Mathematical Modeling and Machine Learning for Force Estimation on a Planar Catheter

  • Author(s) / Creator(s)
  • In this paper, estimation of the applied force on a planar catheter is considered. An image processing approach has been chosen as the most suitable tool. For this purpose, we create a comprehensive database of catheters with different shapes and operating forces. Using image processing algorithms, position data for the catheters' outer, middle, and inner layers in this database are obtained. Finally, using curve fitting, an exponential mathematical function is accepted for the middle layer of each catheter. Feeding this data into different machine learning algorithms, force estimation is obtained with a mean average error of 0.52 N. Later, the force applied to a tendon catheter is estimated using the SOFA framework and deep learning techniques, directing the research into a reliable applied force estimation.

    Part of the Proceedings of the Canadian Society for Mechanical Engineering International Congress 2022.

  • Date created
  • Subjects / Keywords
  • Type of Item
    Article (Published)
  • DOI
  • License
    Attribution-NonCommercial 4.0 International