A deep neural network electrocardiogram analysis framework for left ventricular hypertrophy prediction

Heart disease is ranked second in the top ten death causes in Taiwan in 2016, while the amount of heart disease fatalities is increasing by about 700 individuals each year. Therefore, it is urgent to improve heart disease diagnostic accuracy. This research combines profound neural networks and information from the ECG to create an ECG classifier for left ventricular hypertrophy. The classification system uses data preprocessing to impute missing values of ECG data, which can be collected from IoT equipment and monitor the body situation quickly, and transform these data for use by the deep neural network. Then, the data is input into a six-layers deep neural network designed for this study and is used to predict left ventricular hypertrophy. L2-regularization and dropout are used to avoid overfitting in each model iteration. The system has been named the electrocardiographic left ventricular hypertrophy classifier (ELVHC). The experimental findings demonstrate that the forecast model’s precision is 73%, the sensitivity is 66%, and the specificity is 78%, which is considerably greater than two clinical techniques.