MammoSim Study
Research type
Research Study
Full title
MammoSim: A Simulation-based study to assess radiologist user experience, usability, feasibility and acceptability of a novel artificial-intelligence derived mammography tool
IRAS ID
318024
Contact name
Ara Darzi
Contact email
Sponsor organisation
Imperial College London
Duration of Study in the UK
0 years, 9 months, 3 days
Research summary
The MammoSim study is an observational simulation-based study which aims to evaluate radiologist user experience of a novel Artificial Intelligence (AI) derived mammographic screening tool, Mammo-Reader. As part of the study, we will recruit readers who are naïve to Mammo-Reader and conduct a 3-part simulation-based study using anonymised screening mammograms from a curated retrospective dataset of patients who have historically attended mammographic screening, to ascertain the various primary and secondary aims of this study.
The MammoSim Consortium was established in 2021 and embodies a rigorous clinical-academic-industrial collaboration and partnership between Imperial College London, Imperial College Healthcare NHS Trust, St George’s Hospital London, OPTIMAM and Google Health Limited. OPTIMAM Mammography Image Database (Halling-Brown et al. 2021) is a large curated and centralised database of mammogram images and associated clinical data from multiple NHS Breast Cancer screening sites. Imaging and clinical data will be collected under existing ethical permissions for the OPTIMAM database (REC references 14/SC/0258 and 19/SC/0284), using existing infrastructure and technical methodology. Our team of radiologists, senior clinicians and scientists are representative experts in the fields of breast radiology and AI-derived technology. The MammoSim study aims to independently assess radiologist user experience, usability, feasibility and acceptability of a novel AI-derived mammography tool within a double-read workflow in a simulated environment. This simulation-study is being run in conjunction with the AIMS trial. As this a retrospective non-interventional study of historical anonymised mammograms for women who have previously undergone screening, this will not alter clinical outcomes nor introduce discrepancies in treatment.
REC name
London - Surrey Research Ethics Committee
REC reference
23/PR/0254
Date of REC Opinion
20 Mar 2023
REC opinion
Favourable Opinion