A machine learning cardiac magnetic resonance approach to extract disease features and automate pulmonary arterial hypertension diagnosis.

AIMS: Pulmonary arterial hypertension (PAH) is a progressive condition with high mortality. Quantitative cardiovascular magnetic resonance (CMR) imaging metrics in PAH target individual cardiac structures and have diagnostic and prognostic utility but are challenging to acquire. The primary aim of this study was to develop and test a tensor-based machine learning approach to holistically identify diagnostic features in PAH using CMR, and secondarily, visualize and interpret key discriminative features associated with PAH. METHODS AND RESULTS: Consecutive treatment naive patients with PAH or no evidence of pulmonary hypertension (PH), undergoing CMR and right heart catheterization within 48 h, were identified from the ASPIRE registry. A tensor-based machine learning approach, multilinear subspace learning, was developed and the diagnostic accuracy of this approach was compared with standard CMR measurements. Two hundred and twenty patients were identified: 150 with PAH and 70 with no PH. The diagnostic accuracy of the approach was high as assessed by area under the curve at receiver operating characteristic analysis (P < 0.001): 0.92 for PAH, slightly higher than standard CMR metrics. Moreover, establishing the diagnosis using the approach was less time-consuming, being achieved within 10 s. Learnt features were visualized in feature maps with correspondence to cardiac phases, confirming known and also identifying potentially new diagnostic features in PAH. CONCLUSION: A tensor-based machine learning approach has been developed and applied to CMR. High diagnostic accuracy has been shown for PAH diagnosis and new learnt features were visualized with diagnostic potential.

CT derived left atrial size identifies left heart disease in suspected pulmonary hypertension: Derivation and validation of predictive thresholds.

BACKGROUND: Patients with pulmonary hypertension due to left heart disease (PH-LHD) have overlapping clinical features with pulmonary arterial hypertension making diagnosis reliant on right heart catheterization (RHC). This study aimed to investigate computed tomography pulmonary angiography (CTPA) derived cardiopulmonary structural metrics, in comparison to magnetic resonance imaging (MRI) for the diagnosis of left heart disease in patients with suspected pulmonary hypertension. METHODS: Patients with suspected pulmonary hypertension who underwent CTPA, MRI and RHC were identified. Measurements of the cardiac chambers and vessels were recorded from CTPA and MRI. The diagnostic thresholds of individual measurements to detect elevated pulmonary arterial wedge pressure (PAWP) were identified in a derivation cohort (n = 235). Individual CT and MRI derived metrics were tested in validation cohort (n = 211). RESULTS: 446 patients, of which 88 had left heart disease. Left atrial area was a strong predictor of elevated PAWP>15 mm Hg and PAWP>18 mm Hg, area under curve (AUC) 0.854, and AUC 0.873 respectively. Similar accuracy was also identified for MRI derived LA volume, AUC 0.852 and AUC 0.878 for PAWP > 15 and 18 mm Hg, respectively. Left atrial area of 26.8 cm2 and 30.0 cm2 were optimal specific thresholds for identification of PAWP > 15 and 18 mm Hg, had sensitivity of 60%/53% and specificity 89%/94%, respectively in a validation cohort. CONCLUSIONS: CTPA and MRI derived left atrial size identifies left heart disease in suspected pulmonary hypertension with high specificity. The proposed diagnostic thresholds for elevated left atrial area on routine CTPA may be a useful to indicate the diagnosis of left heart disease in suspected pulmonary hypertension.