Quantitative Imaging Assessment of Pulmonary Vasculature and Parenchyma as A Biomarker in Pulmonary Hypertension: Using CTPA and Machine Learning (QVAC Study)

Objectives: 1. Use and validate novel machine learning algorithms to investigate and quantitatively assess key computed tomography pulmonary angiogram (CTPA) morphologic features of the pulmonary vasculature and parenchyma in pulmonary hypertension which correlates with right heart haemodynamic parameters, known biochemical biomarkers, and functional assessment. 2. Develop a novel risk stratification model/scoring, based on computed tomography pulmonary angiogram (CTPA) imaging biomarkers, with input from machine learning platform, to guide assessment of pulmonary hypertension in order to aid management. Outcomes: 1. Determine key computed tomography pulmonary angiogram (CTPA) imaging biomarkers in correlation with clinical and hemodynamic features, in detecting, describing, and diagnosing pulmonary hypertension with the aid of a machine learning algorithms). 2. Imaging biomarkers to generate risk stratification in aiding diagnosis and patient management with pulmonary hypertension. 3. Machine learning based evaluation of CTPA vascular and parenchymal remodelling to differentiate PH aetiologies/phenotypes.