Postgraduate research project

X-ray imaging using ultrafast laser-generated soft X-rays and machine learning

Funding
Competition funded View fees and funding
Type of degree
Doctor of Philosophy
Entry requirements
2:1 honours degree View full entry requirements
Faculty graduate school
Faculty of Engineering and Physical Sciences
Closing date

About the project

A new area of imaging has opened up, using coherent light to illuminate objects, and computer algorithms to analyse the scattered light and generate images. Developments in ultrafast lasers have made sources of coherent soft X-rays possible in the lab, rather than relying on large-scale installations like synchrotrons.

The combination of these two techniques has been successful in producing a new generation of X-ray microscopes, and at Southampton we have been at the forefront of this development. In particular, we have chosen areas of biology where imaging below the 100nm length scale can provide new information about processes occurring within cells.

The idea of combining algorithmic image reconstruction techniques with techniques based on machine learning has been proposed. Many of the problems of image reconstruction are similar to those addressed by machine learning (ML), particularly using convolutional neural networks. This project aims to use soft X-ray scattering and a combination of algorithmic and ML-based computer techniques to develop new ways of imaging, particularly in the area of biological science.

As part of the project, you will become familiar with high-energy ultrafast laser science, using lasers with pulse lengths below 50 femtoseconds and peak powers in the terawatt regime, and you will also be trained in the use of algorithmic and ML techniques for data analysis and image reconstruction. Significant research expertise in both high-energy ultrafast lasers and machine learning exists in Southampton, and this project will involve working with members of the Engineering and Physical Sciences Research Council EPSRC-funded AI for Scientific Discovery network.