Utilizing Automated Breast Cancer Detection to Identify Spatial Distributions of Tumor Infiltrating Lymphocytes in Invasive Breast Cancer

• Type of Material: Article
• Authors: Han Le, Stony Brook University; Rajarsi Gupta, Stony Brook University; Le Hou, Stony Brook University; Shahira Abousamra, Stony Brook University; Danielle Fassler, Stony Brook University; Tahsin Kurc, Stony Brook University; Dimitris Samaras, Stony Brook University; Rebecca Batiste, Stony Brook University; Tianhao Zhao, Stony Brook University; Alison L. Van Dyke, National Institutes of Health, Bethesda; Ashish Sharma, Emory University; Erich Bremer, Stony Brook University; Jonas S. Almeida, National Cancer Institute Division of Cancer Epidemiology & Genetics; Joel Saltz, Stony Brook University
• Language: English
• Date: 2019
• Publisher: Cornell University
• Publication Version: Final Published Version
• Copyright Statement: ©2019 Cornell University
• License: Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International
• Final Published Version (URL): https://arxiv.org/abs/1905.10841
• Title of Journal or Parent Work: arXiv preprint arXiv:1905.10841
• Grant/Funding Information: NCI Surveillance Research Program overseeing the Virtual Tissue Repository (VTR) Pilot Program, from which participating SEER cancer registries (Greater California, Connecticut, Hawaii, Iowa, Kentucky, and Louisiana) supplied the whole slide images utilized in algorithm development and testing.; This work used the Extreme Science and Engineering Discovery Environment (XSEDE), which is supported by National Science Foundation grant number ACI-1548562.; The SEER VTR Pilot Program is supported by the Division of Cancer Control and Population Sciences at the National Cancer Institute of the National Institutes of Health.; Specifically, it used the Bridges system, which is supported by NSF award number ACI-1445606, at the Pittsburgh Supercomputing Center (PSC).; This work was supported in part by 1U24CA180924-01A1, 3U24CA215109-02, and 1UG3CA225021-01 from the National Cancer Institute, R01LM011119-01 and R01LM009239 from the U.S. National Library of Medicine.
• Keywords: eess.IV; cs.CV

Abstract

Quantitative assessment of Tumor-TIL spatial relationships is increasingly important in both basic science and clinical aspects of breast cancer research. We have developed and evaluated convolutional neural network (CNN) analysis pipelines to generate combined maps of cancer regions and tumor infiltrating lymphocytes (TILs) in routine diagnostic breast cancer whole slide tissue images (WSIs). We produce interactive whole slide maps that provide 1) insight about the structural patterns and spatial distribution of lymphocytic infiltrates and 2) facilitate improved quantification of TILs. We evaluated both tumor and TIL analyses using three CNN networks - Resnet-34, VGG16 and Inception v4, and demonstrated that the results compared favorably to those obtained by what believe are the best published methods. We have produced open-source tools and generated a public dataset consisting of tumor/TIL maps for 1,015 TCGA breast cancer images. We also present a customized web-based interface that enables easy visualization and interactive exploration of high-resolution combined Tumor-TIL maps for 1,015TCGA invasive breast cancer cases that can be downloaded for further downstream analyses.