Research & Evidence
Case Studies
Our work is grounded in peer-reviewed research and real-world NHS collaboration. The publications below document the evidence behind our programmes, tools, and findings on AI adoption in healthcare.
Evaluating Explainable Artificial Intelligence Techniques in Chest Radiology Imaging Through a Human-Centered Lens
A landmark evaluation of XAI techniques in chest radiology, conducted through surveys and assessments with 26 NHS medical professionals. The study revealed significant gaps in AI knowledge among clinicians and demonstrated that human-centred explainability is essential for trust and effective adoption of AI diagnostic tools within the NHS.
Read paper →Explained Deep Learning Framework for COVID-19 Detection in Volumetric CT Images
An explainable deep learning framework for detecting COVID-19 from volumetric CT images, developed in alignment with British Society of Thoracic Imaging guidance. Built in close partnership with NHS radiologists at Sandwell and West Birmingham NHS Trust, the framework demonstrates how responsible AI design and clinical co-development can produce trustworthy diagnostic tools for real clinical environments.
Read paper →Human-Centered User Interface Design for Explainable AI in Chest Radiology: A Multi-Phase Co-Design Approach
A multi-phase co-design study that engaged radiologists and clinical staff in designing a human-centered user interface for explainable AI in chest radiology. The research demonstrates the value of iterative participatory design methods in producing AI tools that clinicians can understand, interrogate, and trust in diagnostic practice.
Read paper →CHERIE: User-Centred Development of an XAI System for Chest Radiology Through Co-Design
Documents the full co-design process behind CHERIE, an explainable AI system built for chest radiology. Working with radiologists, medical physicists, and NHS clinical staff, the team used participatory workshops to ensure the system's outputs are interpretable and actionable. The project exemplifies the team's human-centred approach to responsible AI development in clinical settings.
Read paper →Explainable AI-Assisted Triage in Emergency Departments Using Multi-Modal Clinical Data
Investigates the application of explainable AI to assist clinical triage decisions in emergency departments, combining multi-modal clinical data sources. The study contributes to responsible AI deployment in high-stakes clinical environments by demonstrating how XAI can enhance clinician decision-making transparency, supporting safer and more accountable triage processes in busy NHS settings.
Read paper →