Development of a Model to Identify Cardiac Hypertrophy
Research type
Research Study
Full title
Development of an Artificial Intelligence Model for Differentiation of Pathological vs. Physiological Left Ventricular Hypertrophy (AID-LVH)
IRAS ID
353903
Contact name
Ify Mordi
Contact email
Sponsor organisation
University of Dundee
Duration of Study in the UK
2 years, 5 months, 31 days
Research summary
Thickening of the left ventricle of the heart, known as left ventricular hypertrophy (LVH) is a relatively common finding, affecting around 10-20% of the general population. Diagnosis of LVH is important as it is independently associated with an increased risk of cardiovascular events including hospital admissions and death (~2.5x increase in cardiovascular events compared to individuals without LVH).
It is important not only to diagnose the presence of LVH, but also to identify the cause to allow the appropriate management for the patient. Sometimes LVH can be benign and the patient can be reassured, but in other situations LVH can even be life-threatening and require specific treatments. Current methods for diagnosis are however inaccurate and even experienced clinicians can be left unclear, leading to uncertainty for patients and family members.
We now routinely use cardiac MRI scans (CMR) for assessment of LVH, which has helped increase diagnostic accuracy, however there is still a significant proportion of scans where the diagnosis is uncertain. Studies have previously demonstrated that AI can be used to accurately measure cardiac parameters from CMR. The next step is to develop an AI model that could analyse CMR scans and accurately identify and differentiate causes of LVH.
The overall aim of this study is to develop an AI algorithm using machine learning techniques that can accurately identify causes of LVH from routinely-performed clinical CMR scans.
REC name
East of Scotland Research Ethics Service REC 2
REC reference
25/ES/0091
Date of REC Opinion
7 Nov 2025
REC opinion
Favourable Opinion