Meet the Team
-
Sotirios (Sotos) Tsaftaris
Hub Director
An expert in multimodal AI and applications in health at The University of Edinburgh, working closely with industry, NHS, and other stakeholders.
-
Hana Chockler
London Spoke Lead | EDI Lead
An expert in causal reasoning and explainability at King’s College London, a Google Faculty awardee, and a former Principal Scientist at causaLens.
-
Matthew Sperrin
Manchester Spoke Lead | ECR Champion
An expert in causal inference for decision support at the University of Manchester, leading several projects with industry and NHS partners.
-
Ricardo Silva
Research Theme Lead
An expert on causal inference and AI at University College London. He holds an EPSRC Open Fellowship and is a Turing fellow. Collaborates with industry partners such as DeepMind, Spotify and ByteDance.
-
Ben Glocker
Knowledge Transfer Lead
An expert in causality for image analysis focusing on the safe and ethical deployment of medical imaging AI at Imperial College London and Kheiron Medical Technologies.
-
Niccolo Tempini
Responsible Innovation & Ethics Lead
An expert in the governance, management and sharing of research data, and data infrastructure development at the University of Exeter and also a Turing Fellow.
-
Catherine Gauld
Hub Manager | Environmental Sustainability Champion
An expert having managed several large-scale multi-stakeholder projects, with a keen interest in AI for health and AI Ethics
-
Emily Lekkas
Partnerships Manager
An expert in driving forward high value innovation projects in technology sectors such as Data Science, AI, Digital Health, Medical Devices and Life Sciences.
-
Belgin Davidson
CHAI Hub Administrator
An expert in delivering portfolio activities and projects with 19 years’ experience in higher education.
-
Connie Aitkin
CHAI Hub Events Administrator
An expert in events management, having coordinated workshops, recruitment days, and networking events, ensuring seamless execution and positive outcomes.
-
Erin Johnstone
CHAI Business Development Executive
An expert in developing and managing multi-stakeholder partnerships for cutting-edge projects across healthcare, medical devices and the pharmaceutical industry, with a keen interest in AI.
Co-Investigators
Professor Daniel Alexander, University College London
Professor Kenneth Baillie, University of Edinburgh
Dr Sjoerd Beentjes, University of Edinburgh
Dr Elliot Crowley, University of Edinburgh
Dr Karla Diaz-Ordaz, University College London
Dr Javier Escudero Rodriguez, University of Edinburgh
Dr Henry Gouk, University of Edinburgh
Dr Anita Grigoriadis, King’s College London
Dr Hui Guo, University of Manchester
Professor Bruce Guthrie, University of Edinburgh
Dr Stephan Guttinger, University of Exeter
Professor Ewen Harrison, University of Edinburgh
Professor Yulan He, King’s College London
Dr Ava Khamseh, University of Edinburgh
Dr Yingzhen Li, Imperial College London
Professor Kia Nazarpour, University of Edinburgh
Professor Ram Ramamoorthy, University of Edinburgh
Professor Ginny Russell, University of Exeter
Dr Sohan Seth, University of Edinburgh
Professor Ian Simpson, University of Edinburgh
Dr Eliana Vasquez Osorio, University of Manchester
Professor William Whiteley, University of Edinburgh
Fabio De Sousa Ribeiro
Fabio`s research activity over the past few months has focused primarily on the study of new theoretical identifiability results for deep latent variable models. Identifiability is extremely important for both reliable representation learning and causal inference, as without identifiability guarantees it is difficult to determine whether our model estimates are truly causal or a product of confounding factors, selection bias, spurious correlations.
PDRA
David Kelly
Researches and develops new algorithms for black-box explainability. Since he started, we submitted a paper on explaining absence (specifically, the "no abnormalities" diagnosis for medical AI for images), a paper on multiple explanations (important for healthcare images with multiple abnormalities), and we are currently working on a journal paper describing the underlying algorithm and implementation. David also participates in development of explainability for spectra (applicable to Raman spectroscopy) and for tabular data (useful for tabular data in healthcare), and oversees an UG student working on explainability of 3D images (useful for explaining MRIs).
Nathan Blake
Works on applications of explainability in healthcare. Since he started, Nathan submitted a paper on explanations for MRI classifiers to Nature Communications in Medicine (partly including research that was done in the TAS Node). Currently, he is working on a line of papers on explainability for Raman spectroscopy, with the first paper demonstrating explainability for synthetic data (presented in the SPEC'24 conference in June), and subsequent papers extending the approach to in-vitro and in-vivo data. Nathan also provides guidance to a PhD student, working on personalised prediction of the risk for developing breast cancer for BRCA1&BRCA2 mutation carriers, using AI and causality.