About the Project
A fully funded PhD opportunity to participate in the world-leading research undertaken by the EPSRC Doctoral Landscape Award at the University of Sheffield.
Supervised by Dr Xi Wang and Dr Ning Ma.
The University of Sheffield is one of the leading Russell Group universities in the UK. We carry out cutting-edge research with strong links to industry. When you enrol to do a PhD with us, you will be working with world-leading academics and have access to top of the range facilities. As a PhD student you will have the opportunity to gain skills not only to conduct research, but also to take your career to the next level, whether you want to stay in academia, go into industry or the public sector, or set up your own company. You will have access to a range of training and support services to help you excel in your studies and beyond.
Project Description
Getting help for depression often starts with a wait: for an appointment, for an assessment, for someone to have the time to properly listen. What if a short conversation, available whenever it is needed, could help flag who needs support, and how urgently?
This project will explore that question by developing conversational AI agents that can screen for depression through natural dialogue, grounded in clinically validated tools such as the Patient Health Questionnaire-9 (PHQ-9). It will build simulated patient and clinician agents that generate and refine realistic screening conversations, capturing how depression actually presents across different people, rather than treating everyone the same way.
Because the resulting conversations are synthetic, they can be shared openly as a large-scale, privacy-preserving benchmark, something real clinical interview data cannot offer. This will help researchers build and test screening tools that are more scalable and consistent, and better able to support an overstretched mental health workforce.
You will be a member of the University of Sheffield's Natural Language Processing research group within the School of Computer Science, a large and vibrant team of researchers with a wide range of collaborators within and beyond the University of Sheffield, working alongside the iNLP Lab and NHS clinical partners.
The Challenge: Simulating Real Clinical Conversations
Building conversational depression screening tools that work in practice is hard. Real clinical interviews are difficult to replicate at scale, since participant demographics, rating scales and interview settings all vary between studies. Interview data is also sensitive: it is rarely shared publicly, which makes it difficult to build the large, reproducible benchmarks that training and evaluating screening models requires. Most natural language processing research instead relies on social media data, but the gap between casual online text and a structured clinical interview limits how far these results can be trusted.
The Project: Privacy-Preserving Conversational Simulation
This project addresses these challenges with two core components. The first is a simulation environment that turns self-report tools such as PHQ-9 into synthetic, multi-turn interviews, personalised to reflect individual differences in how depression is expressed. The second is a multi-agent system, in which simulated patient and clinician agents interact within this environment to generate and refine those conversations, with real clinicians in the loop to validate their realism. Because the data is synthetic, it can be released openly, without the ethical and privacy concerns that limit real interview data. Recent progress in large language models (LLMs), the technology behind systems like ChatGPT, makes agents capable of holding this kind of open-ended, context-aware conversation for the first time.
Why Join This Project?
Depression screening sits at the intersection of pressing real-world need and genuinely open research questions, from how to simulate realistic human behaviour to how to build AI systems that are safe to use in healthcare. You will help build tools that could ease pressure on stretched mental health services and reach people who currently go unscreened, in the UK and in countries with even fewer mental health professionals.
Candidate Profile
We welcome applicants with a strong background in computer science or a related quantitative discipline. Programming skills (e.g., Python) are essential. Experience with natural language processing, machine learning, conversational AI or health informatics would be an advantage, as would an interest in working responsibly with sensitive data. The most important qualities are curiosity and a drive to solve complex problems with real-world impact.
How to apply
Interested candidates are strongly encouraged to contact the project supervisors to discuss your interest in and suitability for the project prior to submitting your application.
Please refer to the EPSRC DLA webpage for detailed information about the EPSRC DLA and how to apply.
Applicants must ensure that they use the correct application form linked to the EPSRC DLA webpage and must submit an application for October 2027 entry. The form is unique to the EPSRC DLA. It contains the list of advertised projects from which to select, and three questions specific to the DLA.
The application deadline is 17:00 GMT 13 January 2027.
The majority of our DLA studentships will be allocated to candidates who qualify for Home fee status, in accordance with EPSRC funding regulations which place strict limits on international recruitment.
Funding Notes
The award will fund the full tuition fee and a maintenance stipend at the UKRI rate (currently £21,805 per annum) for 3.5 years, as well as a research grant to support costs associated with the project.