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.

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.

Identifying novel phenotypes of elevated left ventricular end diastolic pressure using hierarchical clustering of features derived from electromechanical waveform data.

Introduction: Elevated left ventricular end diastolic pressure (LVEDP) is a consequence of compromised left ventricular compliance and an important measure of myocardial dysfunction. An algorithm was developed to predict elevated LVEDP utilizing electro-mechanical (EM) waveform features. We examined the hierarchical clustering of selected features developed from these EM waveforms in order to identify important patient subgroups and assess their possible prognostic significance. Materials and methods: Patients presenting with cardiovascular symptoms (N = 396) underwent EM data collection and direct LVEDP measurement by left heart catheterization. LVEDP was classified as non-elevated ( ≤ 12 mmHg) or elevated (≥25 mmHg). The 30 most contributive features to the algorithm output were extracted from EM data and input to an unsupervised hierarchical clustering algorithm. The resultant dendrogram was divided into five clusters, and patient metadata overlaid. Results: The cluster with highest LVEDP (cluster 1) was most dissimilar from the lowest LVEDP cluster (cluster 5) in both clustering and with respect to clinical characteristics. In contrast to the cluster demonstrating the highest percentage of elevated LVEDP patients, the lowest was predominantly non-elevated LVEDP, younger, lower BMI, and males with a higher rate of significant coronary artery disease (CAD). The next adjacent cluster (cluster 2) to that of the highest LVEDP (cluster 1) had the second lowest LVEDP of all clusters. Cluster 2 differed from Cluster 1 primarily based on features extracted from the electrical data, and those that quantified predictability and variability of the signal. There was a low predictability and high variability in the highest LVEDP cluster 1, and the opposite in adjacent cluster 2. Conclusion: This analysis identified subgroups of patients with varying degrees of LVEDP elevation based on waveform features. An approach to stratify movement between clusters and possible progression of myocardial dysfunction may include changes in features that differentiate clusters; specifically, reductions in electrical signal predictability and increases in variability. Identification of phenotypes of myocardial dysfunction evidenced by elevated LVEDP and knowledge of factors promoting transition to clusters with higher levels of left ventricular filling pressures could permit early risk stratification and improve patient selection for novel therapeutic interventions.

Development and validation of a machine learned algorithm to IDENTIFY functionally significant coronary artery disease.

Introduction: Multiple trials have demonstrated broad performance ranges for tests attempting to detect coronary artery disease. The most common test, SPECT, requires capital-intensive equipment, the use of radionuclides, induction of stress, and time off work and/or travel. Presented here are the development and clinical validation of an office-based machine learned algorithm to identify functionally significant coronary artery disease without radiation, expensive equipment or induced patient stress. Materials and methods: The IDENTIFY trial (NCT03864081) is a prospective, multicenter, non-randomized, selectively blinded, repository study to collect acquired signals paired with subject meta-data, including outcomes, from subjects with symptoms of coronary artery disease. Time synchronized orthogonal voltage gradient and photoplethysmographic signals were collected for 230 seconds from recumbent subjects at rest within seven days of either left heart catheterization or coronary computed tomography angiography. Following machine learning on a proportion of these data (N = 2,522), a final algorithm was selected, along with a pre-specified cut point on the receiver operating characteristic curve for clinical validation. An unseen set of subject signals (N = 965) was used to validate the algorithm. Results: At the pre-specified cut point, the sensitivity for detecting functionally significant coronary artery disease was 0.73 (95% CI: 0.68-0.78), and the specificity was 0.68 (0.62-0.74). There exists a point on the receiver operating characteristic curve at which the negative predictive value is the same as coronary computed tomographic angiography, 0.99, assuming a disease incidence of 0.04, yielding sensitivity of 0.89 and specificity of 0.42. Selecting a point at which the positive predictive value is maximized, 0.12, yields sensitivity of 0.39 and specificity of 0.88. Conclusion: The performance of the machine learned algorithm presented here is comparable to common tertiary center testing for coronary artery disease. Employing multiple cut points on the receiver operating characteristic curve can yield the negative predictive value of coronary computed tomographic angiography and a positive predictive value approaching that of myocardial perfusion imaging. As such, a system employing this algorithm may address the need for a non-invasive, no radiation, no stress, front line test, and hence offer significant advantages to the patient, their physician, and healthcare system.

Predicting cardiac disease from interactions of simultaneously-acquired hemodynamic and cardiac signals.

