Development of Advanced Cardiovascular Imaging Methods

Assadi, Hosamadin Sami (2024) Development of Advanced Cardiovascular Imaging Methods. Doctoral thesis, University of East Anglia.

PDF Restricted to Repository staff only until 31 December 2026. Request a copy

Abstract

Cardiovascular magnetic resonance imaging (CMR) is an established non-invasive imaging modality for the diagnosis, management and prognosis of cardiovascular disease. However, comprehensive assessment from cine images remains challenging mainly due to lengthy and cumbersome image acquisition and analysis. The research described in this thesis involves developing and validating advanced cardiovascular imaging techniques that integrate and combine a series of state-of-the-art methods. The first contribution of the thesis is the development and evaluation of the feasibility and accuracy of an automated four-dimensional (4D) flow CMR model for estimating peak mitral inflow diastolic velocities. The proposed method is comparable to Doppler echocardiography and has excellent repeatability for clinical use.

The thesis's main contribution is developing and validating a deep learning artificial intelligence (AI) model of a comprehensive time-resolved segmentation of all four heart chambers using multicentre and multivendor datasets. The proposed AI segmentation technique is rapid, feasible, accurate, reproducible and has the same prognostic value as manual annotations in real-world CMR. We then validated a single plane method against a biplane method to quantify left atrial (LA) volume and function. The single plane method demonstrated significant correlation and superior accuracy for diagnosing dilated LA.

We subsequently shifted the research focus to developing CMR mathematical and physiological models using our fully automated segmentation model. The first was to estimate systolic and diastolic blood pressures using patient characteristics and CMR functional and flow data. The proposed models can help clinicians do an afterload assessment during CMR imaging and aid their management plan for complex diseases. The second was a novel and simple CMR model to predict functional heart age. In the same study, we describe in detail the structural and functional alterations associated with accelerated cardiovascular ageing in all four heart chambers in health and disease in different decades of life.

Item Type:	Thesis (Doctoral)
Faculty \ School:	Faculty of Medicine and Health Sciences > Norwich Medical School
Depositing User:	Chris White
Date Deposited:	03 Dec 2025 10:07
Last Modified:	03 Dec 2025 10:07
URI:	https://ueaeprints.uea.ac.uk/id/eprint/101206
DOI:

Actions (login required)

View Item