellite instability directly from histology in gastrointestinal cancer[J].Nature Medicine,2019,25(7):1054-1056.
[15]YAMASHITA R,LONG J,LONGACRE T,et al.Deep learning model for the prediction of microsatellite instability in colorectal cancer:a diagnostic study[J].The Lancet Oncology,2021,22(1):132-141.
[16]LEE H,SEO J,LEE G,et al.Two-stage classification method for MSI status prediction based on deep learning approach[J].Applied Sciences,2020,11(1):254.
[17]SAI VENKATESH C H,MERIGA C,GEETHIKA M,et al.Modified ResNetModel for MSI and MSS Classification of Gastrointestinal Cancer[C]//High Performance Computing and Networking.Springer,Singapore,2022:273-282.
[18]KHAN Z,LOGANATHAN R.Transfer Learning Based Classification of MSI and MSS Gastrointestinal Cancer[R].Easy Chair,2022.
[19]XUE D,ZHOU X,LI C,et al.An application of transfer learning and ensemble learning techniques for cervical histopathology image classification[J].IEEE Access,2020,8:104603-104618.
[20]BASHIR S,QAMAR U,KHAN F H.Heterogeneous classifiers fusion for dynamic breast cancer diagnosis using weighted vote based ensemble[J].Quality & Quantity,2015,49(5):2061-2076.
[21]SARWAR A,SHARMA V,GUPTA R.Hybrid ensemble learning technique for screening of cervical cancer using Papanicolaou smear image analysis[J].Personalized Medicine Universe,2015,4:54-62.
[22]WIN K P,KITJAIDURE Y,HAMAMOTO K,et al.Computer-assisted screening for cervical cancer using digital image processing of pap smear images[J].Applied Sciences,2020,10(5):1800.
[23]MANNA A,KUNDU R,KAPLUN D,et al.A fuzzy rank-based ensemble of CNN models for classification of cervical cytology[J].Scientific Reports,2021,11(1):1-18.
[24]SZEGEDY C,VANHOUCKE V,IOFFE S,et al.Rethinking the inception architecture for computer vision[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:2818-2826.
[25]HUANG G,LIU Z,VAN DER MAATEN L,et al.Densely connected convolutional networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:4700-4708.
[26]SZEGEDY C,IOFFE S,VANHOUCKE V,et al.Inception-v4,inception-resnet and the impact of residual connections on learning[C]//Thirty-first AAAI Conference on Artificial Intelligence.2017.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发