eduzhai > Applied Sciences > Engineering >

Hierarchical Deep Learning Ensemble to Automate the Classification of Breast Cancer Pathology Reports by ICD-O Topography

  • king
  • (0) Download
  • 20210507
  • Save

... pages left unread,continue reading

Document pages: 4 pages

Abstract: Like most global cancer registries, the National Cancer Registry in SouthAfrica employs expert human coders to label pathology reports using appropriateInternational Classification of Disease for Oncology (ICD-O) codes spanning 42different cancer types. The annotation is extensive for the large volume ofcancer pathology reports the registry receives annually from public and privatesector institutions. This manual process, coupled with other challenges resultsin a significant 4-year lag in reporting of annual cancer statistics in SouthAfrica. We present a hierarchical deep learning ensemble method incorporatingstate of the art convolutional neural network models for the automaticlabelling of 2201 de-identified, free text pathology reports, with appropriateICD-O breast cancer topography codes across 8 classes. Our results show animprovement in primary site classification over the state of the art CNN modelby greater than 14 for F1 micro and 55 for F1 macro scores. We demonstratethat the hierarchical deep learning ensemble improves on state-of-the-artmodels for ICD-O topography classification in comparison to a flat multiclassmodel for predicting ICD-O topography codes for pathology reports.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×