King’s College London has been named the world’s leading institution for research exploring the use of artificial intelligence (AI) in mental health, according to a new international analysis of almost 10,000 scientific papers published over the past decade.
The study, published in the Australian Journal of Psychology in August, analysed 9,623 research articles on AI-driven mental health research published between 2016 and 2025. Researchers found that the field has expanded rapidly in recent years, with applications ranging from mental health diagnosis and digital interventions to suicide prevention and support for people living with post-traumatic stress disorder (PTSD).
King’s College London emerged as the most productive institution in the field worldwide with 182 publications. King’s was followed by Harvard Medical School and the University of Washington, which have 159 and 132 publications respectively. The study underscores King’s long-standing contribution to digital mental health research and highlights the growing global interest in harnessing AI to improve mental health care and outcomes.
King's College London has played a leading role in advancing research at the intersection of AI and mental health, with much of this work centred within the Institute of Psychiatry, Psychology & Neuroscience and its close partnership with the South London and Maudsley NHS Foundation Trust and the NIHR Biomedical Research Centre: Maudsley. As these technologies continue to develop, it is vital that we ensure they are used safely, ethically and in ways that genuinely benefit patients and clinicians. This analysis reflects the breadth and impact of the work that has developed through that research environment.
Professor Richard DobsonProfessor of Medical Bioinformatics, IoPPN; Lead of the Informatics Theme in the NIHR BRC: Maudsley
The researchers looked at the publications by country, with the US ahead with 3,107 publications, followed by China with 1,890 publications, and in third the United Kingdom with 973 publications.
The study also showcased the top 10 authors in the realm of AI and mental health research, ranked according to their publication volume. Professor Robert Stewart, Professor of Psychiatric Epidemiology & Clinical Informatics at the IoPPN ranked first, authoring 49 publications and signifying a substantial contribution to this field.
Professor Stewart is Deputy Lead of the Informatics Theme in the NIHR Biomedical Research Centre: Maudsley. He has been Academic Lead for South London and Maudsley NHS Foundation Trust’s Clinical Record Interactive Search (CRIS) platform since its development in 2007. The study highlighted his collaborative role, working across a wide network.
At King’s and South London and Maudsley, we have been working with mental healthcare information for nearly 20 years now and what we’re ultimately trying to do here is help mental health services improve themselves. If I see a doctor, I’m expecting them to have learned from all the other patients they’ve seen in their career, as well as from whatever they were originally taught at medical school. In the same way, now that all of its information is digital, there ought to be ways in which a mental health service can learn from what has worked well or less well in the past.
Professor Robert StewartProfessor of Psychiatric Epidemiology & Clinical Informatics; Deputy Lead of the Informatics Theme in the NIHR BRC: Maudsley
Professor Stewart continues, “We have been very fortunate at King’s and South London and Maudsley to have had many people with lived experience of mental healthcare who have been willing to give their time and expertise to help direct our research, alongside experts in information governance and data security to ensure that the highest standards are maintained. The progress we have made has also depended on researchers, technicians, and clinical staff working closely together to focus on what will most help improve services and healthcare. There’s nothing individual in data science – it’s all about teams and collaboration.”
Professor Angus Roberts, Biostatistics & Health Informatics Department at IoPPN, Dr Rashmi Patel, Visiting Senior Clinical Lecturer at King’s rank second (jointly with Professor Munmun De Choudhury from Georgia Tech, and Dr Bin Hu from University of Aberdeen, each having 21 publications).
The study found that research is increasingly moving beyond the development of predictive algorithms towards patient-centred clinical applications, including the use of generative AI tools. They highlight the need for future research to address challenges around patient safety, algorithmic bias, data privacy and the responsible integration of AI into healthcare.
