Capsule Networks against Medical Imaging Data Challenges

Abstract

A key component to the success of deep learning is the availability of massive amounts of training data. Building and annotating large datasets for solving medical image classification problems is today a bottleneck for many applications. Recently, capsule networks were proposed to deal with shortcomings of Convolutional Neural Networks (ConvNets). In this work, we compare the behavior of capsule networks against ConvNets under typical datasets constraints of medical image analysis, namely, small amounts of annotated data and class-imbalance. We evaluate our experiments on MNIST, Fashion-MNIST and medical (histological and retina images) publicly available datasets. Our results suggest that capsule networks can be trained with less amount of data for the same or better performance and are more robust to an imbalanced class distribution, which makes our approach very promising for the medical imaging community.

arXiv comment10 pages, 3 figures, accepted at MICCAI-LABELS 2018 Workshop

References in corpus (3)

• Adam: A Method for Stochastic Optimization
• A Survey on Deep Learning in Medical Image Analysis
• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms

Cited by in corpus (7)

• The Multi-Lane Capsule Network (MLCN)
• Hierarchical Taxonomy-Aware and Attentional Graph Capsule RCNNs for Large-Scale Multi-Label Text Classification
• Bayesian Estimate of Mean Proper Scores for Diversity-Enhanced Active Learning
• Imbalanced Data Learning by Minority Class Augmentation using Capsule Adversarial Networks
• Examining the Benefits of Capsule Neural Networks
• Mask of truth: model sensitivity to unexpected regions of medical images
• Enabling Highly Efficient Capsule Networks Processing Through A PIM-Based Architecture Design