Key result
An automated machine learning pipeline using patient-level ECG profiles accurately detected pulmonary arterial hypertension, hypertrophic cardiomyopathy, and cardiac amyloidosis.
Why the study?
Advances in computing, machine learning, and large-scale data may expand clinical inferences from the ECG while preserving interpretability for medical decision-making.
Do machine learning models using patient-level ECG profiles accurately estimate cardiac structure and detect cardiovascular diseases compared to standard clinical measurements?
Observational (n=36,186)
No
Do machine learning models using patient-level ECG profiles accurately estimate cardiac structure and detect cardiovascular diseases compared to standard clinical measurements?
Machine learning models applied to standard 12-lead ECGs can accurately estimate cardiac structure and detect specific cardiovascular diseases such as pulmonary arterial hypertension and hypertrophic cardiomyopathy.
Answers are grounded in this paper and cite their sources.
May aid ECG-based detection of PAH, HCM, and amyloid; leaves open prospective validation before clinical adoption.
BACKGROUND: The ECG remains the most widely used diagnostic test for characterization of cardiac structure and electrical activity. We hypothesized that parallel advances in computing power, machine learning algorithms, and availability of large-scale data could substantially expand the clinical inferences derived from the ECG while at the same time preserving interpretability for medical decision-making. METHODS AND RESULTS: We identified 36 186 ECGs from the University of California, San Francisco database that would enable training of models for estimation of cardiac structure or function or detection of disease. We segmented the ECG into standard component waveforms and intervals using a novel combination of convolutional neural networks and hidden Markov models and evaluated this segmentation by comparing resulting electrical intervals against 141 864 measurements produced during the clinical workflow. We then built a patient-level ECG profile, a 725-element feature vector and used this profile to train and interpret machine learning models for examples of cardiac structure (left ventricular mass, left atrial volume, and mitral annulus e-prime) and disease (pulmonary arterial hypertension, hypertrophic cardiomyopathy, cardiac amyloid, and mitral valve prolapse). ECG measurements derived from the convolutional neural network-hidden Markov model segmentation agreed with clinical estimates, with median absolute deviations as a fraction of observed value of 0.6% for heart rate and 4% for QT interval. Models trained using patient-level ECG profiles enabled surprising quantitative estimates of left ventricular mass and mitral annulus e' velocity (median absolute deviation of 16% and 19%, respectively) with good discrimination for left ventricular hypertrophy and diastolic dysfunction as binary traits. Model performance using our approach for disease detection demonstrated areas under the receiver operating characteristic curve of 0.94 for pulmonary arterial hypertension, 0.91 for hypertrophic cardiomyopathy, 0.86 for cardiac amyloid, and 0.77 for mitral valve prolapse. CONCLUSIONS: Modern machine learning methods can extend the 12-lead ECG to quantitative applications well beyond its current uses while preserving the transparency that is so fundamental to clinical care.
No takes yet. Share an insight, caveat, or question.
Tison et al. (2019) conducted an observational in Normal sinus rhythm (n=36,186). ecgAI (CNN-HMM derived ECG segmentation and patient-level ECG profiles) vs. Clinical estimates (MUSE/echocardiogram) was evaluated on Detection of pulmonary arterial hypertension (PAH) (AUROC 0.94, 95% CI 0.93-0.95). An automated machine learning pipeline using patient-level ECG profiles accurately detected pulmonary arterial hypertension, hypertrophic cardiomyopathy, and cardiac amyloidosis.
Synapse has enriched 10 related papers. Consider them for comparative context: