Cookies on this website

We use cookies to ensure that we give you the best experience on our website. If you click 'Accept all cookies' we'll assume that you are happy to receive all cookies and you won't see this message again. If you click 'Reject all non-essential cookies' only necessary cookies providing core functionality such as security, network management, and accessibility will be enabled. Click 'Find out more' for information on how to change your cookie settings.

Andrey Kormilitzin

Dr Andrey Kormilitzin is a senior researcher in the Department of Psychiatry and leads the AI and Natural Language Processing team within the Translational Neuroscience group. 

Tell us a little about yourself, and what attracted you to studying/working at the University of Oxford?  

I came to Oxford in late 2014 as a postdoctoral researcher in machine learning and stochastic analysis at the Mathematical Institute. I was thrilled to join the Collaborative Oxford Network for Bipolar Research to Improve Outcomes (CONBRIO) project alongside Professor Terry Lyons (Maths) and Professor John Geddes (Psychiatry) to work on the applications of a new mathematical approach – the path signature, to make sense of self-reported mood data from the TrueColours platform. In 2017 I joined the Department of Psychiatry, in Simon Lovestone’s Translational Neuroscience group, to lead the Natural Language Processing team. We have been developing machine learning tools to maximise the utilisation of the largest in the UK secondary mental health electronic medical records database (UK-CRIS) for downstream epidemiological analyses. The NLP research and other machine learning applications have led to the recent NIHR AI Award that allowed us to establish a new collaborative research theme in the Department of Psychiatry dedicated to algorithmic fairness, system efficacy and clinical decision support. This project is called CHRONOSIG.

Apart from my scientific work, I support the Family Friendly and Part-Time Athena SWAN Working Group. I also organise a weekly seminar series on Artificial Intelligence and Machine Learning in Mental and Cognitive Health. As a member of the Mathematical Institute, I supervise a cohort of 4-5 MSc mathematical students each year and their dissertations on topics related to Artificial Intelligence and Machine Learning in Health and Medicine. This provides a unique opportunity for mathematicians to be involved in real-world clinical data-driven research. 

What is your vision for the team/project/research you study/work with?

I am enthusiastic about the promise that abstract mathematical ideas and artificial intelligence hold to advance our understanding of fundamental science and empower clinicians and researchers when carrying out laborious and time-consuming tasks. However, with great power comes great responsibility. Sophisticated algorithms can unintentionally learn from biased historical medical data and generate predictions and recommend decisions that may be erroneous or even harmful for underrepresented communities or patients with rare diseases and conditions. Furthermore, algorithms can learn and disclose unobserved characteristics, such as sexual orientations, that can cause distress to patients and lead to mistrust in new technologies. Recent advances in AI/ML have led to powerful tools that allow us to crunch numbers and support research. However, the socio-technological aspects and the human-computer interaction are largely missed.

Of course, it is impossible to make substantial progress single-handedly. I am honoured to lead a team of seven excellent machine learning researchers, from graduate students to postdocs, with the aim to address the aforementioned challenges and develop a systematic approach to meaningful AI/ML research in mental health. 

What is currently at the top of your To-Do List?

I am totally dedicated to setting up and running the CHRONOSIG programme, alongside my colleagues Dr Dan Joyce, Associate Professor Alejo Nevado-Holgado, Dr Tony James, Julia Hamer-Hunt and Professor Andrea Cipriani. The programme will enable us to translate theoretical advances in AI/ML into clinical research in a reproducible and accessible way. 

How did you get to where you are today?

I stubbornly followed my passion and curiosity. I am also indebted to all the incredible people who have supported and mentored me in the past and at present.  I was very lucky to work alongside, watch and learn from great scientists. Their patience, sincere readiness to help and constructive feedback allowed me to shape my vision and scientific attitudes. My interdisciplinary research spanning mathematics, artificial intelligence, engineering and their translation to medicine, gave me a unique opportunity to communicate effectively with people from very diverse backgrounds and research cultures. 

Who or what inspires you?

While it’s impossible to pin that down to a single person who influenced me, one of them is Richard Feynman – Nobel Prize laureate, physicist, and great educator. His attitude of being honest, open to new ideas and most importantly -- having fun with science is truly inspirational. In particular, I like one of his quotes: “The first principle is that you must not fool yourself and you are the easiest person to fool." 

If you were not in your study programme/job currently, what would you like to be doing?

If I wasn’t a scientist, I would probably be an architect.