CRT-Net: A generalized and scalable framework for the computer-aided diagnosis of Electrocardiogram signals

Electrocardiogram (ECG) signals play critical roles in the clinical screening and diagnosis of many types of cardiovascular diseases. Despite the fact that deep neural networks have greatly facilitated the computer-aided diagnosis (CAD) in many clinical tasks, the variability and complexity of ECG in practical scenarios still pose significant challenges in diagnostic performance and clinical applications, especially under the current demand of medical cloud computing and remote diagnosis. In this paper, we propose a generalized and scalable framework for the clinical recognition of ECG. Considering the fact that hospitals generally record ECG signals in the form of graphic waves of 2-D images, we first extract the graphic waves of 12-lead images into numerical 1-D ECG signals by a proposed bi-directional connected method. Subsequently, a novel deep neural network, named CRT-Net, is designed to explore waveform features, morphological characteristics, and time-domain features of ECG by embedding convolution neural network (CNN), recurrent neural network (RNN), and transformer module in a scalable deep model, which is especially suitable in clinical scenarios with different lengths of ECG signals captured from different devices. The proposed framework is first evaluated on two widely investigated public repositories, demonstrating superior performance to the state-of-the-art. We also evaluate the effectiveness of the developed framework on clinically collected ECG images from a local hospital. Moreover, based on the proposed algorithms, we have developed a cloud-based ECG system for the computer-aided diagnosis of different cardiovascular diseases.