ABSTRACT Multiple Instance Learning (MIL) is a widely employed framework for learning on gigapixel whole-slide images (WSIs) from WSI-level annotations. In most MIL based analytical pipelines for WSI-level analysis, the WSIs are often divided into patches and deep features for patches (i.e., patch embeddings) are extracted prior to training to reduce the overall computational cost and cope with the GPUs' limited RAM. Experiments on the SICAPv2 dataset show that our approach outperforms MIL without augmentation and is on par with traditional Figure 1. The feature extractor can either be pre-trained on an auxiliary task [3] or based on self-supervised learning 1. INTRODUCTION Due to the large number of patches per WSI (can be > 10, 000), this step is computationally intensive and is Computational pathology has made significant progress typically only performed once. Then, in the second step, a in recent years with new methods capable of classifying neural network combines the low-dimensional patch embeddings high-dimensional whole-slide images (WSI) of the order of into a slide-level representation used for classification.

In this paper we propose a new augmentation technique, called patch augmentation, that, in our experiments, improves model accuracy and makes networks more robust to adversarial attacks. In brief, this data-independent approach creates new image data based on image/label pairs, where a patch from one of the two images in the pair is superimposed on to the other image, creating a new augmented sample. The new image's label is a linear combination of the image pair's corresponding labels. Initial experiments show a several percentage point increase in accuracy on CIFAR-10, from a baseline of approximately 81% to 89%. CIFAR-100 sees larger improvements still, from a baseline of 52% to 68% accuracy. Networks trained using patch augmentation are also more robust to adversarial attacks, which we demonstrate using the Fast Gradient Sign Method.