Rotaflip: A New CNN Layer for Regularization and Rotational Invariance in Medical Images
This addresses regularization and rotational invariance for medical imaging applications, but it is incremental as it builds on existing CNN architectures.
The paper tackled the problem of regularization in CNNs, where dropout can be detrimental in convolutional layers, by proposing a new layer that applies random rotations or reflections to feature maps, resulting in improved performance on medical image classification and segmentation tasks.
Regularization in convolutional neural networks (CNNs) is usually addressed with dropout layers. However, dropout is sometimes detrimental in the convolutional part of a CNN as it simply sets to zero a percentage of pixels in the feature maps, adding unrepresentative examples during training. Here, we propose a CNN layer that performs regularization by applying random rotations of reflections to a small percentage of feature maps after every convolutional layer. We prove how this concept is beneficial for images with orientational symmetries, such as in medical images, as it provides a certain degree of rotational invariance. We tested this method in two datasets, a patch-based set of histopathology images (PatchCamelyon) to perform classification using a generic DenseNet, and a set of specular microscopy images of the corneal endothelium to perform segmentation using a tailored U-net, improving the performance in both cases.