Diagnosing Cardiac Abnormalities from 12-Lead Electrocardiograms Using Enhanced Deep Convolutional Neural Networks

We train an enhanced deep convolutional neural network in order to identify eight cardiac abnormalities from the standard 12-lead electrocardiograms (ECGs) using the dataset of 14000 ECGs. Instead of straightforwardly applying an end-to-end deep learning approach, we find that deep convolutional neural networks enhanced with sophisticated hand crafted features show advantages in reducing generalization errors. Additionally, data preprocessing and augmentation are essential since the distribution of eight cardiac abnormalities are highly biased in the given dataset. Our approach achieves promising generalization performance in the First China ECG Intelligent Competition; an empirical evaluation is also provided to validate the efficacy of our design on the competition ECG dataset. Keywords: Electrocardiogram · Deep Convolutional Neural Network · Heart Disease Diagnosis. 1 Introduction The electrocardiogram (ECG) is a diagnostic tool widely utilized for noninvasive diagnosis of various cardiovascular abnormalities in practice of clinical medicine worldwide. For example, there are approximately 250 million ECG recordings being processed by technicians for the diagnosis and treatment of patients with cardiovascular disease in China. The standard 12-lead electrocardiograms are the records of the heart's electrical activity collected from electrodes on arms/legs (known as limb leads) and torso (known as precordial leads). ECG interpretation plays an central role in the assessment of cardiovascular disease based on either a cardiologist's experience or computer-aided diagnosis systems. In practice, compute-aided interpretation has become increasingly important, since such technique improves the accuracy of diagnosis, facilitates health care decision making and reduces costs [17].