Multiple Instance Learning for Heterogeneous Images: Training a CNN for Histopathology

被引:41
作者
Couture, Heather D. [1 ]
Marron, J. S. [2 ,3 ]
Perou, Charles M. [2 ,4 ]
Troester, Melissa A. [2 ,5 ]
Niethammer, Marc [1 ,6 ]
机构
[1] Univ N Carolina, Dept Comp Sci, Chapel Hill, NC 27515 USA
[2] Univ N Carolina, Lineberger Comprehens Canc Ctr, Chapel Hill, NC 27515 USA
[3] Univ N Carolina, Dept Stat & Operat Res, Chapel Hill, NC 27515 USA
[4] Univ N Carolina, Dept Genet, Chapel Hill, NC 27515 USA
[5] Univ N Carolina, Dept Epidemiol, Chapel Hill, NC 27515 USA
[6] Univ N Carolina, Biomed Res Imaging Ctr, Chapel Hill, NC 27515 USA
来源
MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2018, PT II | 2018年 / 11071卷
关键词
D O I
10.1007/978-3-030-00934-2_29
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
Multiple instance (MI) learning with a convolutional neural network enables end-to-end training in the presence of weak image-level labels. We propose a new method for aggregating predictions from smaller regions of the image into an image-level classification by using the quantile function. The quantile function provides a more complete description of the heterogeneity within each image, improving image-level classification. We also adapt image augmentation to the MI framework by randomly selecting cropped regions on which to apply MI aggregation during each epoch of training. This provides a mechanism to study the importance of MI learning. We validate our method on five different classification tasks for breast tumor histology and provide a visualization method for interpreting local image classifications that could lead to future insights into tumor heterogeneity.
引用
收藏
页码:254 / 262
页数:9
相关论文
共 16 条
[1]  
Andrews Stuart, 2002, Proceedings of the 15th International Conference on Neural Information Processing Systems. NIPS'02, P561
[2]  
[Anonymous], ARXIV170100794
[3]  
[Anonymous], THESIS
[4]  
[Anonymous], 2016, ICPR
[5]  
Hiley Crispin T, 2014, Clin Med (Lond), V14 Suppl 6, ps33, DOI 10.7861/clinmedicine.14-6-s33
[6]  
Hou Le, 2016, CVPR
[7]   Computer-aided diagnosis from weak supervision: A benchmarking study [J].
Kandemir, Melih ;
Hamprecht, Fred A. .
COMPUTERIZED MEDICAL IMAGING AND GRAPHICS, 2015, 42 :44-50
[8]   Classifying and segmenting microscopy images with deep multiple instance learning [J].
Kraus, Oren Z. ;
Ba, Jimmy Lei ;
Frey, Brendan J. .
BIOINFORMATICS, 2016, 32 (12) :52-59
[9]   ImageNet Classification with Deep Convolutional Neural Networks [J].
Krizhevsky, Alex ;
Sutskever, Ilya ;
Hinton, Geoffrey E. .
COMMUNICATIONS OF THE ACM, 2017, 60 (06) :84-90
[10]   Biological and Therapeutic Impact of Intratumor Heterogeneity in Cancer Evolution [J].
McGranahan, Nicholas ;
Swanton, Charles .
CANCER CELL, 2015, 27 (01) :15-26