A deep reinforcement learning-based wireless body area network offloading optimization strategy for healthcare services

Chen, Yingqun; Han, Shaodong; Chen, Guihong; Yin, Jiao; Wang, Kate Nana; Cao, Jun

doi:10.1007/s13755-023-00212-3

Cited by 16 publications

(3 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Deep learning evolved from the study of artificial neural networks; however, it is not identical to conventional neural networks. Nevertheless, in terms of vocabulary, the many deep learning algorithms, including deep reinforcement learning, generative adversarial networks, recurrent neural networks, and convolutional neural networks, use the phrase “neural network” [ 47 , 48 , 49 , 50 ]. Deep learning can be thought of as a semi-theoretical, semi-empirical modelling approach that employs human understanding of mathematics and computer algorithms, along with as much training information as is possible, to construct an architectural framework, utilizing the massive computing power of computers to tune the internal criteria to approximate the issue’s objectives as closely as possible [ 51 ].…”

Section: Discussionmentioning

confidence: 99%

Nomograms for Predicting the Risk and Prognosis of Liver Metastases in Pancreatic Cancer: A Population-Based Analysis

Shi

Zhou

et al. 2023

JPM

View full text Add to dashboard Cite

The liver is the most prevalent location of distant metastasis for pancreatic cancer (PC), which is highly aggressive. Pancreatic cancer with liver metastases (PCLM) patients have a poor prognosis. Furthermore, there is a lack of effective predictive tools for anticipating the diagnostic and prognostic techniques that are needed for the PCLM patients in current clinical work. Therefore, we aimed to construct two nomogram predictive models incorporating common clinical indicators to anticipate the risk factors and prognosis for PCLM patients. Clinicopathological information on pancreatic cancer that referred to patients who had been diagnosed between the years of 2004 and 2015 was extracted from the Surveillance, Epidemiology, and End Results (SEER) database. Univariate and multivariate logistic regression analyses and a Cox regression analysis were utilized to recognize the independent risk variables and independent predictive factors for the PCLM patients, respectively. Using the independent risk as well as prognostic factors derived from the multivariate regression analysis, we constructed two novel nomogram models for predicting the risk and prognosis of PCLM patients. The area under the curve (AUC) of the receiver operating characteristic (ROC) curve, the consistency index (C-index), and the calibration curve were then utilized to establish the accuracy of the nomograms’ predictions and their discriminability between groups. Using a decision curve analysis (DCA), the clinical values of the two predictors were examined. Finally, we utilized Kaplan–Meier curves to examine the effects of different factors on the prognostic overall survival (OS). As many as 1898 PCLM patients were screened. The patient’s sex, primary site, histopathological type, grade, T stage, N stage, bone metastases, lung metastases, tumor size, surgical resection, radiotherapy, and chemotherapy were all found to be independent risks variables for PCLM in a multivariate logistic regression analysis. Using a multivariate Cox regression analysis, we discovered that age, histopathological type, grade, bone metastasis, lung metastasis, tumor size, and surgery were all independent prognostic variables for PCLM. According to these factors, two nomogram models were developed to anticipate the prognostic OS as well as the risk variables for the progression of PCLM in PCLM patients, and a web-based version of the prediction model was constructed. The diagnostic nomogram model had a C-index of 0.884 (95% CI: 0.876–0.892); the prognostic model had a C-index of 0.686 (95% CI: 0.648–0.722) in the training cohort and a C-index of 0.705 (95% CI: 0.647–0.758) in the validation cohort. Subsequent AUC, calibration curve, and DCA analyses revealed that the risk and predictive model of PCLM had high accuracy as well as efficacy for clinical application. The nomograms constructed can effectively predict risk and prognosis factors in PCLM patients, which facilitates personalized clinical decision-making for patients.

show abstract

Section: Discussionmentioning

confidence: 99%

Nomograms for Predicting the Risk and Prognosis of Liver Metastases in Pancreatic Cancer: A Population-Based Analysis

