Does an unsupervised deep learning ECG latent space model improve discrimination of prevalent disease compared to standard ECG intervals or demographic models?
Unsupervised deep learning of ECGs enables scalable human disease profiling with better discrimination than standard ECG intervals or demographic models.
). The ECG latent space model demonstrated more associations than models using standard ECG intervals, and offered favorable discrimination of prevalent disease compared to models comprising age, sex, and race. We further demonstrate how latent space models can be used to generate disease-specific ECG waveforms and facilitate individual disease profiling.
Friedman et al. (Sun,) studied this question.