Image quality assessment for machine learning tasks using meta-reinforcement learning.

Journal: Medical image analysis
Published Date: Mar 21, 2022

Abstract

In this paper, we consider image quality assessment (IQA) as a measure of how images are amenable with respect to a given downstream task, or task amenability. When the task is performed using machine learning algorithms, such as a neural-network-based task predictor for image classification or segmentation, the performance of the task predictor provides an objective estimate of task amenability. In this work, we use an IQA controller to predict the task amenability which, itself being parameterised by neural networks, can be trained simultaneously with the task predictor. We further develop a meta-reinforcement learning framework to improve the adaptability for both IQA controllers and task predictors, such that they can be fine-tuned efficiently on new datasets or meta-tasks. We demonstrate the efficacy of the proposed task-specific, adaptable IQA approach, using two clinical applications for ultrasound-guided prostate intervention and pneumonia detection on X-ray images.

Authors

  • Shaheer U Saeed
    Centre for Medical Image Computing, Wellcome/EPSRC Centre for Interventional & Surgical Sciences, and Department of Medical Physics & Biomedical Engineering, University College London, London, UK. Electronic address: shaheer.saeed.17@ucl.ac.uk.
  • Yunguan Fu
    Centre for Medical Image Computing, Wellcome/EPSRC Centre for Interventional & Surgical Sciences, and Department of Medical Physics & Biomedical Engineering, University College London, London, UK; InstaDeep, London, UK.
  • Vasilis Stavrinides
    Division of Surgery & Interventional Science, University College London, London, UK; Department of Urology, University College Hospital NHS Foundation Trust, London, UK.
  • Zachary M C Baum
    Centre for Medical Image Computing, Wellcome/EPSRC Centre for Interventional & Surgical Sciences, and Department of Medical Physics & Biomedical Engineering, University College London, London, UK.
  • Qianye Yang
    Centre for Medical Image Computing, Wellcome/EPSRC Centre for Interventional & Surgical Sciences, and Department of Medical Physics & Biomedical Engineering, University College London, London, UK.
  • Mirabela Rusu
    Department of Radiology, Stanford University, Stanford, CA 94305, USA. Electronic address: mirabela.rusu@stanford.edu.
  • Richard E Fan
    Department of Urology, Stanford University, Stanford, CA 94305, USA.
  • Geoffrey A Sonn
    Department of Radiology, Stanford University School of Medicine, Stanford, CA, USA; Department of Urology, Stanford University School of Medicine, Stanford, CA, USA.
  • J Alison Noble
    Institute of Biomedical Engineering (IBME), University of Oxford, Oxford, England, UK.
  • Dean C Barratt
    Wellcome / EPSRC Centre for Interventional and Surgical Sciences (WEISS), University College London, UK; Centre for Medical Image Computing (CMIC), Departments of Medical Physics & Biomedical Engineering and Computer Science, University College London, UK.
  • Yipeng Hu
    Centre for Medical Image Computing (CMIC), Departments of Medical Physics & Biomedical Engineering and Computer Science, University College London, UK.

Keywords

External Resources

Popular Topics
Recent Journals