Abstract
We introduce a novel, large-scale dataset for microscopy cell annotations. The dataset includes 32 whole slide images (WSI) of canine cutaneous mast cell tumors, selected to include both low grade cases as well as high grade cases. The slides have been completely annotated for mitotic figures and we provide secondary annotations for neoplastic mast cells, inflammatory granulocytes, and mitotic figure look-alikes. Additionally to a blinded two-expert manual annotation with consensus, we provide an algorithm-aided dataset, where potentially missed mitotic figures were detected by a deep neural network and subsequently assessed by two human experts. We included 262,481 annotations in total, out of which 44,880 represent mitotic figures. For algorithmic validation, we used a customized RetinaNet approach, followed by a cell classification network. We find F1-Scores of 0.786 and 0.820 for the manually labelled and the algorithm-aided dataset, respectively. The dataset provides, for the first time, WSIs completely annotated for mitotic figures and thus enables assessment of mitosis detection algorithms on complete WSIs as well as region of interest detection algorithms.
Highlights
Background & SummaryMicroscopy image recognition has seen vast advances in recent years, fostered by the availability of high quality datasets as well as by the application of sophisticated deep learning pipelines
In the field of automatic detection of those mitotic figures, there have been a number of competitions in recent years, e.g. the TUPAC16 challenge[6], the ICPR MITOS-20127 and ICPR MITOS-ATYPIA-2014 challenge[8]
Identification of individual mitotic figures has only moderate agreement between trained pathologists as they include a wide range of morphological variants depending on the phase of cell division and tissue properties as well as atypical morphologies
Summary
Microscopy image recognition has seen vast advances in recent years, fostered by the availability of high quality datasets as well as by the application of sophisticated deep learning pipelines. One of the most important topics in the field of microscopy imaging is the classification of cells, typically stained with hematoxylin and eosin (H&E) dye In this area, one challenging task is the detection of mitotic figures, i.e. cells undergoing division, in tumor tissue. Given the importance of quantifying mitotic figures in various tumor types of animals and humans, it is at first glance surprising that none of the available datasets provide labels for complete WSI Manual annotation of such large areas, is a labor-intense and tedious task. Two experts [CB, RK] classified the annotations in a blinded manner and reviewed the disagreed labels to find common consensus on the label class This collection[19] represents the currently largest data set in number of annotated mitotic figures and annotated tumor area. It provides researchers with new opportunities for the development and refinement of data-driven algorithms for mitotic figure identification on entire whole slide images
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
