Title: Arrhythmia detection in the ECG signals with the fractional corvus escape algorithm optimised activation ensemble DCNN

Authors: T. Neetha; James Visumathi

Addresses: Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, No. 42, Avadi-Vel Tech Road, Vel Nagar, Avadi, Chennai, 600062, Tamil Nadu, India ' Department of Computer Science and Engineering, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, No.42, Avadi-Vel Tech Road, Vel Nagar, Avadi, Chennai, 600062, Tamil Nadu, India

Abstract: The detection of arrhythmia acts as a critical task even in the advanced medical sector that works with advancements in artificial intelligence (AI) such as machine learning (ML), and deep learning techniques. The fractional corvus escape algorithm is used in conjunction with the activation ensemble deep convolutional neural network to address the shortcomings of the existing methods. Methods like Yule-Walker, time, and frequency domain features extraction methods that particularly analyse the signal data to obtain precise features. The FCEA-optimised AE-DCNN permits the ensemble activation function that employs convolutional layers to perform the seamless detection of the arrhythmia. To get around the model's stability loss, the two distinct conventional traits of the sparrow and the cuckoo that are combined in the FCEA optimisation are used in the research. The entire model enables the perfect arrhythmia detection and achieves 98.85%, 97.85%, 98.35%, and 0.023 with precision, recall, f1-score, and error respectively.

Keywords: arrhythmia detection; activation ensemble DCNN; Yule-Walker; discrete wavelet transform; fractional corvus escape algorithm.

DOI: 10.1504/IJBRA.2025.149730

International Journal of Bioinformatics Research and Applications, 2025 Vol.21 No.5, pp.481 - 505

Received: 04 Apr 2024
Accepted: 01 Aug 2024
Published online: 11 Nov 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article