Skip to content


The MIDI (MR Imaging Abnormality Deep Learning Identification) study aims to develop and test an AI tool that can identify MR head abnormalities using routine outpatient as well as dedicated research scans.

Machine learning tools are being developed using a dataset of retrospective and prospective MRI head scans to train, validate and test at the Department of Biomedical Engineering, King’s College London. The reference standard will be consultant radiologist reports of the MRI head scans.

The study has over 30 recruiting centres and is one of the largest neuroimaging studies currently being conducted in the UK.