RSET projects

Using artificial intelligence in chest diagnostics for lung disease

Using AI to diagnose disease may help with accurate diagnosis, reduce errors, and improve efficiency, but evidence on the implementation and use of AI in real-world settings is limited. This RSET evaluation is funded and commissioned by NIHR as part of the national AI Diagnostic Fund aiming to accelerate the deployment of promising AI tools to help diagnose patients more quickly.

Full study title: Mixed-method evaluation of implementing artificial intelligence in chest diagnostics for lung disease

Evidence suggests that implementing artificial intelligence (AI) in diagnostics may contribute to accurate diagnosis, reduce errors, and improve efficiency. However, there is limited evidence on implementation and use of AI in real-world settings, including staff experiences, patient and carer experience, effectiveness, and costs.

In 2023, NHS England launched a pilot scheme to support the introduction of AI for chest x-rays and chest CT scans at selected NHS hospitals across England, where AI is to support specialists in making treatment decisions.

The NIHR Rapid Service Evaluation Team (RSET) will evaluate early deployment and implementation of AI for chest diagnostics as part of AIDF, to explore factors influencing implementation, and identify settings and data sources for a potential phase 2 evaluation and/or future longer-term evaluations.

This is the first phase of a planned two-phase evaluation. Our findings will both inform a Phase 2 evaluation and/or any future longer-term evaluations.

Our research questions

  1. How can we best collect patient and public perceptions of using AI diagnostic tools in clinical practice?
  2. How can services best measure the impact of AI deployment on patients and the clinical pathway?
  3. How can we best measure the costs and resources involved in using AI tools for chest diagnostics?
  4. How is AI being used to support analysis and reporting of chest x-rays and chest CT scans?
  5. What do healthcare staff and AI suppliers think about these processes?
  6. What helps and gets in the way of using AI to look at chest x-rays and chest CT scans? (including any impacts on equalities, diversity, and inclusion)?

Our approach

To address our research questions, we will:

  • conduct a rapid scoping review followed by stakeholder consultation discussions (RQ1-3).
  • use stakeholder interviews, non-participant observations of oversight meetings, and analysis of relevant planning and progress documents.

Project team

  • Steve Morris
  • Angus Ramsay
  • Naomi Fulop
  • Pei Li Ng
  • Chris Sherlaw-Johnson
  • Nadia Crellin
  • Raj Mehta
  • Holly Walton
  • Efthalia Massou
  • Kevin Herbert
  • Emily Slade
  • Joanne Lloyd
  • Rachel Lawrence
  • Emma Dodsworth
  • Holly Elphinstone

To find out more, contact Principal Investigator, Angus Ramsay: angus.ramsay@ucl.ac.uk