Guest Editors
Huiling Chen
College of Computer Science and Artificial Intelligence, Wenzhou University, China
Email: chenhuiling.jlu@gmail.com
Gaige Wang
School of Computer Science and Technology, Ocean University of China, China
Email: wgg@ouc.edu.cn
Summary
Over the past decades, swarm intelligence (SI) has shown its potential in solving complex biological optimization problems. SI algorithms, also known as the modern evolutionary algorithms, have dynamic and flexible global optimization performance and strong versatility and are suitable for parallel processing. Well-established SI algorithms with a strong evolutionary basis cover many areas, from land to sea, water to air, living animals and plants, microorganisms to inanimate natural phenomena, physics and chemistry mathematics, and nonlinear science to complex adaptive systems. An SI algorithm extracts the behaviors of living organisms, including genetic information exchanges and dynamic searching strategies as abstracted algorithmic models and operations for tackling optimization problems. Some popular and well-established SI algorithms include Particle Swarm Optimization (PSO), Ant Colony Optimization (ACO), Harris Hawks Optimization (HHO), Slime Mould Algorithm (SMA), Colony Predation Algorithm (CPA), Moth Search Algorithm (MSA), and Monarch Butterfly Optimization (MBO).
The SI algorithm combined with machine learning or deep learning can select the optimal subset of features for transcriptome or methylome biomarkers and use the validation set to obtain the best combination of hyperparameters for building an efficient model. The former can obtain a subset of features with predictive performance, and the latter can obtain a predictive model with optimal hyperparameters. SI algorithms mimic the evolutionary patterns and searching behaviors of living entities, and they have demonstrated promising capabilities in selecting optimal features to help developers build machine learning models with good performances. The SI algorithm can be combined with machine learning or deep learning techniques for early disease diagnosis and targeted therapy. Notably, the SI algorithm can use epithelial cancer cell’s scaffold as a carrier, its properties as a distributed perception mechanism, and its motility patterns as the swarm movements such as anti-durotaxis, blebbing, and chemotaxis.
An essential aim of this special issue is to make available recent research in the field of swarm intelligence and evolutionary methods and their applications to various biological optimization problems. We welcome authors to submit original research, review, and perspective articles focusing on, but not limited to, new findings in the following areas:
1. Computational biology
2. Bioinformatics
3. Swarm intelligence in biomedicine
4. Biometrics and evolutionary computing
5. Optimized machine learning models for e-healthcare systems
6. Optimized feature selection methods for bio-medicine
7. Efficient prediction methods for healthcare and disease diagnosis
8. Deep learning-based optimized computer-aided diagnosis
9. Efficient disease classification models
10. Efficient image segmentation and thresholding methods for Biomedicine
11. Optimized medical image segmentation and thresholding
etc.
Keywords
Swarm Intelligence, Evolutionary Computation, Biological Applications, Biomedicine, Image Processing, Biology
Published Papers
Login
Register
Submit a Paper
Propose a Special lssue