An interpretable automated detection system for FISH-based HER2 oncogene amplification testing in histo-pathological routine images of breast and gastric cancer diagnostics

Sarah Schmell, Falk Zakrzewski, Walter de Back, Martin Weigert, Uwe Schmidt, Torsten Wenke, Silke Zeugner, Robert Mantey, Christian Sperling, Ingo Roeder, Pia Hoenscheid, Daniela Aust, Gustavo Baretton

Track: short paper

TL;DR: We developed a comprehensible deep learning-based pipeline which automates the evaluation of FISH images with respect to HER2 gene amplification testing for breast and gastric cancer to support pathologists in their everyday work.

Keywords: FISH imaging, HER2 amplification status, gastric cancer, breast cancer, digital pathology, deep learning, image classification, object segmentation and localization, interpretability

Abstract: Histo-pathological diagnostics are an inherent part of the everyday work but are particularly laborious and often associated with time-consuming manual analysis of image data. In order to cope with the increasing diagnostic case numbers due to the current growth and demographic change of the global population and the progress in personalized medicine, pathologists ask for assistance. Profiting from digital pathology and the use of artificial intelligence, individual solutions can be offered (e.g. to detect labeled cancer tissue sections). The testing of the human epidermal growth factor receptor 2 (HER2) oncogene amplification status via fluorescence in situ hybridization (FISH) is recommended for breast and gastric cancer diagnostics and is regularly performed at clinics. Here, we developed a comprehensible, multi-step deep learning-based pipeline which automates the evaluation of FISH images with respect to HER2 gene amplification testing. It mimics the pathological assessment and relies on the detection and localization of interphase nuclei based on instance segmentation networks. Furthermore, it localizes and classifies fluorescence signals within each nucleus with the help of image classification and object detection convolutional neural networks (CNNs). Finally, the pipeline classifies the whole image regarding its HER2 amplification status. The visualization of pixels on which the networks' decision occurs, complements an essential part to enable interpretability by pathologists.

Paper Type: methodological development

Presentation Upload Agreement: I agree that my presentation material (videos and slides) will be made public.

Community Implementations: [ 3 code implementations](https://www.catalyzex.com/paper/an-interpretable-automated-detection-system/code)