Artificial Intelligence in Mammography Screening (AIMS) Part C

  • Research type

    Research Study

  • Full title

    Feasibility of an artificial intelligence system to improve the quality and efficiency of breast cancer screening

  • IRAS ID

    307842

  • Contact name

    Ara Darzi

  • Contact email

    a.darzi@imperial.ac.uk

  • Duration of Study in the UK

    1 years, 5 months, 31 days

  • Research summary

    The objective of this study is to demonstrate feasibility for the technical integration of an artificial intelligence (AI) system into the standard clinical workflow. This is the third part (Part C) of a multi-stage project which overall aims to evaluate the potential for AI-enabled NHS breast screening to increase accuracy, safety, cost-effectiveness, and clinician/patient experience, while demonstrating evidence of clinical and technical feasibility.

    In the UK breast screening programme, two expert readers (radiologists) assess each mammogram (x-ray of the breast), with any disagreements in opinion reviewed by an arbitration panel of two further readers. However, a radiologist workforce crisis threatens the screening programme’s long-term sustainability. Google’s AI system identified cancer in mammograms with greater accuracy than specialists (Nature, 2020), suggesting potential to: reduce radiologist workload; increase service capacity; improve accuracy and outcomes, and reduce variability; and reduce time to results, improving patient experience. This project brings this initial research towards real world impact.

    Interventional use of the AI system within the health system cannot be commenced before comprehensive assessments of feasibility and safety, alongside modelling of likely clinical, workflow, and economic impacts. This proposal outlines our plans to translate this research towards real world patient impact through a prospective non-interventional feasibility study that will test the AI system running ‘silently’ within breast screening clinics at two NHS sites.

    Together with planned retrospective diagnostic accuracy studies (Part A) and large scale consensus panel reader studies (Part B), and this prospective observational feasibility study (Part C), we will design an AI integration strategy for each site, and aim to provide the necessary evidence to form the basis for a proposed deployment framework to support progression to future interventional use in a way that delivers measurable benefits to public health.

  • REC name

    East Midlands - Nottingham 1 Research Ethics Committee

  • REC reference

    22/EM/0198

  • Date of REC Opinion

    22 Sep 2022

  • REC opinion

    Favourable Opinion