Skip to content

Image Reconstruction

Enabling clinicians to infer physical qualities from hybrid medical imaging methods, creating 2D and 3D images from indirect imaging techniques

What does this
Methodology Involve?

Imaging has revolutionised the role of data acquisition within healthcare, allowing non-invasive visualisation, both qualitatively and quantitatively, of a vast range of clinically critical parameters. Whereas direct imaging allows for the capture of data across the electromagnetic and acoustic spectra, in many cases continuously and in real-time, indirect imaging incorporates also a reconstruction step that allows the clinician to infer many other physical quantities through the solution of an inverse problem based, for example, on a physical or statistical model of the data acquisition process.

This theme addresses the reconstruction task at the intersection of advanced skills in mathematics, natural sciences, computing, statistics, and engineering. The UK has an outstanding track record of invention in imaging and measurement systems, for example the early development of X-ray CT and Magnetic Resonance Imaging, both awarded Nobel prizes.

Emerging hybrid medical imaging methods such as photoacoustic tomography and magnetic resonance impedance tomography are at a phase of rapid technological and mathematical development, and improvements of existing methods including fast CT, MRI for lung dynamics and limited data dental CT will continue to ensure such success.The demand for a trained cohort of experts with these diverse skills is constant and high, both nationally and internationally.

Partners and Facilities

Work on this methodology spans the department of Computer Science at UCL and be applied to any of the clinical themes.

Studying Under This Theme

Projects listed below are available for applicants to include in their 5 project choices. Successful candidates will meet with both primary supervisors from their selection before being assigned to their allocated research project to ensure that each student gets the best opportunity for their skills and development. To view all available project videos please visit our YouTube channel.

Theme Leader

Prof. Simon Arridge
UCL Centre for Medical Image Computing

Simon Arridge’s research interests primarily lie in the area of tomography in medical imaging, specifically the application of inverse problem techniques to image reconstruction. Inverse problems can be linear or non-linear and either well posed or ill-posed. Ill-posed inverse problems usually require regularisation techniques which can be placed within the general frame work of Bayesian estimation, where the assumed prior distribution of the image under consideration plays the role of a penalty term in a constrained or unconstrained posterior probability optimisation.