Introduction:
Accurate analysis of electrocardiogram (ECG) signals is crucial for early and effective detection of many heart abnormalities and diseases. Traditional machine-learning techniques have shown promise; however, they often require complex architectures and may struggle with complex, broad biomedical data that often includes artifacts and other abnormalities.

Bioinspired intelligence algorithms such as Ant Colony Optimization (ASO) offer more lightweight, adaptive alternatives, inspired by the adaptive intelligence of ants observed over the years. ACO has shown significant potential in enhancing feature extraction and classification in ECG signal processing.

Methods:
This study reviews the following recent scientific advances applying ACO to ECG analysis:

• • Gao et al. (2023)¹ introduced a self-adjusting ant colony clustering algorithm with a correction mechanism, achieving high robustness and accuracy in arrhythmia classification using the MIT-BIH dataset.

  • Korürek & Nizam (2010)² applied ACO combined with Principal Component Analysis (PCA) and Discrete Wavelet Transform (DWT) for feature selection and dimensionality reduction, improving classification performance over neural networks and linear discriminant analysis.

  • Cheng et al. (2016)³ used population-based ACO to reconstruct ECG signals, optimizing signal quality and reducing noise through adaptive parameter tuning.

Results:
The self-adjusting ACO model proposed by Gao et al. achieved 99% classification accuracy, outperforming conventional clustering methods.
ACO combined with other models allowed for efficient classification of different arrhythmia types with improved sensitivity and specificity.

Conclusions:
Ant Colony Optimization algorithms provide a resource-efficient approach to ECG signal analysis, with proven accuracy in detecting heart rhythm abnormalities. Their bioinspired adaptability makes them promising candidates for implementation in real-life medical scenarios such as portable cardiac monitoring systems, especially in resource-limited healthcare settings.

Bibliography:
(1) Gao et al., Biomedical Signal Processing and Control, 2023
(2)Korürek & Nizam, Digital Signal Processing, 2010
(3)Cheng et al., Advances in Intelligent Systems and Computing, 2016.