Shi

Zhou

et al. 2023

JPM

View full text Add to dashboard Cite

show abstract

“…The primary objective of this approach is to empower decision-makers in strategically planning and optimizing healthcare resources, thereby enabling streamlined processes and enhanced efficiency in patient allocation. In [137], the authors focused on optimizing the computational offloading and resource allocation in Mobile Edge Computing (MEC) for healthcare scenarios by using reinforcement learning algorithms. They propose an offloading approach called Deep Deterministic Policy Gradient-based WBAN Offloading Strategy (DDPG-WOS) to address the challenges related to energy consumption and latency in the transmission channels.…”

Section: ) Machine Learning Algorithmsmentioning

confidence: 99%

Empowering Healthcare With Cyber-Physical System—A Systematic Literature Review

Khater,

Sallabi,

Serhani

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Cyber-physical systems (CPS) are recognized for their intelligence as they seamlessly interact with humans, enhancing the physical world through computation, communication, and control. Over the last few years, the evolution of CPS, including IoT components, has significantly impacted many facets of people's lifestyles. It has been immersed in a wide range of services and applications in various areas, including manufacturing, healthcare, and energy. However, the interrelationship between the cyber and physical worlds gives rise to a multitude of research problems and challenges. In the healthcare field, for instance, CPS introduces complexities related to interoperability, privacy, data security, and real-time data processing with critical implications for the reliability and safety of medical processes. To address these challenges and harness the full potential of CPS in healthcare, this paper, through a comprehensive literature review, aims to discuss cutting-edge CPS technologies and solutions that hold promise for healthcare applications. To this end, we propose a comprehensive architectural model that can serve as a benchmark for implementing CPS in healthcare applications. This model thoroughly details how services, components, and technologies can be integrated to transform massive raw data collected from the physical world into valuable information for an enhanced decision-making process. Finally, a use case on healthcare CPS is presented, outlining its characteristics, the role that different technologies have played in its development, and the major challenges in implementing such systems successfully. This study provides a cohesive understanding of the role CPS can play in empowering healthcare while offering insights into the challenges, and future research trends.INDEX TERMS Cyber-physical systems, Healthcare, Real-Time processing, IoT.

show abstract

“…Wireless Body Area Networks (WBANs) are a specialized form of wireless networks that involve the use of small, lowpower sensors placed on or inside the human body to monitor various physiological parameters [1]- [4]. WBANs have gained significant attention in recent years due to their potential applications in healthcare, sports monitoring, and wellness tracking.…”

Section: Introductionmentioning

confidence: 99%

Current security and privacy posture in wireless body area networks

Auko¹

2023

World J. Adv. Res. Rev.

View full text Add to dashboard Cite

Wireless Body Area Networks (WBANs) have emerged as a promising technology for remote health monitoring and healthcare applications. However, ensuring the security and privacy of sensitive health data in WBANs is crucial to foster user trust and prevent unauthorized access or data breaches. This paper provides an overview of the key challenges, techniques, and research gaps in WBAN security and privacy. The findings indicate that the challenges in WBAN security and privacy include resource constraints, compatibility issues, privacy concerns, dynamic network environments, security and usability trade-offs, emerging threat landscape, and user awareness and education. To address these challenges, various security techniques have been developed, such as authentication and authorization mechanisms, encryption, access control, secure communication protocols, intrusion detection systems, and privacy-preserving data handling techniques. Despite the progress made, there are research gaps that require further investigation. These research gaps include the development of secure and lightweight authentication mechanisms, privacy-preserving data analysis techniques, trust and security management frameworks, resilience to insider threats, security of data aggregation and fusion, user-centric security designs, and addressing legal and ethical considerations. Addressing these research gaps and challenges requires collaboration between researchers, device manufacturers, policymakers, and end-users. Ongoing research and innovation are necessary to develop robust security techniques, privacy-enhancing technologies, and user-friendly solutions tailored for WBANs. Additionally, compliance with privacy regulations, user education, and awareness are critical for responsible and ethical use of WBANs.

show abstract

A deep reinforcement learning-based wireless body area network offloading optimization strategy for healthcare services

Cited by 16 publications

References 41 publications

Nomograms for Predicting the Risk and Prognosis of Liver Metastases in Pancreatic Cancer: A Population-Based Analysis

Nomograms for Predicting the Risk and Prognosis of Liver Metastases in Pancreatic Cancer: A Population-Based Analysis

Empowering Healthcare With Cyber-Physical System—A Systematic Literature Review

Current security and privacy posture in wireless body area networks

Contact Info

Product

Resources

About