Researchers have developed a neural network approach which they say can accurately identify congestive heart failure (CHF) with 100 per cent accuracy through analysis of just one raw electrocardiogram (ECG) heartbeat.
The researchers trained and tested a Convolutional Neural Network (CNN) model on large publicly available ECG data sets, comprising a total of 490,505 heartbeats, to detect CHF. The data set included subjects with CHF as well as healthy, non-arrhythmic hearts.
By checking just one heartbeat, the researchers were able to detect whether a person had heart failure. The authors noted that the model also identifies heartbeat sequences and ECG morphological characteristics, which are class-discriminative and prominent for CHF detection.
They say the research drastically improves existing CHF detection methods typically focused on heart rate variability that, whilst effective, are time-consuming and prone to errors.
“Overall, our contribution substantially advances the current methodology for detecting CHF and caters to clinical practitioners’ needs by providing an accurate and fully transparent tool to support decisions concerning CHF detection,” the authors wrote in Biomedical Signal Processing and Control Journal.
They have called for future works "to expand the results presented in this study towards improving human wellbeing and reducing current healthcare financial and societal burdens."