Developing and validating multi-outcome transformer models for clinical prediction

risk of an outcome that matters for a particular clinical decision. Models are often evaluated across broad sets of diagnosis codes, with performance summarised across many predictions. These overall results can obscure weaker performance for specific, clinically important outcomes and may give an overly favourable impression of how the model would perform in its intended use. Many applications also focus on single outcomes, whereas clinicians may need the risks of several related outcomes and the probability that they occur together. Separate predictions may be inconsistent, and good discrimination does not establish that predicted probabilities are well calibrated or more useful than those from simpler models.

This methodological PhD will address these challenges through the development and validation of multi-outcome transformer prediction. A focused review will establish the state of the field and inform the choice of a clinically meaningful use case, including its population, outcomes, prediction time and risk horizon. Reproducible baseline models will provide comparators for a structured transformer method that estimates coherent risks for individual outcomes and their combinations. Targeted simulations and real-data analyses will assess calibration, predictive performance and potential clinical value for that specified use case. Clinical examples and findings will be integrated throughout the PhD, while established BEHRT pipelines will provide the infrastructure for the methodological work.

Research question

How can longitudinal EHR transformers be developed and validated to produce coherent and clinically meaningful predictions for multiple related outcomes?

Objectives

1. Review the field and establish a clinical use case and baseline models. Conduct a focused review of multi-outcome prediction with longitudinal EHR transformers and its validation to identify the methodological gap and inform use-case selection. Define a small, clinically coherent set of outcomes, population, prediction time and horizon, then develop reproducible single-outcome and simple multi-outcome comparator models as benchmarks.
2. Develop multi-outcome transformer methodology. Extend an established EHR transformer with a structured prediction approach that estimates clinically relevant marginal and joint risks while respecting the relationships between them.
3. Develop and apply multi-outcome validation methods. Evaluate whether the proposed risks are coherent, calibrated and informative for clinical decisions, using targeted simulations and real-data comparisons with the baseline models.

Desirable Student Background:

The project would suit a student with a strong quantitative degree in statistics, mathematics, computer science, data science or a related discipline. Essential preparation includes Python programming, probability and statistical modelling, machine learning, and careful work with complex data. Experience with PyTorch, transformers, survival analysis, multivariate modelling or health data would be advantageous. The successful candidate should be motivated by methodological development.

How to apply:

Please apply under the University of Manchester application portal and select 'PhD in Artificial Intelligence' OAA Application portal

When applying, you’ll need to specify the full name of this project, the name of your supervisor, if you already having funding or if you wish to be considered for available funding through the university, details of your previous study, and names and contact details of two referees.

Your application will not be processed without all of the required documents submitted at the time of application, and we cannot accept responsibility for late or missed deadlines. Incomplete applications will not be considered.

You will be asked to upload the following supporting documents:

• Final Transcript and certificates of all awarded university level qualifications
• Interim Transcript of any university level qualifications in progress
• CV
• Supporting statement: A one or two page statement outlining your motivation to pursue postgraduate research and why you want to undertake postgraduate research at Manchester, any relevant research or work experience, the key findings of your previous research experience, and techniques and skills you’ve developed. (This is mandatory for all applicants and the application will be put on hold without it).
• Contact details for two referees (Please ensure referee email addresses are official university or workplace accounts, as we may need to verify the references.)
• English Language Certificate (if applicable)

We actively encourage applicants from diverse career paths and backgrounds and from all sections of the community, regardless of age, disability, ethnicity, gender, gender expression, sexual orientation and transgender status.

We also support applications from those returning from a career break or other roles. We are dedicated to supporting work-life balance and offer flexible working arrangements to accommodate individual needs. Our selection process is free from bias, and we are committed to ensuring fair and equal opportunities for all applicants.

We are dedicated to supporting work-life balance and offer flexible working arrangements to accommodate individual needs, including a part-time study option.

Join us and help grow the UK’s national capability in artificial intelligence, part of the UK’s modern industrial strategy. Everyone is welcome if you are passionate about uncertainty quantification, decision making with humans in the loop and decision making for machine learning systems and their applications in Science and Engineering, and looking to develop your research and innovation skills at the doctoral level to further your career in the AI sector.

Equality, diversity and inclusion is fundamental to the success of The University of Manchester, and is at the heart of all of our activities. We know that diversity strengthens our research community, leading to enhanced research creativity, productivity and quality, and societal and economic impact. (Equality, diversity and inclusion | The University of Manchester)

We actively encourage applicants from diverse career paths and backgrounds and from all sections of the community, regardless of age, disability, ethnicity, gender, gender expression, sexual orientation and transgender status.

Funding Notes

Fully funded AI UKRI CDT 4 year program; The Project will be based in University of Manchester.

The UKRI AI CDT in Decision Making for Complex Systems is part of the TechFirst offering stipends of £30,000 for students eligible for home fee status. The start date is September 2027.

References

Selected references

1. Li Y, Rao S, Solares JRA, et al. BEHRT: Transformer for Electronic Health Records. Scientific Reports. 2020;10:7155. doi:10.1038/s41598-020-62922-y.
2. Yildiz Y, Nenadic G, Jani M, Jenkins DA. Will Large Language Models Transform Clinical Prediction? Diagnostic and Prognostic Research. 2025. doi:10.1186/s41512-025-00211-w.
3. Yildiz Y, Nenadic G, Jani M, Jenkins DA. Extending BEHRT to UK Biobank: assessing transformer model performance in clinical prediction. Frontiers in Digital Health. 2026;8:1715506. doi:10.3389/fdgth.2026.1715506.
4. Martin GP, Sperrin M, Snell KIE, Buchan I, Riley RD. Clinical prediction models to predict the risk of multiple binary outcomes: a comparison of approaches. Statistics in Medicine. 2021;40:498–517. doi:10.1002/sim.8787.
5. Pate A, Martin GP, Riley RD, et al. Developing prediction models to estimate the risk of two survival outcomes both occurring: a comparison of techniques. Statistics in Medicine. 2023;42:3184–3207. doi:10.1002/sim.9771.
6. Pate A, Sperrin M, Riley RD, et al. Calibration plots for multistate risk prediction models. Statistics in Medicine. 2024;43:2830–2852. doi:10.1002/sim.10094.
7. Martin G, Pate A, Bladon S, Sperrin M, Riley R. A decision-analytical perspective on incorporating multiple outcomes in the production of clinical prediction models: defining a taxonomy of risk estimands. BMC Medicine. 2025;23:142. doi:10.1186/s12916-025-03978-3.