Deep Learning Utilities for Pathology
Contents
Deep Learning Utilities for Pathology#
The evaluation of histopathological tissue sections is an often used tool in the diagnosis and prognosis of cancer. While historically studied with a microscope, these images are digitized at a high pace. These image are often so large that a collection of them will not fit into today’s computer memory, as well as resulting in complex shapes with hard-to-define boundaries. Dlup is a collection of software tools used to facilitate computational pathology research, with a special focus on deep learning. Dlup aims to ease the development of new models by simplifying the retrieval and management of whole-slide images, so that they can be quickly turned into datasets ready to be used for model training and inference.