Orientation-Disentangled Unsupervised Representation Learning for Computational Pathology
This work addresses the issue of irrelevant orientation variations in computational pathology, which can improve analysis of large histopathology datasets, though it is incremental as it builds on existing frameworks.
The paper tackled the problem of orientation entanglement in unsupervised representation learning for histopathology images by extending Variational Auto-Encoders with rotation-equivariant networks to separate oriented and isotropic components, resulting in higher performance in subsequent tasks compared to classical approaches.
Unsupervised learning enables modeling complex images without the need for annotations. The representation learned by such models can facilitate any subsequent analysis of large image datasets. However, some generative factors that cause irrelevant variations in images can potentially get entangled in such a learned representation causing the risk of negatively affecting any subsequent use. The orientation of imaged objects, for instance, is often arbitrary/irrelevant, thus it can be desired to learn a representation in which the orientation information is disentangled from all other factors. Here, we propose to extend the Variational Auto-Encoder framework by leveraging the group structure of rotation-equivariant convolutional networks to learn orientation-wise disentangled generative factors of histopathology images. This way, we enforce a novel partitioning of the latent space, such that oriented and isotropic components get separated. We evaluated this structured representation on a dataset that consists of tissue regions for which nuclear pleomorphism and mitotic activity was assessed by expert pathologists. We show that the trained models efficiently disentangle the inherent orientation information of single-cell images. In comparison to classical approaches, the resulting aggregated representation of sub-populations of cells produces higher performances in subsequent tasks.