Optimized Tiny Machine Learning and Explainable AI for Trustable and Energy-Efficient Fog-Enabled Healthcare Decision Support System DOI Creative Commons

R. Arthi,

S. Krishnaveni

International Journal of Computational Intelligence Systems, Journal Year: 2024, Volume and Issue: 17(1)

Published: Sept. 2, 2024

The Internet of things (IoT)-based healthcare decision support system plays a crucial role in modern medicine, especially with the rise chronic illnesses and an aging population necessitating continuous remote health monitoring. Current systems struggle to deliver timely accurate decisions minimal latency due limited real-time data inefficient computational resources. There is critical need for optimized, energy-efficient machine learning model that reliably supports monitoring within IoT fog computing environments. Our study proposes Optimized Tiny Machine Learning (TinyML) Explainable AI (XAI) binary classification trustable system, leveraging optimize performance. fog-based approach improves response times enhances bandwidth usage, addressing needs such as reduced latency, higher utilization, decreased packet loss. To further improve efficiency, we incorporate innovative mLZW compression technique, significantly enhancing communication efficiency reducing time alerts. However, records challenge By implementing TinyML algorithm, our demonstrates superior performance other models. proposed optimized achieves impressive F1 score 0.93 abnormalities detection, emphasizing its robustness effectiveness. This paper highlights potential XAI delivering robust, trustworthy, energy-aware solutions, making significant contributions toward effective fog-enabled networks.

Language: Английский

Enhancing patient information performance in internet of things-based smart healthcare system: Hybrid artificial intelligence and optimization approaches DOI
Ali Ala, Vladimir Šimić, Dragan Pamučar

et al.

Engineering Applications of Artificial Intelligence, Journal Year: 2024, Volume and Issue: 131, P. 107889 - 107889

Published: Jan. 16, 2024

Language: Английский

Citations

24
5G technology for healthcare: Features, serviceable pillars, and applications DOI Creative Commons
Mohd Javaid, Abid Haleem, Ravi Pratap Singh

et al.

Intelligent Pharmacy, Journal Year: 2023, Volume and Issue: 1(1), P. 2 - 10

Published: May 10, 2023

5G refers to wireless network technology and has opened up new healthcare possibilities in innovation expanded access treatment. is a unified, powerful air interface built with increased capacity support next-generation user experiences services. one of the essential technologies for societal digital transformation, it also prerequisite interconnection everything smart healthcare. Promoting implementing can reduce inconsistencies allocating medical resources expedite advancements. This article studied its need Smart Principal Features Serviceable Pillars Technology are discussed briefly. Finally, identify discuss significant applications promises give people more control over their health. With implementation 5G, we will most certainly witness introduction technology, allowing patients test monitor health from comfort homes. The combination Artificial Intelligence (AI) result devices that connect and, as result, broaden backdrop decision-making. It creates potential growth internal ecosystem. connection coverage may be limited areas tall broad trees buildings. In future, operators device makers collaborate care.

Language: Английский

Citations

32
AI-driven approaches for optimizing power consumption: a comprehensive survey DOI Creative Commons

Parag Biswas,

Abdur Rashid,

Angona Biswas

et al.

Discover Artificial Intelligence, Journal Year: 2024, Volume and Issue: 4(1)

Published: Dec. 27, 2024

Reduced environmental impacts, lower operating costs, and a stable, sustainable energy supply for current future generations are the main reasons why power optimization is important. Power ensures that used more efficiently, reducing waste optimizing utilization of resources. In today's world, integration artificial intelligence (AI) essential transforming how produced, used, distributed. AI-driven algorithms predictive analytics enable real-time monitoring analysis usage trends, allowing dynamic adjustments to effectively meet demand. Efficiency sustainability enhanced across various sectors by consumption through intelligent systems. This survey paper provides an extensive review different AI techniques optimization, along with systematic literature on application systems diverse areas consumption. The evaluates performance outcomes 17 distinct research methodologies, highlighting their strengths limitations. Additionally, this article outlines directions in optimization.

Language: Английский

Citations

8
Editorial: The New Era of Computer Network by using Machine Learning DOI Open Access
Suyel Namasudra, Pascal Lorenz, Uttam Ghosh

et al.

