Semi-supervised learning in cancer diagnostics

被引:9
|
作者
Eckardt, Jan-Niklas [1 ,2 ]
Bornhaeuser, Martin [1 ,3 ,4 ]
Wendt, Karsten [2 ,5 ]
Middeke, Jan Moritz [1 ,2 ]
机构
[1] Univ Hosp Carl Gustav Carus, Dept Internal Med 1, Dresden, Germany
[2] Tech Univ Dresden, Else Kroner Fresenius Ctr Digital Hlth, Dresden, Germany
[3] German Consortium Translat Canc Res, Heidelberg, Germany
[4] Natl Ctr Tumor Dis NCT, Dresden, Germany
[5] Tech Univ Dresden, Inst Software & Multimedia Technol, Dresden, Germany
来源
FRONTIERS IN ONCOLOGY | 2022年 / 12卷
关键词
semi-supervised learning; cancer; diagnostics; artificial intelligence; machine learning; SEGMENTATION;
D O I
10.3389/fonc.2022.960984
中图分类号
R73 [肿瘤学];
学科分类号
100214 ;
摘要
In cancer diagnostics, a considerable amount of data is acquired during routine work-up. Recently, machine learning has been used to build classifiers that are tasked with cancer detection and aid in clinical decision-making. Most of these classifiers are based on supervised learning (SL) that needs time- and cost-intensive manual labeling of samples by medical experts for model training. Semi-supervised learning (SSL), however, works with only a fraction of labeled data by including unlabeled samples for information abstraction and thus can utilize the vast discrepancy between available labeled data and overall available data in cancer diagnostics. In this review, we provide a comprehensive overview of essential functionalities and assumptions of SSL and survey key studies with regard to cancer care differentiating between image-based and non-image-based applications. We highlight current state-of-the-art models in histopathology, radiology and radiotherapy, as well as genomics. Further, we discuss potential pitfalls in SSL study design such as discrepancies in data distributions and comparison to baseline SL models, and point out future directions for SSL in oncology. We believe well-designed SSL models to strongly contribute to computer-guided diagnostics in malignant disease by overcoming current hinderances in the form of sparse labeled and abundant unlabeled data.
引用
收藏
页数:10
相关论文
共 50 条
  • [1] Semi-supervised Learning
    Adams, Niall
    [J]. JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES A-STATISTICS IN SOCIETY, 2009, 172 : 530 - 530
  • [2] On semi-supervised learning
    Cholaquidis, A.
    Fraiman, R.
    Sued, M.
    [J]. TEST, 2020, 29 (04) : 914 - 937
  • [3] On semi-supervised learning
    A. Cholaquidis
    R. Fraiman
    M. Sued
    [J]. TEST, 2020, 29 : 914 - 937
  • [4] Learning Semi-Supervised Representation Towards a Unified Optimization Framework for Semi-Supervised Learning
    Li, Chun-Guang
    Lin, Zhouchen
    Zhang, Honggang
    Guo, Jun
    [J]. 2015 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2015, : 2767 - 2775
  • [5] Semi-supervised Sequence Learning
    Dai, Andrew M.
    Le, Quoc V.
    [J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 28 (NIPS 2015), 2015, 28
  • [6] Semi-supervised learning by disagreement
    Zhou, Zhi-Hua
    Li, Ming
    [J]. KNOWLEDGE AND INFORMATION SYSTEMS, 2010, 24 (03) : 415 - 439
  • [7] A survey on semi-supervised learning
    Jesper E. van Engelen
    Holger H. Hoos
    [J]. Machine Learning, 2020, 109 : 373 - 440
  • [8] A semi-supervised learning approach for bladder cancer grading
    Wenger, Kenneth
    Tirdad, Kayvan
    Cruz, Alex Dela
    Mari, Andrea
    Basheer, Mayada
    Kuk, Cynthia
    van Rhijn, Bas W. G.
    Zlotta, Alexandre R.
    van der Kwast, Theodorus H.
    Sadeghian, Alireza
    [J]. MACHINE LEARNING WITH APPLICATIONS, 2022, 9
  • [9] Semi-Supervised Incremental Learning
    Bouchachia, Abdelhamid
    Prossegger, Markus
    Duman, Hakan
    [J]. 2010 IEEE INTERNATIONAL CONFERENCE ON FUZZY SYSTEMS (FUZZ-IEEE 2010), 2010,
  • [10] Semi-supervised learning by disagreement
    Zhi-Hua Zhou
    Ming Li
    [J]. Knowledge and Information Systems, 2010, 24 : 415 - 439