FindA University Ltd Featured PhD Programmes
Xi’an Jiaotong-Liverpool University Featured PhD Programmes
FindA University Ltd Featured PhD Programmes

Multimodal image-to-image translation for drug discovery


School of Computer Science

This project is no longer listed on FindAPhD.com and may not be available.

Click here to search FindAPhD.com for PhD studentship opportunities
Dr I Styles No more applications being accepted Funded PhD Project (European/UK Students Only)

About the Project

This project will develop state of the art computer vision and machine learning approaches to help scientists understand images from some of the most advanced optical biological imaging techniques currently available in a collaboration between the University of Birmingham, GlaxoSmithKline (GSK), and Professor Stephen Boppart, Director of the GSK Center for Optical Molecular Imaging (COMI) at the Beckman Institute for Advanced Science and Technology, on the campus of the University of Illinois at Urbana-Champaign (USA).

GSK-COMI develops multi-modal real-time imaging techniques that do not require biological samples to be chemically labelled. They enable real-time imaging of biology without labels that can interfere with the processes that are being studied. This capability is central to addressing core problems in drug design by enabling longitudinal studies over the timespan of a disease’s progression in a single sample or subject, and across multiple dimensional scales from the cell to clinic. COMI’s unique instrumentation allows multiple imaging techniques to be deployed in rapid succession, giving a multi-modal view where each modality reveals different information about the sample.

The project will develop techniques to help scientists understand the enormous quantities of data that are generated when imaging multiple samples over time. Humans cannot manually screen this data and automated analysis techniques are essential. We will develop techniques that will combine information from across the modalities and the time points to summarise the data in a form that can be qualitatively and quantitively interpreted by a human expert. Our approach will leverage recent advances in generative adversarial modelling to learn to map information across modalities. This will allow identification of events in an image mode that are inconsistent with the other modes. The key features of the events in each mode will then be extracted, mapped and summarised into a common latent space. The project will explore a range of techniques from machine learning and artificial intelligence, aiming to establish robust applicability and generalizability across imaging techniques and data scales. The successful candidate will develop considerable expertise in these methods and their application to advanced imaging and the drug development process. This is therefore a rare opportunity to explore the interface between AI, drug development, and advanced optical imaging. The insights gained may also serve to guide future development of the imaging technology in order to improve the ability to identify key features of the data that are found to be important biomarkers that reflect disease state and drug effect.

The successful applicant will have a good (at least 2.1) first degree in Computer Science, Physics, Mathematics or similar numerate disciplines. A strong background in programming, ideally in Python, is essential. Applications are especially welcomed from underrepresented communities. The University of Birmingham holds Athena Swan and Race Equality Charter Bronze Awards.
Funding is available to pay tuition fees and a stipend at standard UKRI rates for Home/EU students. Additional financial support from GSK will provide an additional stipend payment and funding for consumables and travel. There will be opportunities to spend time at GSK, and COMI. The project will also benefit from access to excellent high-performance computing facilities.
Interviews will be held in the week beginning Monday 17 August in advance of a start date of 28 September 2020. Informal enquiries are welcomed and should be directed to Professor Iain Styles ([Email Address Removed]).

Funding Notes

Main source of funding is EPSRC Institutional DTA and so UKRI elicibility rules apply. A minimum 2.1 is a numerate discipline is essential.
Search Suggestions

Search Suggestions

Based on your current searches we recommend the following search filters.



FindAPhD. Copyright 2005-2020
All rights reserved.