Patient opinions on AI use in breast cancer prediction (Version 1)
Research type
Research Study
Full title
Assessing patient opinions on the use of artificial intelligence in predicting breast cancer outcome
IRAS ID
355098
Contact name
Valerie Speirs
Contact email
Sponsor organisation
University of Aberdeen
Clinicaltrials.gov Identifier
OSF registration, https://doi.org/10.17605/OSF.IO/X9SGZ
Duration of Study in the UK
3 years, 0 months, 0 days
Research summary
This qualitative research study aims to explore the attitudes of breast cancer patients towards the use of artificial intelligence as part of their diagnostic workup. This study is part of a larger machine learning study where we aim to develop an AI tool to predict patients who may suffer early recurrence of their breast cancer and therefore would benefit from closer monitoring and earlier treatment. We are particularly interested in triple negative breast cancer (TNBC). TNBC is the most aggressive form of breast cancer, which lacks drug-targetable receptors, limiting treatment options to chemotherapy. This treatment is effective in many patients but around 40% of women have a recurrence of their cancer early on and survival is poor, often less than one year. Previous work has shown that haemtoxylin and eosin (H&E)-stained sections of breast tissue can be used in a machine learning model to extract features of importance that could be used to provide a more accurate and immediate assessment of TNBC aggressiveness which may not be otherwise apparent with standard diagnostic models. This approach is perceived by many to be a “black box”, so we want to understand patient thoughts and feelings on the use of AI in their breast cancer care journey. Understanding their thoughts and feelings on the use of AI in breast cancer care will allow us to tailor how the AI algorithm would be implemented into clinical care.
REC name
North West - Liverpool Central Research Ethics Committee
REC reference
25/NW/0345
Date of REC Opinion
2 Dec 2025
REC opinion
Favourable Opinion