Central AsianResearch Index

DEEP NEURAL NETWORK CONV-LSTM FOR ECG-BASED CARDIAC DISORDER IDENTIFICATION

. In the growing field of health diagnostics, the electrocardiogram (ECG) serves as a fundamental instrument for evaluating cardiac health. This paper presents an innovative methodology that leverages the capabilities of the Deep Conv-LSTM architecture for rapid and accurate arrhythmia detection via ECG measurements. Historically, cardiac anomalies have been identified through heuristic evaluations, necessitating in-depth investigation and expert expertise. The Deep Conv-LSTM model addresses the limitations of traditional methodologies by integrating the spatial feature extraction capabilities of convolutional neural networks (CNN) with the temporal pattern recognition strengths of LSTM networks. Initial findings, derived from a comprehensive dataset rich in ECG waveform anomalies, indicated a significant increase in accuracy, reducing false detections and accelerating essential actions. This model demonstrated an aptitude for handling the unexpected characteristics of ECG signals, reflecting a range of heart rhythms and the complexities involved in identifying subtle arrhythmic events. Furthermore, the model's proficiency in identifying prolonged, complex sequences coupled with transient anomalies indicates its potential use in remote diagnostics and continuous patient monitoring systems. The integration of CNN and LSTM architectures is set to initiate a transformative era in automated arrhythmia detection, creating a connection between technical advancements and the intricate details of cardiac activity, with the potential to enhance patient care quality.

Сведения

Опубликовано
2025 · № 4
Язык
Английский
ISSN
1991-346X · 2518-1726
Издатель
Central Asian Academic Research Center LLP