Parallel Capsule Networks for Classification of White Blood Cells

TL;DR

This paper introduces parallel Capsule Networks that improve stability and rotational invariance in classifying white blood cells, outperforming traditional CapsNets and ResNeXt-50 on an unbalanced leukemia dataset.

Contribution

The paper proposes a novel parallel CapsNet architecture that isolates capsules through branching, enhancing stability and invariance in complex image classification tasks.

Findings

Parallel CapsNets outperform ResNeXt-50 in accuracy.

Parallel CapsNets are more stable than conventional CapsNets.

Parallel CapsNets exhibit better rotational invariance.

Abstract

Capsule Networks (CapsNets) is a machine learning architecture proposed to overcome some of the shortcomings of convolutional neural networks (CNNs). However, CapsNets have mainly outperformed CNNs in datasets where images are small and/or the objects to identify have minimal background noise. In this work, we present a new architecture, parallel CapsNets, which exploits the concept of branching the network to isolate certain capsules, allowing each branch to identify different entities. We applied our concept to the two current types of CapsNet architectures, studying the performance for networks with different layers of capsules. We tested our design in a public, highly unbalanced dataset of acute myeloid leukaemia images (15 classes). Our experiments showed that conventional CapsNets show similar performance than our baseline CNN (ResNeXt-50) but depict instability problems. In contrast, parallel CapsNets can outperform ResNeXt-50, is more stable, and shows better rotational invariance than both, conventional CapsNets and ResNeXt-50.