Meme Generator

Cardiac arrhythmias, posing serious health risks, are diagnosed using electrocardiogram (ECG) signals, crucial for detecting irregularities. Recent advancements in signal processing and machine learning enable efficacious ECG signal analysis, improving prediction and diagnosis of arrhythmias. This paper proposes a novel approach centered on Internet of Things (IoT) schemes for long term monitoring, while also focusing on developing a robust predictive model for detecting irregular heartbeats through ECG signal analysis. The proposed system utilizes an ECG pulse sensor to acquire data, which is subsequently sent to cloud through a Node MCU module for visualization on IoT display. Simultaneously, the input ECG data undergoes a series of technical processes to enable prediction. Initially, the ECG data undergoes preprocessing and feature extraction using Fractional Fourier Transform (FrFT). FrFT extracts relevant features from the data in time frequency domain without overwhelming computational resources which is important for monitoring and analysis. Subsequently, the Improved Bat Algorithm (IBA) refines the feature set by selecting most informative features for classification. Finally, hybrid Modified Convolutional Neural Network Long Short Term Memory (MCNN LSTM) architecture accurately classifies the ECG signals, enabling timely intervention and improved patient care in cardiovascular health management. Experimental validation using Python software demonstrates the superiority of the proposed system, in predicting irregular heartbeats using ECG signal analysis.