In this article, we present a resource-efficient approach for electrocardiogram (ECG) based heartbeat classification using multi-feature fusion and bidirectional long short-term memory (Bi-LSTM). The dataset comprises five original classes from the MIT-BIH Arrhythmia Database: Normal (N), Left Bundle Branch Block (LBBB), Right Bundle Branch Block (RBBB), Premature Ventricular Contraction (PVC), and Paced Beat (PB). Preprocessing methods including the discrete wavelet transform and dual moving average windows are used to reduce noise and artifacts in the raw ECG signal, and extract the main points (PQRST) of the ECG waveform. Multi-feature fusion is achieved by utilizing time intervals and the proposed under-the-curve areas, which are inherently robust against noise, as input features. Simulations demonstrated that incorporating under-the-curve area features improved the classification accuracy for the challenging RBBB and LBBB classes from 31.4% to 84.3% for RBBB, and from 69.6% to 87.0% for LBBB. Using a Bi-LSTM network, rather than a conventional LSTM network, resulted in higher accuracy (33.8% vs 21.8%) with a 28% reduction in required network parameters for the RBBB class. Multiple neural network models with varying parameter sizes, including tiny (84k), small (150k), medium (478k), and large (1.25M) models, are developed to achieve high accuracy across all classes, a more crucial and challenging goal than overall classification accuracy.

Abstract--We present an integrated approach by combining analog computing and deep learning for electrocardiogram (ECG) arrhythmia classification. Experimental evaluations on PhysionNet's MIT-BIH and PTB Diagnostics datasets demonstrate the effectiveness of the proposed Despite the challenges associated with The electrocardiogram (ECG) is crucial for monitoring analog circuits, such as susceptibility to noise and device heart health in medical practice [1], [2]. However, accurately variation, they can be effectively utilized for inferring neural detecting and categorizing different waveforms and network algorithms. The presence of inherent system noise in morphologies in ECG signals is challenging, similar to other analog circuits can be leveraged to enhance robustness and time-series data. Moreover, manual analysis is time-consuming improve classification accuracy, aligning with the desirable and prone to errors. Given the prevalence and potential lethality properties of AI algorithms [24]-[26]. of irregular heartbeats, achieving accurate and cost-effective In this paper, we propose EKGNet, a fully analog neural diagnosis of arrhythmic heartbeats is crucial for effectively network with low power consumption (10.96μW) that achieves managing and preventing cardiovascular conditions [3], [4].