Corpus ID: 222341358

A Pathologist-Annotated Dataset for Validating Artificial Intelligence: A Project Description and Pilot Study.

  title={A Pathologist-Annotated Dataset for Validating Artificial Intelligence: A Project Description and Pilot Study.},
  author={Sarah N. Dudgeon and Si Wen and M. Hanna and Rajarsi Gupta and Mohamed Amgad and Manasi Sheth and H. Marble and R. Huang and M. Herrmann and C. Szu and Darick Tong and Bruce Werness and Evan Szu and Denis Larsimont and A. Madabhushi and E. Hytopoulos and W. Chen and R. Singh and S. Hart and J. Saltz and R. Salgado and Brandon D Gallas United States Food and Drug Administration and Center for Devices and Radiological Health and Office of Science and Engineering Laboratories and Division of Imaging DiagnosticsSoftware Reliability and White Oak and Md. and Memorial Sloan Kettering Cancer Center and N. York. and Ny and Stony Brook Medicine Dept of Biomedical Informatics and S. Brook and Department of Pathology and N. University and Rubloff Building and C. Illinois and U. S. Food and Office of Product Quality and Evaluation and Office of Clinical Evidence and Analysis and Division of Biostatistics and Massachusetts General HospitalHarvard Medical School and Boston and Ma. and Computational Pathology and Margaret Hospital and Harvard Medical School and Arrive Origin and S. Francisco and Ça and Institut Jules Bordet and Brussels and Belgium. and Case Western Reserve University and Cleveland and Oh and iRhythm Technologies Inc. and Northwell health and Zucker School of Medicine and Department of Health Sciences Research and Mayo Clinic and M. Rochester and Division of Research and Peter Mac Callum Cancer Centre and Melbourne. and Australia. and GZA-ZNA Hospitals and Antwerp},
  journal={arXiv: Quantitative Methods},
  • Sarah N. Dudgeon, Si Wen, +69 authors Antwerp
  • Published 2020
  • Computer Science, Biology, Engineering
  • arXiv: Quantitative Methods
  • Purpose: In this work, we present a collaboration to create a validation dataset of pathologist annotations for algorithms that process whole slide images (WSIs). We focus on data collection and evaluation of algorithm performance in the context of estimating the density of stromal tumor infiltrating lymphocytes (sTILs) in breast cancer. Methods: We digitized 64 glass slides of hematoxylin- and eosin-stained ductal carcinoma core biopsies prepared at a single clinical site. We created training… CONTINUE READING
    1 Citations

    Figures and Tables from this paper


    Structured crowdsourcing enables convolutional segmentation of histology images
    • 17
    • PDF
    Report on computational assessment of Tumor Infiltrating Lymphocytes from the International Immuno-Oncology Biomarker Working Group
    • 10
    • PDF
    1399 H&E-stained sentinel lymph node sections of breast cancer patients: the CAMELYON dataset
    • 70
    • PDF
    Evaluation environment for digital and analog pathology: a platform for validation studies
    • 10
    • Highly Influential
    • PDF
    Validation of mitotic cell quantification via microscopy and multiple whole-slide scanners
    • 4
    Computational Pathology: Challenges and Promises for Tissue Analysis
    • 163
    • PDF
    A Containerized Software System for Generation, Management, and Exploration of Features from Whole Slide Tissue Images.
    • 22
    • PDF