Research Papers

A Passive Parallel Master–Slave Mechanism for Magnetic Resonance Imaging-Guided Interventions

[+] Author and Article Information
Santhi Elayaperumal

Department of Mechanical Engineering,
Stanford University,
Stanford, CA 94305
e-mail: santhie@stanford.edu

Mark R. Cutkosky

Fellow ASME
Department of Mechanical Engineering,
Stanford University,
Stanford, CA 94305
e-mail: cutkosky@stanford.edu

Pierre Renaud

CNRS-INSA-Strasbourg University,
Strasbourg 67000, France
e-mail: pierre.renaud@insa-strasbourg.fr

Bruce L. Daniel

Department of Radiology,
Stanford University,
Stanford, CA 94305
e-mail: bdaniel@stanford.edu

KMS Bearings, Inc., Anaheim, CA.

Xi’an Hua Ying Te Bearing Co., Ltd., Shaanxi Province, China.

THK Co. Ltd., Tokyo, Japan.

GE Healthcare, Waukesha, WI.

ATI Industrial Automation, Apex, NC.

M4-50, Mark-10 Corp., Copiague, NY.

Manuscript received December 23, 2013; final manuscript received October 12, 2014; published online November 26, 2014. Assoc. Editor: Carl Nelson.

J. Med. Devices 9(1), 011008 (Mar 01, 2015) (11 pages) Paper No: MED-13-1302; doi: 10.1115/1.4028944 History: Received December 23, 2013; Revised October 12, 2014; Online November 26, 2014

A passive, parallel master–slave mechanism is presented for magnetic resonance imaging (MRI)-guided interventions in the pelvis. The mechanism allows a physician to stand outside the MRI scanner while manipulating a needle inside the bore and, unlike a powered robot, does not place actuators in proximity to the patient. The manipulator combines two parallel mechanisms based on the Delta robot architecture. The mechanism also includes a two-axis gimbal to allow for tool angulation, giving a total of five degrees of freedom so that the physician can insert and steer a needle using continuous natural arm and wrist movements, unlike simple needle guides. The need for access between the patient’s legs and within the MRI scanner leads to an unusual asymmetric design in which the sliding prismatic joints form the vertices of an isosceles triangle. Kinematic analysis shows that the dexterity index of this design is improved over the desired workspace, as compared to an equilateral design. The analysis is extended to estimate the effect of friction and model the input:output force transmission. Prototypes, with final dimensions selected for transperineal prostate interventions, showed force transmission behavior as predicted by simulation, and easily withstood maximum forces required for tool insertion.

Copyright © 2015 by ASME
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Fig. 1

Illustration of the double Delta-based (three P-U-U) manipulator with two-axis gimbals (+2 R). Design parameters such as the angle between the prismatic axes and strut lengths can be changed to improve manipulability in the workspace of interest.

Grahic Jump Location
Fig. 2

Schematic of half the double parallel mechanism, with joint and end-effector positions labeled. For clarity of the illustration, only some universal joint axes are indicated. Gimbal not shown.

Grahic Jump Location
Fig. 3

End-view diagrams of the manipulator to illustrate dimensions of interest. Note: LSj and LPj (j = 1, 2) lie in the XY plane because the platform does not tilt or twist; and the strut lengths are not purely in the XY plane. S1, S2, and S3 represent the slider locations, and P1, P2, and P3 represent the corners of the platform. In the equilateral case, r intersects the origin at the triangle centroid, in the isosceles case, r intersects the circumcenter.

Grahic Jump Location
Fig. 4

Manipulability polyhedra as viewed in the XY plane and isometrically for manipulators with the following parameters: (a) a = b = 60 deg, LS2 = 0.178 m, LT1 = LT2 = 0.152 m, and LP2/LS2=0.36. (b) a = 45, b = 90 deg, LS2 = 0.171 m, LT1 = LT2 = 0.135 m, and LP2/LS2=0.44.

Grahic Jump Location
Fig. 5

Isotropy measure throughout the workspace ±4 cm from the origin for an equilateral tip-up and isosceles tip-down design as shown in the photos. Values close to 1 imply high isotropy. RIM calculated is for w ± 4 cm and Nw = 900. (Top) Equilateral parameters: a = b = 60 deg, LS2 = 0.191 m, LT1 = LT2 = 0.152 m, and LP2/LS2=0.40. (Bottom) Isosceles parameters (tip-down): a = 45 deg, b = 90 deg, LS2 = 0.203 m, LT1 = 0.117 m, LT2 = 0.165 m, and LP2/LS2=0.4.

Grahic Jump Location
Fig. 6

Regional dexterity index for varying single design parameters, over a workspace ±4 cm from the origin. Values are normalized with respect to frame radius r. Initial design: a = 70 deg, b = 40 deg, LS2 = 0.19 m, LT2 = 0.19 m, and LP2/LS2=0.50. LT1 was changed according to Eq. (14).

Grahic Jump Location
Fig. 7

Simulation of forces at the slave for an input force of F = (1 N, 1 N, 1 N) at the master, with coefficient of friction μ = 0.05. Manipulator parameters: (top) tip-up equilateral − a = b = 60 deg, LS2 = 0.191 m, LT1 = LT2 = 0.152 m, and LP2/LS2=0.40; (bottom) tip-down isosceles −a = 45 deg, b = 90 deg, LS2 = 0.171 m, LT1 = LT2 = 0.135 m, and LP2/LS2=0.44.

Grahic Jump Location
Fig. 8

(a) Prototype on a scanner table with a model patient and torso coil. (b) End-view from patient’s feet, with patient and manipulator in procedure position for transperineal access. (c) Axial localizer image of the pelvis, with prostate outlined. (d) Fast spin echo image (ETL 14, TE 96.9, TR 4205) of the prostate of a canine taken from an animal study using the manipulator to guide a needle to targets in the prostate. Workspace of interest in (a) and (b) outlined as a square in the XY plane.

Grahic Jump Location
Fig. 9

(a) Oblique view of the passive masterslave manipulator, annotated with its 3DOF in translation. (b) The mechanism includes a set of gimbals for pitch and yaw rotations. The manipulator thus has five total degrees of freedom, and is shown in different articulations, with the platform (c) up and (d) lower down.

Grahic Jump Location
Fig. 10

(a) Benchtop test showing the manipulator mounted vertically to compare force transmission for input load Fz. (b) Predicted force variations for Fz = 4.9 N and μ = 0.05 over the workspace of interest.

Grahic Jump Location
Fig. 11

Benchtop maximum loading test apparatus showing prototype 1 with (a) digital force scale at slave end and (b) six-axis sensor at master end to test force capability. Configuration shown is to test Fz loading.




Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In