Automatic Detection of Irregular Contraction and Relaxation of Cardiac Muscle using Alexnet

Congestive heart failure and arrhythmias are the two cardiovascular conditions that afflict the most people worldwide. A significant majority of modern techniques for automated classification of cardio-vascular diseases rely on traits that are manually generated. The major objective of this research is to create a deep learning-based method that can perform feature identification without the aid of humans. Alex Net convolutional neural network (CNN) using 5,655 singlelead ECGs has been used in this research. In the beginning, the researchers employed the Continuous Wavelet Transform (CWT) to transform all of the 30s signals into spectrograms, which were then converted into RGB images. As a result, with a few modest modifications to the metrics, it has been sent to Alex Net and trained. In this research, Alex net CNN has been used to find out the two diseases which the result is superior when compared to all other conventional techniques, with a remarkable accuracy of 97.9%, an F1 score of 98.82 percent, and higher overall sensitivity (97.8%) and specificity (91.6%).