An improved cuckoo search algorithm with deep learning approach for classifying arrhythmia based on <scp>ECG</scp> signal

Abstract Arrhythmias are variations in the heartbeat rhythm that occur frequently a human's life. These arrhythmias can result potentially deadly consequences, putting one's life danger. As result, detection and classification of is an important issue cardiac diagnostics. Electrocardiogram one easiest ways to diagnose heart disease but complexities due noise present it. This research introduced Improved Cuckoo Search Algorithm (ICSA) which utilized optimize features. Initially, data gathered from MIT‐BIH arrhythmia dataset pre‐processing performed using Discrete Wavelet Transformation (DWT) removes unwanted noises signals. The major limitation standard cuckoo search algorithm increased number iterations. Whenever value probability distribution convergence small then efficiency will be poor enhance ICSA eliminate these drawbacks by fixing values for at early stage increase integrity among solutions. Thus, process optimizing features finally, Support Vector Machine with Feed Forward Back Propagation Neural Network (SVM‐FFBPNN). experimental results s how proposed effectively offers better accuracy 98.21% comparatively higher than Monarch Butterfly Optimization (IMBO) Bat‐Rider (BROA) 97.75% 91.32% respectively.