Objective
SAFE-MRI aims to prevent needless mortality from breast cancer by optimizing the use and performance of dedicated breast MRI.
First SAFE-MRI aims at detecting breast cancer early in women that are currently deemed too young for population screening (age 45-50). Early diagnosis of breast cancer in these women will lead to a large gain in DALYs averted. I will initiate a prospective screening study inviting women to a one-time only abbreviated breast MRI at the age of 45. The primary outcome will be the cancer detection rate per 1000 women actually screened. This will validate abbreviated breast MRI as standalone screening technique. Psycho-social and risk information will be collected from all participants. Cost-effectiveness modelling will be included. 5-year follow-up will be performed in all women up to, and including, the first screening round at the age of 50.
Second SAFE-MRI aims at improving breast MRI by reducing the need for contrast administration. This is a major downside of breast MRI - for women (iv-canulation and a small risk of allergic reactions) and the environment (pollution of surface water). First, I aim to enhance signal changes using a strongly reduced dose of contrast with advanced deep learning (DL). Further studies are tailored at assessment of unenhanced MRI. Particularly, I will validate novel diffusion weighted imaging (DWI) sequences capable of generating high resolution diagnostic images within short scan times using DL-based reconstruction. Furthermore, I will investigate arterial spin labelling (ASL) based breast imaging strategies that show perfusion without the need for contrast administration. Finally, we will combine unenhanced MRI techniques in a general adversarial network to construct synthetic breast MRI images. Repeated reader studies will be conducted to assess image quality and diagnostic value. SAFE-MRI will thus consolidate my leading role in breast imaging and enable me to design the breast screening of the future.
Fields of science (EuroSciVoc)
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
CORDIS classifies projects with EuroSciVoc, a multilingual taxonomy of fields of science, through a semi-automatic process based on NLP techniques.
- social sciencessociologydemographymortality
- medical and health sciencesclinical medicineoncology
- natural sciencesearth and related environmental sciencesenvironmental sciencespollution
You need to log in or register to use this function
Programme(s)
- HORIZON.1.1 - European Research Council (ERC) Main Programme
Funding Scheme
HORIZON-ERC - HORIZON ERC GrantsHost institution
6525 GA Nijmegen
Netherlands