Can uncertainty estimation predict segmentation performance in ultrasound bone imaging?

Journal: International journal of computer assisted radiology and surgery
Published Date:

Abstract

PURPOSE: Segmenting bone surfaces in ultrasound (US) is a fundamental step in US-based computer-assisted orthopaedic surgeries. Neural network-based segmentation techniques are a natural choice for this, given promising results in related tasks. However, to gain widespread use, we must be able to know how much to trust segmentation networks during clinical deployment when ground-truth data is unavailable.

Authors

  • Prashant U Pandey
    School of Biomedical Engineering, University of British Columbia, Vancouver, Canada. prashant@ece.ubc.ca.
  • Pierre Guy
    Department of Orthopaedics, Faculty of Medicine, University of British Columbia, Vancouver, Canada.
  • Antony J Hodgson
    Department of Mechanical Engineering, School of Biomedical Engineering, Surgical Technologies Lab, University of British Columbia, Vancouver, British Columbia, Canada.