A BERT model generates diagnostically relevant semantic embeddings from pathology synopses with active learning.

Journal: Communications medicine
Published Date: Jul 5, 2021

Abstract

BACKGROUND: Pathology synopses consist of semi-structured or unstructured text summarizing visual information by observing human tissue. Experts write and interpret these synopses with high domain-specific knowledge to extract tissue semantics and formulate a diagnosis in the context of ancillary testing and clinical information. The limited number of specialists available to interpret pathology synopses restricts the utility of the inherent information. Deep learning offers a tool for information extraction and automatic feature generation from complex datasets.

Authors

  • Youqing Mu
    McMaster University, Hamilton, ON Canada.
  • Hamid R Tizhoosh
    Kimia Lab, University of Waterloo, Waterloo, ON Canada.
  • Rohollah Moosavi Tayebi
    McMaster University, Hamilton, ON Canada.
  • Catherine Ross
    McMaster University, Hamilton, ON Canada.
  • Monalisa Sur
    McMaster University, Hamilton, ON Canada.
  • Brian Leber
    McMaster University, Hamilton, ON Canada.
  • Clinton J V Campbell
    McMaster University, Hamilton, ON Canada.

Keywords

No keywords available for this article.

External Resources

Popular Topics
Recent Journals