My Research
My research in Medical Physics has
mainly focused on the application of
mechanical engineering techniques to ultrasound imaging research.
This has led to my work as part of an EPSRC funded project led by Peter Hoskins of Medical Physics and Bill
Easson of the School of Engineering at the University of Edinburgh.
I have also worked as a mechanical design and optics engineer
in the
research division of Lab901, which develops laboratory automation products for the life
sciences.
My most recent work is in the field of Magnetic
Resonance Imaging (MRI). I have mainly been developing an
actuator for MR elastography (MRE) acquisition in the Clinical Research Imaging Centre for Neil Roberts. I have also been developing a two chamber calibration phantom for MR spectroscopy measurement of brain temperature for Ian Marshall.
You can find out more about the ultrasound imaging research I carried out in my PhD thesis, which you can download here as a PDF file.
I also have three other pages related to my work - see my LabVIEW page,
my Stradwin page and my
Solid Edge page.
3D ultrasound
I developed a 3D ultrasound system for arterial scanning based on
the Stradwin 3D ultrsound system. The system consisted of
- a Philips
HDI5000 APM medical ultrasound scanner
- a fast PC with a video capture card running Stradwin
- an NDI Polaris optical tracking system
- a Traxtal Adaptrax multi-faceted optical tracking tool.
This was used to scan healthy volunteers and patient
volunteers with diagnosed arterial disease in the common femoral
arteries or the common
carotid arteries.
For most patients and healthy volunteers, a 3D scan of their artery was
successfully made.
Some patients and healthy volunteers with abdominal aortic anuerysm
(AAA) were also scanned with less success due to the presence of bowel
gas.
Image-guided modelling
In image-guided modelling (IGM) I was responsible for
- segmenting arterial 3D ultrasound images using the
Sheffield
Image Registration Toolkit (ShIRT) and MATLAB
- converting segmented images into a format suitable for
creating a
3D mesh model
- creating 3D mesh models using the Rhino surface modelling
package.
- visualising Computational Fluid Dynamics (CFD) images using
ParaView.
I also carried out an evaluation of the performance of the ShIRT
segmentation method and the assumed "gold standard" of manual
segmentation using a set of segmentation performance metrics.
The 3D ultrasound and IGM work was published in Ultrasound in Medicine
and Biology (doi:10.1016/j.ultrasmedbio.2009.06.1099).
Arterial
wall motion
There is some interest in developing arterial wall motion (AWM)
techniques for directly measuring arterial wall displacement and
velocity in vivo using ultrasound.
A vital requirement for any such technique is being able to measure the
accuracy of the technique.
To achieve this, I designed a phantom that simulated a cross-section
through an artery. The artery section could move with a programmed set
of physiological displacements and velocities to provide a realistic
test of the AWM technique.
This work was published in Ultrasound in Medicine and Biology
(doi:10.1016/j.ultrasmedbio.2007.04.005
).