• CN: 11-2187/TH
  • ISSN: 0577-6686

Journal of Mechanical Engineering ›› 2021, Vol. 57 ›› Issue (11): 52-60.doi: 10.3901/JME.2021.11.052

Previous Articles     Next Articles

Trajectory Prediction of Bevel-Tip Flexible Needle Based on Force and Vision Perception

WANG Linze1, GAO Dedong1, BAI Huiquan2, ZHAO Yan1, CUI Jiali3, LI Murong4, LEI Yong4   

  1. 1. School of Mechanical Engineering, Qinghai University, Xining 810016;
    2. Xining First People's Hospital, Xining 810016;
    3. Department of Computer Technology and Application, Qinghai University, Xining 810016;
    4. State Key Laboratory of Fluid Power & Mechatronic Systems, Zhejiang University, Hangzhou 310027
  • Received:2020-11-10 Revised:2021-02-09 Online:2021-06-05 Published:2021-07-23

Abstract: It is a challenge to precisely control the flexible needle to reach the target in clinical applications. The force acting on the flexible needle can cause tissue deformation and needle bending, resulting in misalignment of the needle tip. The interaction between the needle and soft tissue involves a large number of biophysical characteristics, and these parameters cannot be estimated through physical modeling or directly. In order to solve this problem, a method to predict the needle trajectory is proposed. The force analysis for the flexible needle is carried out and the corresponding mechanical model is established; on the basis of the mechanical model, a force-vision perception prediction model based on BP neural network is established and needle tip trajectory is predicted. Three different types of flexible needles are tested, data are collected to train the model. Finally, the needle tip trajectory was obtained through experiments, which are compared with the model prediction. The results show that the displacements in x and y directions predicted by the model can accord with the experiments and model error is within 2mm, which can predict the insertion trajectory more accurately.

Key words: needle insertion, BP neural network, soft tissue, image processing

CLC Number: