Steerable Needles


Needles are one of the most useful medical devices. They can be used to make diagnoses (e.g. biopsy) as well as deliver a wide range of treatments including liquid injections, thermal treatments, or targeted doses of chemo or radiotherapy. The effectiveness of all such diagnoses and treatments is strongly correlated with the accuracy with which the needle tip is placed at the desired target.

A bevel-steered needle traversing a curved path in phantom tissue.


Nearly all surgical needles are inserted manually using a procedure that relies completely on hand-eye coordination and is highly dependent on the physician's skill and experience. This procedure is described on our image-guided needle placement page. Image-guided robotic systems have the potential to improve the accuracy of needle tip placement by accurately aligning a needle before insertion begins. 

However, even perfect pre-entry alignment cannot guarantee accurate tip placement. Needles can still miss their targets due to deflection at membranes, tissue deformation and inhomogeneity, registration and calibration tolerances, etc. The only way to compensate for these sources of error is to use steerable needles capable of controllable trajectories within tissue. Another compelling motivation for steerable needles is the potential to maneuver around sensitive structures to reach locations behind them, where straight trajectories may not be feasible or desirable.

A variety of needle steering methods have been invented within the past few years, and in the MED lab we are currently exploring several of them. One method that is particularly appealing because of its simplicity is the bevel steering technique. Bevel steering harnesses the forces generated by the standard wedge-like bevel tip (shown in the inset line drawing on the figure above). The bevel tip is a common feature of most surgical needles because its single grind provides a cost effective means of creating a sharp needle.

Normally, the bevel-induced bending effect is viewed as an unwanted source of error in needle placement.  However, we harness the effect and use it to our advantage, amplifying it by employing a thin, flexible needle shaft. By controlling axial rotation ("aiming" the bevel) during insertion, it is possible to control trajectory of the needle. We have designed and constructed a robot for controlling bevel-based steering.  Currently, we are developing control algorithms to guide the needle along desired trajectories under image guidance.
 
  
            
(Left) Steerable needle experimental setup (Right) Robotic needle driver.


Our needle steering work in the MED lab is an outgrowth of Dr. Webster's prior work at Johns Hopkins. There, with Noah Cowan, Gregory Chirikjian, and Allison Okamura he developed a kinematic model that describes tip pose as a function of the translation and axial rotation velocities of the needle. This model is a starting point for model-based control techniques being developed at Vanderbilt and also JHU (link). Collaboration with Ron Alterovitz to plan paths for the needle and model tissue deformation as a steerable needle enters tissue (link) is ongoing. We have also investigated teleoperation strategies for steerable needles, which enable the human to assume a portion of the planning and control tasks, and in so doing provides a rapid path to clinical adoption of needle steering. However, a fully automated solution is eventually expected to be more accurate. Thus, we plan in the near future to integrate modeling, planning, and control to automatically steer needles along desired trajectories using real-time image feedback.

Videos (a bit dated, but still useful):

  1. Low Res 5Mb mov

  2. High Res 10Mb divX 

Patents:

  1. R. J. Webster III, A. M. Okamura, N. J. Cowan, G. S. Chirikjian, K. Y. Goldberg, and R. Alterovitz.  Distal Bevel-Tip Needle Control Device and Algorithm.  US Patent Pending.

Journal Publications:

  1. R. J. Webster III, J. S. Kim, N. J. Cowan, G. S. Chirikjian, and A. M. Okamura. Nonholonomic Modeling of Needle Steering,  International Journal of Robotics Research, 25(5–6), 509-525, 2006.

Book Chapters:

  1. R. J. Webster III, N. J. Cowan, G. S. Chirikjian, and A. M. Okamura.  Nonholonomic Modeling of Needle Steering.  9th International Symposium on Experimental Robotics 2004, Springer Tracts in Advanced Robotics, 21, 35-44, 2006.

Conference Publications:

  1. J. M. Romano, R. J. Webster III, and A. M. Okamura.  Teleoperation of Steerable Needles.  IEEE International Conference on Robotics and Automation, 934-939, 2007.

  2. R. J. Webster III, J. Memisevic, and A. M. Okamura.  Design Considerations for Robotic Needle Steering.  IEEE International Conference on Robotics and Automation, 3599-3605, 2005.