Are you applying to universities? | SHARE YOUR EXPERIENCE Are you applying to universities? | SHARE YOUR EXPERIENCE

EPSRC DTP PhD project: Natural Language Processing to Improve Healthcare Guidelines and Reduce Medication Errors

   Department of Life Sciences

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

Click here to search for PhD studentship opportunities
  Dr Matthew Jones  No more applications being accepted  Competition Funded PhD Project (Students Worldwide)

About the Project

The University of Bath is inviting applications for the following PhD project commencing in October 2023.

Eligible applicants will be considered for a fully-funded studentship – for more information, see the Funding Notes section below.

Supervisory Team:

Lead supervisor: Dr Matthew Jones, Department of Life Sciences

Co-supervisor: Dr Harish Tayyar Madabushi, Department of Computer Sciences

Overview of the Research:

Hundreds of millions of medication errors occur globally each year, many of which harm patients with a cost of $42 billion/year. Healthcare professionals rely on written guidelines to provide the information needed to prescribe and administer medicines. However, difficulties finding and understanding information in such guidelines also contribute to errors.

A University of Bath study applied a user testing technique to improve the design of the NHS Injectable Medicines Guide, leading to 2.5x reduction in medication errors, a 13% reduction in staff time and an estimated five-year net benefit of >£3m/drug. However, user testing of the thousands of current healthcare guidelines is limited by lack of resources and skills. A semi-automated platform to user test and suggest improvements to guidelines would therefore increase the impact of this research.

As such, the goal of this project is to develop a system that automatically rewrites guidelines to be more understandable. Crucially, such a system will have a human in the loop to ensure that information critical to the welfare of patients is not missed or misrepresented by the automated system. Feedback from human experts additionally allows continuous feedback based on which the system will continue to evolve with increased adoption across healthcare practitioners. 

State-of-the-art deep learning methods used in Natural Language Processing (e.g., automatic summarisation), have the potential to automate this process. However, such methods, which rely on opaque deep learning methods, could potentially generate guidelines missing key information. Consequently, this project aims to develop methods that guide deep learning models to: a) retain critical pieces of medical information, b) generate output in clear, possibly predefined sections, and c) ensure that the output guidelines maintain the overall structure and style that pilot studies have found to be effective.

The selected candidate will be expected to deliver state-of-the-art methods in computational linguistics. While work will involve the use of medical data, students are not expected to have any prior knowledge of the medical domain. 

Project keywords: natural language processing, deep learning, NLP for medicine, medication errors, healthcare guidelines.

Candidate Requirements:

Applicants should hold, or expect to receive, a First Class or good Upper Second Class UK Honours degree (or the equivalent). A master’s level qualification would also be advantageous.

Excellent programming skills are highly desirable as is experience working with deep learning and natural language processing (e.g., pre-trained language models).

Non-UK applicants must meet our English language entry requirement.

Enquiries and Applications:

Applicants are encouraged to contact Dr Harish Tayyar Madabushi on email address [Email Address Removed] before applying to find out more about the project and to discuss their suitability for the role.

Formal applications should be made via the University of Bath’s online application form for a PhD in Pharmacy & Pharmacology.

More information about applying for a PhD at Bath may be found on our website.

Equality, Diversity and Inclusion:

We value a diverse research environment and aim to be an inclusive university, where difference is celebrated and respected. We welcome and encourage applications from under-represented groups.

If you have circumstances that you feel we should be aware of that have affected your educational attainment, then please feel free to tell us about it in your application form. The best way to do this is a short paragraph at the end of your personal statement.

Funding Notes

Candidates applying for this project may be considered for a 3.5-year Engineering and Physical Sciences Research Council (EPSRC DTP) studentship. Funding covers tuition fees, a stipend (£17,668 per annum, 2022/23 rate) and research/training expenses (£1,000 per annum). EPSRC DTP studentships are open to both Home and International students; however, in line with guidance from UK Research and Innovation (UKRI), the number of awards available to International candidates will be limited to 30% of the total.


Jones et al., 2022. User testing to improve retrieval and comprehension of information in guidelines to improve medicines safety. Journal of Patient Safety, 18(1), pp. e172-e179.
Jones et al., 2021 User-testing guidelines to improve the safety of intravenous medicines administration: a randomised in-situ simulation study. BMJ Quality and Safety, 30(1), pp. 17-26.
Jones et al., 2022. Costs and cost-effectiveness of user-testing of health professionals’ guidelines to reduce the frequency of intravenous medicines administration errors by nurses in the United Kingdom: a probabilistic model based on voriconazole administration. Applied Health Economics and Health Policy, 20, pp. 91-104.
Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W. and Liu, P.J., 2020. Exploring the limits of transfer learning with a unified text-to-text transformer. J. Mach. Learn. Res., 21(140), pp.1-67.
Brown, T., Mann, B., Ryder, N., Subbiah, M., Kaplan, J.D., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A. and Agarwal, S., 2020. Language models are few-shot learners. Advances in neural information processing systems, 33, pp.1877-1901.
PhD saved successfully
View saved PhDs