Hamid Dehghani

PhD Studentships

Applications are invited for an internationally funded studentship in the School of Computer Science, University of Birmingham, beginning September 2015, working under the supervision of Dr Hamid Dehghani, with partnership with our collaborators at University of Washington in St Louis, USA.

Background Information

Diffuse optical imaging (DOI), is an emerging, non-invasive technique with unique portability and hemodynamic contrast capabilities, which can record evoked brain function in enriched or clinical environments. However, despite unique strengths, DOI as a standard tool for functional mapping has been limited by low spatial resolution, limited depth penetration and a lack of reliable and repeatable mapping. Though DOI of brain activity is commonly performed using topography with sparse imaging arrays, high-density arrays and tomography algorithms provide a means to dramatically increase image quality.
The overall goal of this project is to develop fast optical tomography methods for mapping resting state functional connectivity in order to study the childhood development of brain functions. We recently demonstrated the feasibility of a DOT prototype that solves several basic challenges in inter-channel cross talk and enables tomography of the adult visual cortex (Eggebrecht et. al. Nature Photonics 2014).
This PhD reserach project will specifically investigate the creation and use of CPU/GPU based implementation of our NIRFAST code (www.nirfast.org) to allow real-time model based image reconstruction. Additionally, we will be interested in the development of novel, efficient parameter recovery algorithms to allow multi-parameter, whole space imaging capabilities. More specifically, the aims of this project will involve (1) Fast anatomical head modeling and image reconstruction using atlas based models, (2) Development of robust CPU/GPU computational libraries under MATLAB to allow multi-platform parallelisation to allow (3) DOT brain mapping in a common brain atlas and (4) Validation methods.
Applicants should have (or be about to receive) an honours degree (at least 2.1 or equivalent) in Computer Sciences or other relevant discipline. Experience in the following areas is crucial: MATLAB, Computer Programming (OpenCL, CUDA, JAVA, C++), Image Processing and Finite Element Methods.

Further Information

For further more details please contact: H.Dehghani@cs.bham.ac.uk

Funding

The studentship offers tuition fees and a maintenance grant and is available to HOME/EU applicants.

How to apply

The admission process for PhD studentships is detailed here.