A convolutional neural network-based method for the generation of super-resolution 3D models from clinical CT images.

Journal: Computer methods and programs in biomedicine
PMID:

Abstract

BACKGROUND AND OBJECTIVE: The accurate evaluation of bone mechanical properties is essential for predicting fracture risk based on clinical computed tomography (CT) images. However, blurring and noise in clinical CT images can compromise the accuracy of these predictions, leading to incorrect diagnoses. Although previous studies have explored enhancing trabecular bone CT images to super-resolution (SR), none of these studies have examined the possibility of using clinical CT images from different instruments, typically of lower resolution, as a basis for analysis. Additionally, previous studies rely on 2D SR images, which may not be sufficient for accurate mechanical property evaluation, due to the complex nature of the 3D trabecular bone structures. The objective of this study was to address these limitations.

Authors

  • Yijun Zhou
    Division of Biomedical Engineering, Department of Materials Science and Engineering, Ångströmlaboratoriet, Uppsala University, Lägerhyddsvägen 1, Uppsala 75237, Sweden.
  • Eva Klintström
    Center for Medical Image Science and Visualization (CMIV), Linköping University, Sweden; Department of Radiology and Department of Health, Medicine and Caring Sciences, Linköping University, Sweden.
  • Benjamin Klintström
    Department of Biomedical Engineering and Health Systems, KTH Royal Institute of Technology, Huddinge, Sweden.
  • Stephen J Ferguson
    Institute for Biomechanics, ETH Zürich, Zürich, Switzerland.
  • Benedikt Helgason
    Institute for Biomechanics, ETH Zürich, Zürich, Switzerland.
  • Cecilia Persson
    Division of Biomedical Engineering, Department of Materials Science and Engineering, Ångströmlaboratoriet, Uppsala University, Lägerhyddsvägen 1, Uppsala 75237, Sweden. Electronic address: Cecilia.Persson@angstrom.uu.se.