Machine Learning in Fabry disease version 1
Research type
Research Study
Full title
Development of a Machine Learning method to diagnose Fabry disease in Cardiology
IRAS ID
305043
Contact name
Duncan Cole
Contact email
Sponsor organisation
Cardiff and Vale UHB
Duration of Study in the UK
2 years, 0 months, 1 days
Research summary
Fabry disease is a rare inherited condition affecting many parts of the body, including the heart, kidneys and nerves. It is difficult to diagnose as the problems it causes look similar to lots of other conditions. We are aiming to develop a machine learning approach to diagnosis of Fabry disease using data routinely gathered in hospital cardiology clinics, including from blood tests, ECGs, and echocardiograms. This will use data from tests already completed, which will then be anonymised before it is used. One we have developed a machine learning program to do this, we will test the program by mixing data from patients with Fabry disease and those with another condition and seeing if the program can correctly identify the Fabry patients, and to compare this with a cardiologist's diagnosis.
REC name
N/A
REC reference
N/A