Strategies to Reduce the Expert Supervision Required for Deep Learning-Based Segmentation of Histopathological Images

Abstract

The emergence of computational pathology comes with a demand to extract more and more information from each tissue sample. Such information extraction often requires the segmentation of numerous histological objects (e.g., cell nuclei, glands, etc.) in histological slide images, a task for which deep learning algorithms have demonstrated their effectiveness. However, these algorithms require many training examples to be efficient and robust. For this purpose, pathologists must manually segment hundreds or even thousands of objects in histological images, i.e., a long, tedious and potentially biased task. The present paper aims to review strategies that could help provide the very large number of annotated images needed to automate the segmentation of histological images using deep learning. This review identifies and describes four different approaches: the use of immunohistochemical markers as labels, realistic data augmentation, Generative Adversarial Networks (GAN), and transfer learning. In addition, we describe alternative learning strategies that can use imperfect annotations. Adding real data with high-quality annotations to the training set is a safe way to improve the performance of a well configured deep neural network. However, the present review provides new perspectives through the use of artificially generated data and/or imperfect annotations, in addition to transfer learning opportunities.

Image Descriptors for Weakly Annotated Histopathological Breast Cancer Data

“…Based on the results gathered from the experiments, we concluded that WAID is able to achieve state-of-the-art performance on a database that contains weakly annotated images. As personalized medicine becomes prevalent, medical experts are faced with high demands to create automation of their most recurrent tasks and for a more complex set of analyses to be done (34). The average patient waits approximately 10 days for a pathology result, which can be critical for some patients when it comes to treatment plans as their safety and health are at risk 2 .…”

Quick Annotator: an open‐source digital pathology based rapid image annotation tool

“…For example, the number of nuclei in a single WSI can order into the hundreds of thousands, making accurately individually annotating each cell unfeasible. While a number of image analysis based algorithms have been proposed to help reduce annotation effort, they are not yet integrated into tools with user interfaces enabling their employment [ 5 , 6 , 7 , 8 ]. Other efforts have resulted in proprietary closed‐source tools [ 9 , 10 ] which can be too costly to purchase in academic settings, or do not provide an open environment for facile testing and integration of new algorithms.…”

A User-Friendly Tool For Cloud-Based Whole Slide Image Segmentation, With Examples From Renal Histopathology

“…CNNs have been successfully utilized by many research groups for the segmentation of WSIs [4][5][6][7][8][9] . However, thus far tools to segment WSIs have been complex to deploy and use, requiring knowledge of the command line interface and computational expertise [10][11][12] . The ideal user for these tools is the pathologist or biological scientist, whose clinical work ow or research questions could bene t from fast and accurate segmentation of relevant structures 2 .…”