Swarm Intelligence in Internet of Medical Things: A Review

Roohallah Alizadehsani, Mohamad Roshanzamir, Navid Hoseini Izadi, Raffaele Gravina, Hussain Mohammed Dipu Kabir, Darius Nahavandi, Hamid Alinejad-Rokny, Abbas Khosravi, U. Rajendra Acharya, Saeid Nahavandi, Giancarlo Fortino

TL;DR

This review explores how Swarm Intelligence (SI) algorithms can revolutionize the Internet of Medical Things (IoMT) by enabling more efficient, intelligent, and responsive healthcare systems. It highlights the growing importance of IoMT and the potential of SI to address its inherent challenges, from data management to patient safety.

What Problem Does This Paper Address?

The paper addresses the growing need for efficient and accessible healthcare solutions in the face of an aging global population and increasing chronic diseases. The Internet of Medical Things (IoMT) offers a promising avenue for remote healthcare, leveraging connected devices to collect, process, and analyze health data. However, IoMT systems face significant challenges, including:

  • Complexity and Dynamism: The sheer number of interconnected devices and the dynamic nature of healthcare environments make traditional management approaches difficult.
  • Resource Constraints: Many IoMT devices, especially wearables, have limited processing power and battery life.
  • Data Management: Handling vast amounts of sensitive patient data, ensuring its security, and dealing with missing values are critical issues.
  • Real-time Responsiveness: For critical medical applications, systems need to be highly responsive and reliable.

The authors propose Swarm Intelligence (SI) algorithms as a powerful tool to tackle these challenges, drawing inspiration from the collective behavior of social species to create decentralized, robust, and adaptive systems.

Methods

This paper is a comprehensive review of existing research, rather than a presentation of novel methods. The authors systematically searched the Scopus database for relevant literature, focusing on the application of SI algorithms in IoT and, specifically, IoMT. Their methodology involved:

  • Literature Search: Utilizing specific search queries on Scopus to identify relevant papers.
  • Paper Selection: A multi-stage selection process where papers were reviewed by four authors, and inclusion was based on agreement among at least three authors, considering publisher, citation count, and relevance.
  • Categorization: Organizing the reviewed works based on the primary SI algorithm used (e.g., PSO, ACO, ABC) and their application domain within IoT/IoMT.
  • Analysis: Summarizing the findings, contributions, and limitations of existing studies.

The paper discusses various SI algorithms, including:

  • Particle Swarm Optimization (PSO): Inspired by bird flocking, used for optimization tasks like routing, resource allocation, and node localization.
  • Ant Colony Optimization (ACO): Mimicking ant foraging behavior, applied to path planning, routing, and cluster head selection.
  • Artificial Bee Colony (ABC): Based on bee foraging strategies, used for clustering, routing, and optimization.

Datasets

Not discussed in the paper. The paper is a review of existing literature and does not present new experimental results or datasets.

Key Insights and Ideas

  • SI as a Natural Fit for IoMT: The decentralized, adaptive, and robust nature of SI algorithms makes them exceptionally well-suited for the complex and dynamic environment of IoMT, where individual devices may have limited capabilities.
  • Wearables as a Key IoMT Component: The review emphasizes the growing role of wearable devices in IoMT, highlighting their potential for continuous patient monitoring and personalized healthcare.
  • Addressing Core IoMT Challenges: SI algorithms are shown to be effective in tackling critical IoMT issues such as disease prediction, data security, missing value imputation, and resource allocation.
  • The Rise of IoMT Research: The paper indicates that while IoT research is mature, IoMT is still in its early stages, with significant potential for growth and further research.
  • PSO Dominance: Particle Swarm Optimization (PSO) emerges as the most widely applied SI algorithm in IoMT research, likely due to its simplicity, efficiency in continuous optimization, and fewer parameters compared to other methods.

Key Results

The paper does not present quantitative results from new experiments. Instead, it synthesizes the findings from numerous reviewed studies. Key qualitative findings include:

  • Improved Efficiency: SI algorithms have demonstrated improvements in energy consumption, network longevity, and data transmission efficiency in various IoT/IoMT applications.
  • Enhanced Security: SI-based approaches have been proposed for securing medical data and detecting intruders in IoMT systems.
  • Better Data Management: SI algorithms have shown promise in handling missing data and optimizing data storage and analysis.
  • Increased Accuracy: In applications like disease prediction and medical image analysis, SI-based methods have contributed to improved accuracy.

Limitations and Future Work

Limitations:

  • Computational Cost: The paper acknowledges that SI algorithms can have high computational costs, which can be a challenge for resource-constrained IoMT devices.
  • Limited Research in Specific Areas: While the paper reviews many applications, it notes that some SI methods are underutilized, and further research is needed in areas like wireless cognitive networks.
  • Early Stage of IoMT: The IoMT field itself is still developing, meaning that the full potential of SI in this domain is yet to be realized.

Future Work:

  • Wireless Cognitive Networks: Exploring the application of ABC and other SI algorithms in wireless cognitive networks for IoMT.
  • Enhanced QoS Frameworks: Developing more sophisticated Quality of Service (QoS) frameworks for WSNs that are suitable for IoMT, focusing on scalability and energy efficiency.
  • Load Balancing in Edge Computing: Investigating SI methods for effective load balancing between cloud servers and edge devices in IoMT.
  • Patient Safety and Security: Continued research into SI algorithms for enhancing patient safety, device immunization, and robust security mechanisms in IoMT.
  • Dynamic Routing: Further development of SI-based dynamic routing protocols for mobile wireless sensor networks to ensure efficient energy consumption for wearables.
  • Addressing Complex Optimization: Continued exploration of SI algorithms for complex optimization problems in IoMT, such as resource allocation and data management.

Summary updated

Abstract

Continuous advancements of technologies such as machine-to-machine interactions and big data analysis have led to the internet of things (IoT) making information sharing and smart decision-making possible using everyday devices. On the other hand, swarm intelligence (SI) algorithms seek to establish constructive interaction among agents regardless of their intelligence level. In SI algorithms, multiple individuals run simultaneously and possibly in a cooperative manner to address complex nonlinear problems. In this paper, the application of SI algorithms in IoT is investigated with a special focus on the internet of medical things (IoMT). The role of wearable devices in IoMT is briefly reviewed. Existing works on applications of SI in addressing IoMT problems are discussed. Possible problems include disease prediction, data encryption, missing values prediction, resource allocation, network routing, and hardware failure management. Finally, research perspectives and future trends are outlined.
BibTeX
@inproceedings{alizadehsani2023swarm,
  author = {Roohallah Alizadehsani and Mohamad Roshanzamir and Navid Hoseini Izadi and Raffaele Gravina and Hussain Mohammed Dipu Kabir and Darius Nahavandi and Hamid Alinejad-Rokny and Abbas Khosravi and U. Rajendra Acharya and Saeid Nahavandi and Giancarlo Fortino},
  title = {{Swarm Intelligence in Internet of Medical Things: A Review}},
  booktitle = {Sensors},
  year = {2023},
  url = {https://openreview.net/forum?id=apCNUqLpdG},
}

Loading related content...

Figure