Swarm intelligence in biomedical engineering

Abstract

Swarm Intelligence (SI), a specialized branch of Artificial Intelligence (AI), is founded on the collective behaviors observed in biological systems, such as those of ants, bees, and bird flocks. This bio-inspired approach enables SI to develop computational algorithms that tackle complex problems, which traditional methods often struggle to address. Over the past few decades, SI has gained substantial traction in biomedical engineering due to its capacity to address multifaceted issues with higher adaptability and efficiency. This review presents key advancements in SI applications across three primary areas: neurorehabilitation, Alzheimer's Disease Diagnosis (ADD), and medical image processing. In neurorehabilitation, SI has played a pivotal role in improving the precision and adaptability of devices such as exoskeletons and neuroprostheses, enhancing motor function recovery for patients. Similarly, in ADD, SI algorithms have shown significant promise in analyzing neuroimaging and neurophysiological data, increasing diagnostic accuracy and enabling earlier intervention. Furthermore, in medical image processing, SI techniques have been effectively applied to tasks such as image segmentation, tumor detection, feature extraction, and artifact reduction, particularly in modalities like Magnetic Resonance Imaging (MRI), Computed Tomography (CT), and ultrasound imaging (such as Sonography). Based on the literature analyzed, SI methods have demonstrated consistent strengths in global optimization, adaptability to noisy data, and robustness in feature selection tasks when compared to traditional machine learning techniques. However, SI still faces challenges, especially regarding computational complexity, model interpretability, and limited clinical translation. Overcoming these hurdles is crucial for the full-scale adoption of this technology in clinical settings. This review not only highlights the progress in these areas but also synthesizes current limitations and future directions to guide the effective integration of SI into real-world biomedical applications.