BACKGROUND AND OBJECTIVE: Coronary artery disease (CAD) and heart failure are the most common cardiovascular diseases. Non-invasive diagnostic testing for CAD requires radiation, heart rate acceleration, and imaging infrastructure. Early detection of left ventricular dysfunction is critical in heart failure management, the best measure of which is an elevated left ventricular end-diastolic pressure (LVEDP) that can only be measured using invasive cardiac catheterization. There exists a need for non-invasive, safe, and fast diagnostic testing for CAD and elevated LVEDP. This research employs nonlinear dynamics to assess for significant CAD and elevated LVEDP using non-invasively acquired photoplethysmographic (PPG) and three-dimensional orthogonal voltage gradient (OVG) signals. PPG (variations of the blood volume perfusing the tissue) and OVG (mechano-electrical activity of the heart) signals represent the dynamics of the cardiovascular system. METHODS: PPG and OVG were simultaneously acquired from two cohorts, (i) symptomatic subjects that underwent invasive cardiac catheterization, the gold standard test (408 CAD positive with stenosis≥ 70% and 186 with LVEDP≥ 20 mmHg) and (ii) asymptomatic healthy controls (676). A set of Poincaré-based synchrony features were developed to characterize the interactions between the OVG and PPG signals. The extracted features were employed to train machine learning models for CAD and LVEDP. Five-fold cross-validation was used and the best model was selected based on the average area under the receiver operating characteristic curve (AUC) across 100 runs, then assessed using a hold-out test set. RESULTS: The Elastic Net model developed on the synchrony features can effectively classify CAD positive subjects from healthy controls with an average validation AUC=0.90±0.03 and an AUC= 0.89 on the test set. The developed model for LVEDP can discriminate subjects with elevated LVEDP from healthy controls with an average validation AUC=0.89±0.03 and an AUC=0.89 on the test set. The feature contributions results showed that the selection of a proper registration point for Poincaré analysis is essential for the development of predictive models for different disease targets. CONCLUSIONS: Nonlinear features from simultaneously-acquired signals used as inputs to machine learning can assess CAD and LVEDP safely and accurately with an easy-to-use, portable device, utilized at the point-of-care without radiation, contrast, or patient preparation.

Machine learning: at the heart of failure diagnosis.

PURPOSE OF REVIEW: Refinement in machine learning (ML) techniques and approaches has rapidly expanded artificial intelligence applications for the diagnosis and classification of heart failure (HF). This review is designed to provide the clinician with the basics of ML, as well as this technologies future utility in HF diagnosis and the potential impact on patient outcomes. RECENT FINDINGS: Recent studies applying ML methods to unique data sets available from electrocardiography, vectorcardiography, echocardiography, and electronic health records show significant promise for improving diagnosis, enhancing detection, and advancing treatment of HF. Innovations in both supervised and unsupervised methods have heightened the diagnostic accuracy of models developed to identify the presence of HF and further augmentation of model capabilities are likely utilizing ensembles of ML algorithms derived from different techniques. SUMMARY: This article is an overview of recent applications of ML to achieve improved diagnosis of HF and the resultant implications for patient management.

Cardiac Phase Space Tomography: A novel method of assessing coronary artery disease utilizing machine learning.

BACKGROUND: Artificial intelligence (AI) techniques are increasingly applied to cardiovascular (CV) medicine in arenas ranging from genomics to cardiac imaging analysis. Cardiac Phase Space Tomography Analysis (cPSTA), employing machine-learned linear models from an elastic net method optimized by a genetic algorithm, analyzes thoracic phase signals to identify unique mathematical and tomographic features associated with the presence of flow-limiting coronary artery disease (CAD). This novel approach does not require radiation, contrast media, exercise, or pharmacological stress. The objective of this trial was to determine the diagnostic performance of cPSTA in assessing CAD in patients presenting with chest pain who had been referred by their physician for coronary angiography. METHODS: This prospective, multicenter, non-significant risk study was designed to: 1) develop machine-learned algorithms to assess the presence of CAD (defined as one or more ≥ 70% stenosis, or fractional flow reserve ≤ 0.80) and 2) test the accuracy of these algorithms prospectively in a naïve verification cohort. This report is an analysis of phase signals acquired from 606 subjects at rest just prior to angiography. From the collective phase signal data, features were extracted and paired with the known angiographic results. A development set, consisting of signals from 512 subjects, was used for machine learning to determine an algorithm that correlated with significant CAD. Verification testing of the algorithm was performed utilizing previously untested phase signals from 94 subjects. RESULTS: The machine-learned algorithm had a sensitivity of 92% (95% CI: 74%-100%) and specificity of 62% (95% CI: 51%-74%) on blind testing in the verification cohort. The negative predictive value (NPV) was 96% (95% CI: 85%-100%). CONCLUSIONS: These initial multicenter results suggest that resting cPSTA may have comparable diagnostic utility to functional tests currently used to assess CAD without requiring cardiac stress (exercise or pharmacological) or exposure of the patient to radioactivity.