Unsupervised feature extraction using deep learning empowers discovery of genetic determinants of the electrocardiogram
Sieliwonczyk, E.; Sau, A.; Patlatzoglou, K.; McGurk, K. A.; Pastika, L.; Thami, P.; Mangino, M.; Zheng, S. L.; Powell, G.; Curran, L.; Buchan, R.; Theotokis, P.; Peters, N. S.; Loeys, B.; Kramer, D. B.; Waks, J. W.; Ng, F. S.; Ware, J. S.
Show abstract
Advanced data-driven methods can outperform conventional features in electrocardiogram (ECG) analysis, but often lack interpretability. The variational autoencoder (VAE), a form of unsupervised machine learning, can address this shortcoming by extracting comprehensive and interpretable new ECG features. Our novel VAE model, trained on a dataset comprising over one million secondary care median beat ECGs, and validated using the UK Biobank, reveals 20 independent features that capture ECG information content with high reconstruction accuracy. Through phenome- and genome-wide association studies, we illustrate the increased power of the VAE approach for gene discovery, compared with conventional ECG traits, and identify previously unrecognised common and rare variant determinants of ECG morphology. Additionally, to highlight the interpretability of the model, we provide detailed visualisation of the associated ECG alterations. Our study shows that the VAE provides a valuable tool for advancing our understanding of cardiac function and its genetic underpinnings.
Matching journals
The top 4 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- Simple Models Versus Deep Learning in Detecting Low Ejection Fraction From The Electrocardiogram 96%
- Development and Multinational Validation of an Ensemble Deep Learning Algorithm for Detecting and Predicting Structural Heart Disease Using Noisy Single-lead Electrocardiograms 96%
- Clinical and genetic associations of asymmetric apical and septal left ventricular hypertrophy 95%
Similar papers in this journal
Similar papers in this journal
- The autism-associated Meis2 gene is necessary for cardiac baroreflex regulation in mice 94%
- Altered Intercellular Communication and Extracellular Matrix Signaling as a Potential Disease Mechanism in Human Hypertrophic Cardiomyopathy 94%
- In Silico Evaluation of Cell Therapy in Acute versus Chronic Infarction: Role of Automaticity, Heterogeneity and Purkinje in Human 93%
Similar papers in this journal
- Biomarker panels for improved risk prediction and enhanced biological insights in patients with atrial fibrillation 95%
- Genome-wide association analysis and Mendelian randomization proteomics identify novel protein biomarkers and drug targets for primary prevention of heart failure 95%
- Tissue-specific multiOMICs analysis of atrial fibrillation 94%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.