Multicenter validation of a machine learning phase space electro-mechanical pulse wave analysis to predict elevated left ventricular end diastolic pressure at the point-of-care

Abstract

Background Phase space is a mechanical systems approach and large-scale data representation of an object in 3-dimensional space. Whether such techniques can be applied to predict left ventricular pressures non-invasively and at the point-of-care is unknown. Objective This study prospectively validated a phase space machine-learned approach based on a novel electro-mechanical pulse wave method of data collection through orthogonal voltage gradient (OVG) and photoplethysmography (PPG) for the prediction of elevated left ventricular end diastolic pressure (LVEDP). Methods Consecutive outpatients across 15 US-based healthcare centers with symptoms suggestive of coronary artery disease were enrolled at the time of elective cardiac catheterization and underwent OVG and PPG data acquisition immediately prior to angiography with signals paired with LVEDP (IDENTIFY; NCT #03864081). The primary objective was to validate a ML algorithm for prediction of elevated LVEDP using a definition of ≥25 mmHg (study cohort) and normal LVEDP ≤ 12 mmHg (control cohort), using AUC as the measure of diagnostic accuracy. Secondary objectives included performance of the ML predictor in a propensity matched cohort (age and gender) and performance for an elevated LVEDP across a spectrum of comparative LVEDP (<12 through 24 at 1 mmHg increments). Features were extracted from the OVG and PPG datasets and were analyzed using machine-learning approaches. Results The study cohort consisted of 684 subjects stratified into three LVEDP categories, ≤12 mmHg (N = 258), LVEDP 13–24 mmHg (N = 347), and LVEDP ≥25 mmHg (N = 79). Testing of the ML predictor demonstrated an AUC of 0.81 (95% CI 0.76–0.86) for the prediction of an elevated LVEDP with a sensitivity of 82% and specificity of 68%, respectively. Among a propensity matched cohort (N = 79) the ML predictor demonstrated a similar result AUC 0.79 (95% CI: 0.72–0.8). Using a constant definition of elevated LVEDP and varying the lower threshold across LVEDP the ML predictor demonstrated and AUC ranging from 0.79–0.82. Conclusion The phase space ML analysis provides a robust prediction for an elevated LVEDP at the point-of-care. These data suggest a potential role for an OVG and PPG derived electro-mechanical pulse wave strategy to determine if LVEDP is elevated in patients with symptoms suggestive of cardiac disease.

Pulmonary Hypertension Detection Non-Invasively at Point-of-Care Using a Machine-Learned Algorithm

“…The PH algorithm is the series of processing steps to take in a signal from the CorVista Capture device and return a prediction reflective of PH status (i.e., PH Score). The development process began with assessment of the quality of captured signal, then feature extraction from OVG and PPG signals [15], followed by univariate feature selection to identify discriminative features. Statistical tests were employed to retain only the significant features, reducing the dimensionality of the dataset.…”

“…Several techniques have been employed for feature engineering, such as spectral, scalogram, time-series, dynamical and topological analysis. Features have previously demonstrated utility in the assessment of CAD [12,15] and elevated left ventricular end diastolic pressure [15,16]. Detailed description of the features' calculation and their reported utility can be found in Supplement Section S2.…”

“…Arterial compliance features employ the first or second derivative of the PPG (i.e., velocity plethysmogram and acceleration plethysmogram), both of which are known to embed characteristics of arterial compliance [25]. 'Perfusion response to cardiac contraction' features characterize the interplay between the OVG signal and the PPG signal, therefore embedding the perfusion response to cardiac pulsation [15]. Atrial structure features capture heterogeneity in atrial composition, including atrial enlargement [26].…”

Identifying Novel Phenotypes of Elevated Left Ventricular End Diastolic Pressure Using Hierarchical Clustering of Features Derived From Electromechanical Waveform Data

“…An algorithm was previously developed to predict LVEDP elevation status based on manually engineered features calculated from CorVista Capture signals (22). The signal acquisition modality was previously described ( 26), but briefly, is the simultaneous acquisition of orthogonal voltage gradient (OVG) data via electrodes placed on the torso, and photoplethysmogram using transmission of red and infrared light via a clip placed on the finger.…”

“…Predicting outcomes in those with known moderate to severe disease allows modification and testing of current therapies but provides little information that might permit interventions at early stages of heart failure to prevent progression. Machine learning affords the opportunity with a single rapid test to detected features of similarities between groups with early stage LVEDP elevation and those not yet manifesting hemodynamic changes (22). ML clustering techniques that identify such features facilitate the development and future validation of new risk models.…”