DECAPS: Detail-Oriented Capsule Networks

Capsule Networks (CapsNets) have demonstrated to be a promising alternative to Convolutional Neural Networks (CNNs). However, they often fall short of state-of-the-art accuracies on large-scale high-dimensional datasets. We propose a Detail-Oriented Capsule Network (DECAPS) that combines the strength of CapsNets with several novel techniques to boost its classification accuracies. First, DECAPS uses an Inverted Dynamic Routing (IDR) mechanism to group lower-level capsules into heads before sending them to higher-level capsules. This strategy enables capsules to selectively attend to small but informative details within the data which may be lost during pooling operations in CNNs. Second, DECAPS employs a Peekaboo training procedure, which encourages the network to focus on fine-grained information through a second-level attention scheme. Finally, the distillation process improves the robustness of DECAPS by averaging over the original and attended image region predictions. We provide extensive experiments on the CheXpert and RSNA Pneumonia datasets to validate the effectiveness of DECAPS. Our networks achieve state-of-the-art accuracies not only in classification (increasing the average area under ROC curves from 87.24% to 92.82% on the CheXpert dataset) but also in the weakly-supervised localization of diseased areas (increasing average precision from 41.7% to 80% for the RSNA Pneumonia detection dataset).

[1]  Yu Qiao,et al.  A Discriminative Feature Learning Approach for Deep Face Recognition , 2016, ECCV.

[2]  Lorenzo Torresani,et al.  STAR-Caps: Capsule Networks with Straight-Through Attentive Routing , 2019, NeurIPS.

[3]  Saurabh Prasad,et al.  GRAPH CONVOLUTIONAL NEURAL NETWORKS FOR HYPERSPECTRAL DATA CLASSIFICATION , 2018, 2018 IEEE Global Conference on Signal and Information Processing (GlobalSIP).

[4]  Yifan Yu,et al.  CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison , 2019, AAAI.

[5]  M. Mildner,et al.  Re-epithelialization and immune cell behaviour in an ex vivo human skin model , 2020, Scientific Reports.

[6]  Geoffrey E. Hinton,et al.  Dynamic Routing Between Capsules , 2017, NIPS.

[7]  Han Zhang,et al.  Dynamic Routing Capsule Networks for Mild Cognitive Impairment Diagnosis , 2019, MICCAI.

[8]  Tao Mei,et al.  Learning Multi-attention Convolutional Neural Network for Fine-Grained Image Recognition , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[9]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[10]  Yee Whye Teh,et al.  Stacked Capsule Autoencoders , 2019, NeurIPS.

[11]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[12]  Badrinath Roysam,et al.  Automated Classification of Apoptosis in Phase Contrast Microscopy Using Capsule Network , 2020, IEEE Transactions on Medical Imaging.

[13]  Yong Jae Lee,et al.  Hide-and-Seek: Forcing a Network to be Meticulous for Weakly-Supervised Object and Action Localization , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[14]  K. Plataniotis,et al.  \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\text {DRTOP}$$\end{document}DRTOP: deep learning-based radiomics for , 2020, Scientific Reports.

[15]  Ulas Bagci,et al.  Capsules for Object Segmentation , 2018, ArXiv.

[16]  Aryan Mobiny,et al.  Risk-Aware Machine Learning Classifier for Skin Lesion Diagnosis , 2019, Journal of clinical medicine.

[17]  Robertas Alzbutas,et al.  Convolutional capsule network for classification of breast cancer histology images , 2018, ICIAR.

[18]  Geoffrey E. Hinton,et al.  Matrix capsules with EM routing , 2018, ICLR.

[19]  Farnoosh Naderkhani,et al.  3D-MCN: A 3D Multi-scale Capsule Network for Lung Nodule Malignancy Prediction , 2020, Scientific Reports.

[20]  Kaiming He,et al.  Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[21]  Hien Van Nguyen,et al.  Fast CapsNet for Lung Cancer Screening , 2018, MICCAI.

[22]  Andrew Y. Paek,et al.  Assaying neural activity of children during video game play in public spaces: a deep learning approach , 2019, Journal of neural engineering.

[23]  Lukasz Kaiser,et al.  Attention is All you Need , 2017, NIPS.

[24]  Shadi Albarqouni,et al.  Capsule Networks against Medical Imaging Data Challenges , 2018, CVII-STENT/LABELS@MICCAI.

[25]  Sergey Ioffe,et al.  Rethinking the Inception Architecture for Computer Vision , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).