Engineering
Engineers at the Institute of Biomedical Engineering, led by Professor Alison Noble, develop robust algorithms and open, reproducible workflows for clinical AI. See Noble Group.
SHAIPE
A cross-disciplinary collaboration designing, evaluating, and translating trustworthy AI to enhance patient safety.
SHAIPE - Safer Healthcare AI through Psychology and Ergonomics
Psychology
Researchers in the Department of Experimental Psychology, led by Nick Yeung, study how clinicians use AI tools in decision-making and how to support effective human–AI teaming. See Attention & Cognitive Control Lab.
Clinical practice
Clinicians at OxSTaR, led by Professor Helen Higham, focus on human factors and simulation-based training to evaluate AI-enabled pathways and improve safety in real clinical settings. See OxSTaR.
Research themes
- Trustworthy and transparent AI
- Human factors and clinical simulation
- Decision support and workflow integration
- Evaluation frameworks and safety metrics
Recent highlights
- Collaborative projects across engineering, psychology, and clinical simulation
- Workshops on safe deployment and evaluation of AI tools
- Preprints and publications on human–AI decision-making
About SHAIPE
SHAIPE (Safer Healthcare AI through Psychology and Ergonomics) is a cross-disciplinary initiative uniting:
- Engineers at the Institute of Biomedical Engineering (IBME), led by Professor Alison Noble;
- Psychologists in the Department of Experimental Psychology, led by Professor Nick Yeung;
- Clinicians at OxSTaR, led by Professor Helen Higham, focusing on human factors and simulation.
We design, evaluate, and translate AI technologies to improve patient safety and support clinical decision-making.
SHAIPE team at the 3rd Human–AI for Clinical Decision-Making Workshop, IBME, University of Oxford — 12th September 2025
People
SHAIPE is a collaborative group spanning engineering, psychology, and clinical practice.
Leads
Alison Noble
Institute of Biomedical Engineering, University of Oxford.
Nick Yeung
Department of Experimental Psychology, University of Oxford.
Helen Higham
OxSTaR Centre, University of Oxford — Clinical Lead.
Members
Joshua Strong
DPhil Student — Engineering
Interests: Human–AI Collaboration, healthcare, LLMs
Working on: Deferral applications in healthcare
Email: joshua.strong@eng.ox.ac.uk
Cheng Ouyang
Faculty Member — Engineering
Working on: multi-modal analysis for human–AI collaborative image analysis.
Personal webpage: cheng-01037.github.io. MICCAI workshops on Human–AI Collaboration and Uncertainty for Machine Learning in Medical Imaging.
Harry Rogers
Postdoc — Engineering
Interests: Human–AI Collaboration, VLMs
Working on: AI as An Instructor
Nicholas Phillips
Academic Specialised Foundation Doctor — Clinical
Interests: AI in healthcare and education
Working on: Predictive Melanoma Models, AI in Dermatology
Email: nicholas.phillips@msd.ox.ac.uk
Open to collaboration: LinkedIn
Rubeta Matin
Consultant Dermatologist (OUHFT) & Honorary Senior Clinical Lecturer (RDM) — Clinical
International academic clinical expert in diagnosis and management of skin cancer and skin disease in immunosuppressed patients, with a strong background in AI for dermatology. Committed to safe, equitable, and ethically responsible AI.
Ben Attwood
Chief Digital & Information Officer (OUHFT) & Consultant in Anaesthesia & Intensive Care — Clinical
Ben is Chief Digital and Information Officer for OUH NHS Foundation Trust and a Consultant in Anaesthesia and Intensive Care. He is passionate about ensuring information and data is presented to the right people in the right place to deliver higher quality care.
Anna Louise Todsen
DPhil Student — Psychology
Interests: Cognitive and systems-level perspectives on clinical AI development and implementation
Working on: Evidence from real clinical environments across multiple specialties
Email: anna.todsen@lincoln.ox.ac.uk
Pramit Saha
Postdoc — Engineering
Interests: Federated learning, human–AI collaboration, multimodal learning, agentic AI in healthcare
Research
Themes
- Trustworthy, transparent AI methods for clinical settings
- Human factors, simulation, and evaluation
- Decision support and workflow integration
- Safety metrics, audit, and governance
Projects
Understanding AI integration in dermatology
Ethnographic study at Oxford University Hospitals NHS Foundation Trust on how AI fits dermatology practice, aligned with the NHS 10‑year plan.
- Focus: feasibility, acceptability, and clinical need (the “applicability of AI”).
- Methods: observations, interviews, focus groups, and co‑design with clinicians; guided by the In‑Situ System Observation Guide (I‑SOG).
- Outcomes: workflow fit, clinician trust, and adoption barriers and enablers.
AI‑assisted assessment in VR paediatric training
We explore a Human–AI Collaboration (HAIC) approach to assessing medical students in VR paediatric emergency simulations. The system auto‑grades clear performances and flags ambiguous cases for expert review, aiming for scalable, consistent, and objective skills assessment with high‑quality feedback.
Publications
Selected publications and preprints relevant to clinical AI and human–AI decision-making.
- Coming soon — add BibTeX or auto-import from ORCID/DBLP/Google Scholar.
Welcome to SHAIPE
Posted 2026
We are launching a cross-disciplinary effort to improve safety in clinical AI. Watch this space for updates.
Events
Upcoming talks, workshops, and symposia on clinical AI safety.
- LLMs@Oxford: “LLMs as clinical decision support tools” — talk by Professor Helen Higham.
- Human–AI for Clinical Decision‑Making Workshop (third) — hosted by SHAIPE at the Institute of Biomedical Engineering, University of Oxford.
View the original website HERE