Deep neural heart rate variability analysis
This work addresses the need for non-invasive and accurate screening tools for cardiovascular disease in clinical practice, offering a potential alternative to blood tests.
The paper tackled the problem of early cardiovascular disease recognition by developing a deep learning model for heart rate variability analysis, which significantly outperformed traditional predictors and approached or exceeded the accuracy of clinical blood tests using only 60 seconds of inter-beat intervals.
Despite of the pain and limited accuracy of blood tests for early recognition of cardiovascular disease, they dominate risk screening and triage. On the other hand, heart rate variability is non-invasive and cheap, but not considered accurate enough for clinical practice. Here, we tackle heart beat interval based classification with deep learning. We introduce an end to end differentiable hybrid architecture, consisting of a layer of biological neuron models of cardiac dynamics (modified FitzHugh Nagumo neurons) and several layers of a standard feed-forward neural network. The proposed model is evaluated on ECGs from 474 stable at-risk (coronary artery disease) patients, and 1172 chest pain patients of an emergency department. We show that it can significantly outperform models based on traditional heart rate variability predictors, as well as approaching or in some cases outperforming clinical blood tests, based only on 60 seconds of inter-beat intervals.