Field Of Interest Proposal for Augmented Mitotic Cell Count: Comparison of two Convolutional Networks

TL;DR

This study compares two CNN-based methods for identifying regions with the highest mitotic activity in whole slide images to improve tumor grading accuracy, demonstrating high correlation with ground truth despite minimal differences in proposed regions.

Contribution

The paper introduces and evaluates two CNN-based segmentation approaches at different resolutions for selecting regions of interest in histopathology slides.

Findings

High correlation (0.94 vs. 0.83) with ground truth for the fine resolution method.

Both methods effectively identify regions with high mitotic counts.

Proposed regions contain mitotic counts in the upper quartile of slides.

Abstract

Most tumor grading systems for human as for veterinary histopathology are based upon the absolute count of mitotic figures in a certain reference area of a histology slide. Since time for prognostication is limited in a diagnostic setting, the pathologist will often almost arbitrarily choose a certain field of interest assumed to have the highest mitotic activity. However, as mitotic figures are commonly very sparse on the slide and often have a patchy distribution, this poses a sampling problem which is known to be able to influence the tumor prognostication. On the other hand, automatic detection of mitotic figures can't yet be considered reliable enough for clinical application. In order to aid the work of the human expert and at the same time reduce variance in tumor grading, it is beneficial to assess the whole slide image (WSI) for the highest mitotic activity and use this as a reference region for human counting. For this task, we compare two methods for region of interest proposal, both based on convolutional neural networks (CNN). For both approaches, the CNN performs a segmentation of the WSI to assess mitotic activity. The first method performs a segmentation at the original image resolution, while the second approach performs a segmentation operation at a significantly reduced resolution, cutting down on processing complexity. We evaluate the approach using a dataset of 32 completely annotated whole slide images of canine mast cell tumors, where 22 were used for training of the network and 10 for test. Our results indicate that, while the overall correlation to the ground truth mitotic activity is considerably higher (0.94 vs. 0.83) for the approach based upon the fine resolution network, the field of interest choices are only marginally better. Both approaches propose fields of interest that contain a mitotic count in the upper quartile of respective slides.