Spring 2014 MAE 156B Project #6

MRI Positioning System

BACKGROUND

The Center for Functional Magnetic Resonance Imaging (CFMRI) houses three imaging systems: two 3 tesla short bore scanners--GE Discovery MR750--for human studies; and one 7 tesla scanner--Bruker Biospec 70/20 USR--for small animal imaging. MRI scanners produce a rapidly changing magnetic field which aligns the hydrogen nuclei within the sample body. These hydrogen nuclei absorb energy from an applied radio frequency (RF) pulse, which then creates a brief and faint electromagnetic field signals that are detected by the RF coils in the MRI system. Samples are positioned at the bore’s isocenter, the optimal position for producing high-resolution images. Our sponsor, Robert Bussell, is an associate development engineer at the CFMRI specialized in the 7 Tesla scanner, Bruker Biospec, used for animal and tissue imaging. The current positioning system used at the center comprises of tooling manually positioned along a dovetail shaped guide. Our goal is to design an automated positioning system with high accuracy and repeatability to replace their existing setup.


OBJECTIVE

Functional Requirements
  • Linear position system with a tolerance of 0.5mm
  • Ability to transport a sample mass of up to 2kg
  • Little to no noise or interference with MRI scanning capabilities
  • Accurate repeatability
  • Easy-to-use graphical user interface
  • Emergency stop switch
  • As few magnetic components in the system as possible
Deliverables
  • Belt-driven linear motion system with a tolerance of 0.2mm
  • LabVIEW-based user interface communicating with an Arduino controller
  • Design and maintenance manual, including parts list

FINAL DESIGN

There are three main components to the overall positioning system: the linear slider, fiber optics communication, and the LabVIEW-based user interface.

Linear Slider


The belt-driven linear slider is comprised of a NEMA 23 stepper motor, a 1/2" wide timing belt, two timing pulleys, and the carriage itself. The motor was chosen based upon its high torque capabilities, and its ability to achieve high resolution microstepping when used with a stepper motor driver. Using timing pulleys with a 1.432" pitch diameter, we were able to achieve a step size of approximately 0.2 mm.

The carriage itself rolls on nylon wheels housing glass ball bearings. Tooling that the center uses to hold research samples will be mounted to the top plate, which has been drilled with a generic hole pattern.

Fiber Optic Communication
When running copper wires between the control room to the scanning room, there is a risk of picking up noise from the control room and channeling them into the scanning room and potentially distorting the MR image. Fiber optic cables avoid this problem by converting signals from the Arduino into high/low signals that can be interpreted and used to control the stepper motor driver.

LabVIEW-based Graphical User Interface

The graphical user interface has an intuitive design to allow the end user to control the position of the mechanical system with little instruction. Based in LabVIEW, the program gives developer flexibility for modifying and updating the interface as needs change. Researchers can easily position their samples using the "Home" and "Isocenter" buttons while being given the opti
on to incrementally adjust to the optimal position and save the information for repeated trials.

Final Test with MRI

MRI Positioning System Demonstration




Results

 Requirements   Deliverables
  • Resolution of < 0.5 mm
  • Precision within 0.5 mm
  • Accuracy within 0.5
  • Few magnetic compoents
  • Resolution approximately 0.057 mm
  • Precision within +/- 0.25 mm
  • Accuracy within +/- 0.25 mm
  • Magnetic components:
    • stepper motor & motor driver  (little to no effect on scanned images)

Scans from Positioning Test with MRI

Top left: Moved sample to isocenter location
Top right: Adjusted position by 4.4mm to move desired location to isocenter
Bottom left: Moved the sample to home position and back to isocenter
Bottom right: Moved sample eight 0.5mm steps forward, eight 0.5mm steps back