Guest Editors
Dr. Xin Zhang
Email: mexzyl@ust.hk
Affiliation: Department of Mechanical and Aerospace Engineering, The Hong Kong University of Science and Technology,999077, Hong Kong
Homepage:
Research Interests: Complex equipment condition recognition, digital twin, deep learning
Dr. Ruyi Huang
Email: snowxiaoyu@hotmail.com
Affiliation: Department of Mechanical & Aerospace Engineering ,Case Western Reserve University, Cleveland,44106,USA
Homepage:
Research Interests: Interpretable AI-based Method and Its Industrial Applications, Diagnostic and Prognostic based on Industrial Big Data
Dr. Yadong Xu
Email: yadongseu.xu@polyu.edu.hk
Affiliation: Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, 100872, Hong Kong
Homepage:
Research Interests: condition recognition, digital twin, fault diagnosis, signal processing, deep learning
Assist. Prof. Baoru Huang
Email: baoru.huang@liverpool.ac.uk
Affiliation: Department of Computer Science,University of Liverpool, Liverpool, L69 3BX, United Kingdom
Homepage:
Research Interests: Embodied AI, Robotics, Surgical Vision
Summary
Complex systems are ubiquitous across various industries, including advanced mechanical equipment, human physiological systems, and large-scale software systems. Monitoring and analyzing the operational states of such systems is crucial to ensure their safety, reliability, and functionality. In recent years, artificial intelligence (AI) technologies have revolutionized approaches to system monitoring and analysis. Numerous machine learning and deep learning models have been developed to perform pattern recognition and prediction tasks using diverse data types, such as time-series data (e.g., EEG signals, vibration signals), images (e.g., medical imaging), and graph data (e.g., social networks). These AI-driven strategies have laid the foundation for intelligent, generalized operational maintenance and mechanism analysis of complex systems.
Nevertheless, in extreme scenarios, the accuracy of pattern recognition and prediction can be compromised. For instance, external noise can significantly alter data distributions, making analysis challenging. To address these issues, the attention mechanism—an influential deep learning approach—has gained prominence in various fields, including natural language processing, computer vision, equipment health management, and healthcare. By dynamically focusing on the most relevant parts of the input and assigning weights to features based on their importance, attention mechanisms enhance model precision and interpretability. However, designing effective attention modules tailored to specific tasks and scenarios remains a significant challenge.
This special issue invites original research and review articles on the development of attention-based deep learning models for addressing the challenges in complex system pattern recognition and prediction. We encourage submissions on the following topics, but are not limited to:
• Prognostics and health management of complex systems
• Pattern recognition and prediction in healthcare for complex systems
• Attention-based medical image analysis
• Attention-based electroencephalogram (EEG) signal processing and analysis
• Model interpretability for complex systems
• Digital twin technologies for complex systems
• Mechanism analysis of complex systems
• Large language models for complex systems
• Multi-modal data processing in complex systems
• Other deep learning-based algorithms and applications
Keywords
Attention Mechanism; Deep Learning; Complex Systems; Pattern Recognition; Prognostics and Health Management (PHM); Medical Image Analysis; Electroencephalogram (EEG) Signal Processing; Multi-Modal Data Fusion; Digital Twin; Model Interpretability
Published Papers
Login
Register
Submit a Paper
Propose a Special lssue