Mobile Networks and Applications, Journal Year: 2023, Volume and Issue: 28(2), P. 764 - 766

Published: March 7, 2023

Language: Английский

Citations

10
PEFTOSPRO: A Power-Efficient and Fault-Tolerant Scheme for Permutation Routing in Multi-hop Wireless Sensor Networks DOI
Alain Bertrand Bomgni, Miguel Landry Foko Sindjoung, Clémentin Tayou Djamegni

et al.

International Journal of Wireless Information Networks, Journal Year: 2024, Volume and Issue: 31(2), P. 96 - 108

Published: Jan. 24, 2024

Language: Английский

Citations

3
Energy-efficient routing protocol for reliable low-latency Internet of Things in oil and gas pipeline monitoring DOI Creative Commons
Sana Nasim Karam, Kashif Bilal, Abdul Nasir Khan

et al.

PeerJ Computer Science, Journal Year: 2024, Volume and Issue: 10, P. e1908 - e1908

Published: Feb. 29, 2024

The oil and gas industries (OGI) are the primary global energy source, with pipelines as vital components for OGI transportation. However, pipeline leaks pose significant risks, including fires, injuries, environmental harm, property damage. Therefore, maintaining an effective maintenance system is critical ensuring a safe sustainable supply. Internet of Things (IoT) has emerged cutting-edge technology efficient leak detection. deploying IoT in monitoring faces challenges due to hazardous environments limited communication infrastructure. Energy efficiency fault tolerance, typical concerns, gain heightened importance context. In monitoring, devices linearly deployed no alternative mechanism available along pipelines. Thus, absence both routes can disrupt crucial data transmission. energy-efficient fault-tolerant paramount. Critical needs reach control center on time faster actions avoid loss. Low latency another challenge environment. Moreover, gather plethora parameter redundant values that hold relevance transmission center. optimizing essential conserve monitoring. This article presents Priority-Based, Energy-Efficient, Optimal Data Routing Protocol (PO-IMRP) tackle these challenges. model congestion optimize packets congestion-free network. PO-IMRP, nodes aware their status communicate node’s depletion timely network robustness. Priority-based routing selects low-latency losses. Comparative analysis against linear LEACH highlights PO-IMRP’s superior performance terms total packet by completing fewer rounds more packet’s transmissions, attributed optimization technique implemented at each hop, which helps mitigate congestion. MATLAB simulations affirm effectiveness protocol efficiency, fault-tolerance, low communication.

Language: Английский

Citations

3
Dynamic Resource Allocation Techniques for Wireless Network Data in Elastic Optical Network Applications DOI

Jing Ge,

Kangcheng Wu,

Nasir Jamal

et al.

Mobile Networks and Applications, Journal Year: 2023, Volume and Issue: 28(5), P. 1712 - 1723

Published: Oct. 1, 2023

Language: Английский

Citations

4
Optimized Tiny Machine Learning and Explainable AI for Trustable and Energy-Efficient Fog-Enabled Healthcare Decision Support System DOI Creative Commons

R. Arthi,

S. Krishnaveni

International Journal of Computational Intelligence Systems, Journal Year: 2024, Volume and Issue: 17(1)

Published: Sept. 2, 2024

The Internet of things (IoT)-based healthcare decision support system plays a crucial role in modern medicine, especially with the rise chronic illnesses and an aging population necessitating continuous remote health monitoring. Current systems struggle to deliver timely accurate decisions minimal latency due limited real-time data inefficient computational resources. There is critical need for optimized, energy-efficient machine learning model that reliably supports monitoring within IoT fog computing environments. Our study proposes Optimized Tiny Machine Learning (TinyML) Explainable AI (XAI) binary classification trustable system, leveraging optimize performance. fog-based approach improves response times enhances bandwidth usage, addressing needs such as reduced latency, higher utilization, decreased packet loss. To further improve efficiency, we incorporate innovative mLZW compression technique, significantly enhancing communication efficiency reducing time alerts. However, records challenge By implementing TinyML algorithm, our demonstrates superior performance other models. proposed optimized achieves impressive F1 score 0.93 abnormalities detection, emphasizing its robustness effectiveness. This paper highlights potential XAI delivering robust, trustworthy, energy-aware solutions, making significant contributions toward effective fog-enabled networks.

Language: Английский

Citations

1