Back to Projects List
An ongoing effort in the WPI AIM Lab is to use our MRI-compatible 6-DOF needle insertion robot to perform closed-loop image-guided insertions using bevel-tipped clinical biopsy needles. Our focus has been on compensating for target shift and unmodelled needle tip deflection during nominally straight insertions, since the stiffness of clinical-style needles precludes drastic changes in heading during insertions.
A key part of this project is needle tip localization. I am developing a Slicer module that uses actively-updated MR images to detect the pose of a moving needle tip and determine the next best scan plane to capture the tip. The measured tip pose will be an input to the robot control algorithm, which will assess the error relative to the desired trajectory and the remaining insertion depth to the target and adjust the insertion velocity and needle rotation speed accordingly.
See Closed-loop Autonomous Needle Steering during Cooperatively Controlled Needle Insertionsfor MRI-guided Pelvic Interventions for a similar approach to this problem, albeit using cameras and a transparent phantom.
 R. J. Webster, J. S. Kim, N. J. Cowan, G. S. Chirikjian, and A. M. Okamura, “Nonholonomic Modeling of Needle Steering,” The International Journal of Robotics Research. 2006 May;25(5-6):509-25.
 J. P. Swensen, M. Lin, A. M. Okamura, and N. J. Cowan, “Torsional Dynamics of Steerable Needles: Modeling and Fluoroscopic Guidance,” IEEE Trans Biomed Eng. 2014 Nov;61(11):2707-17.
 F. Zijlstra, J. G. Bouwman, I. Braškutė, M. A. Viergever, and P. R. Seevinck, “Fast Fourier-based simulation of off-resonance artifacts in steady-state gradient echo MRI applied to metal object localization,” Magn Reson Med. 2017 Nov;78(5):2035-41.
 A. Mastmeyer, G. Pernelle, R. Ma, L. Barber, and T. Kapur, “Accurate Model-based Segmentation of Gynecologic Brachytherapy Catheter Collections in MRI-images,” Med Image Anal. 2017 Dec;42:173-88