Leveraging voxel-wise segmentation uncertainty to improve reliability in assessment of paediatric dysplasia of the hip.

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

Abstract

PURPOSE: Estimating uncertainty in predictions made by neural networks is critically important for increasing the trust medical experts have in automatic data analysis results. In segmentation tasks, quantifying levels of confidence can provide meaningful additional information to aid clinical decision making. In recent work, we proposed an interpretable uncertainty measure to aid clinicians in assessing the reliability of developmental dysplasia of the hip metrics measured from 3D ultrasound screening scans, as well as that of the US scan itself. In this work, we propose a technique to quantify confidence in the associated segmentation process that incorporates voxel-wise uncertainty into the binary loss function used in the training regime, which encourages the network to concentrate its training effort on its least certain predictions.

Authors

  • Arunkumar Kannan
    Department of Electrical and Computer Engineering, University of British Columbia, Vancouver, Canada. arunk@ece.ubc.ca.
  • Antony Hodgson
    Department of Mechanical Engineering, University of British Columbia, Vancouver, Canada.
  • Kishore Mulpuri
    Department of Orthopedic Surgery, British Columbia Children's Hospital, Vancouver, Canada.
  • Rafeef Garbi
    Department of Electrical and Computer Engineering, University of British Columbia, Vancouver, Canada.