Internet of Medical Things Privacy and Security: Challenges, Solutions, and Future Trends from a New Perspective

Sustainability, Journal Year: 2023, Volume and Issue: 15(4), P. 3317 - 3317

Published: Feb. 10, 2023

The Internet of Medical Things (IoMT), an application the (IoT) in medical domain, allows data to be transmitted across communication networks. In particular, IoMT can help improve quality life citizens and older people by monitoring managing body’s vital signs, including blood pressure, temperature, heart rate, others. Since has become main platform for information exchange making high-level decisions, it is necessary guarantee its reliability security. growth recent decades attracted interest many experts. This study provides in-depth analysis IoT focusing on security concerns from different points view, this comprehensive survey unique compared other existing studies. A total 187 articles 2010 2022 are collected categorized according type applications, year publications, variety novel perspectives. We compare current studies based above criteria provide a pave way researchers working area. addition, we highlight trends future work. have found that blockchain, as key technology, solved problems security, authentication, maintenance systems due decentralized nature blockchain. study, technology examined fields’ especially health sector, additional importance fields.

Language: Английский

---

Deep learning in public health: Comparative predictive models for COVID-19 case forecasting

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(3), P. e0294289 - e0294289

Published: March 14, 2024

The COVID-19 pandemic has had a significant impact on both the United Arab Emirates (UAE) and Malaysia, emphasizing importance of developing accurate reliable forecasting mechanisms to guide public health responses policies. In this study, we compared several cutting-edge deep learning models, including Long Short-Term Memory (LSTM), bidirectional LSTM, Convolutional Neural Networks (CNN), hybrid CNN-LSTM, Multilayer Perceptron’s, Recurrent (RNN), project cases in aforementioned regions. These models were calibrated evaluated using comprehensive dataset that includes confirmed case counts, demographic data, relevant socioeconomic factors. To enhance performance these Bayesian optimization techniques employed. Subsequently, re-evaluated compare their effectiveness. Analytic approaches, predictive retrospective nature, used interpret data. Our primary objective was determine most effective model for predicting Malaysia. findings indicate selected algorithms proficient cases, although efficacy varied across different models. After thorough evaluation, architectures suitable specific conditions UAE Malaysia identified. study contributes significantly ongoing efforts combat pandemic, providing crucial insights into application sophisticated precise timely cases. hold substantial value shaping strategies, enabling authorities develop targeted evidence-based interventions manage virus spread its populations confirms usefulness methodologies efficiently processing complex datasets generating projections, skill great healthcare professional settings.

Language: Английский

---

A Review of Graph Neural Networks in Epidemic Modeling

Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Journal Year: 2024, Volume and Issue: unknown, P. 6577 - 6587

Published: Aug. 24, 2024

Since the onset of COVID-19 pandemic, there has been a growing interest in studying epidemiological models. Traditional mechanistic models mathematically describe transmission mechanisms infectious diseases. However, they often fall short when confronted with challenges today. Consequently, Graph Neural Networks (GNNs) have emerged as progressively popular tool epidemic research. In this paper, we endeavor to furnish comprehensive review GNNs tasks and highlight potential future directions. To accomplish objective, introduce hierarchical taxonomies for both methodologies, offering trajectory development within domain. For tasks, establish taxonomy akin those typically employed methodology, categorize existing work into Models Hybrid Models. Following this, perform an exhaustive systematic examination encompassing their technical details. Furthermore, discuss limitations methods from diverse perspectives systematically propose research This survey aims bridge literature gaps promote progression promising field. We hope that it will facilitate synergies between communities epidemiology, contribute collective progress.

Language: Английский

---

Machine learning impact of radiative blood flow over a wedge in a time-dependent MHD Williamson fluid

Annals of Mathematics and Computer Science, Journal Year: 2024, Volume and Issue: 22, P. 12 - 44

Published: March 28, 2024

The main objective of this research is to examine the radiation effects an unsteady MHD Williamson biofluid (Blood) over a wedge that interacts with thermophoresis diffusion and Brownian motion. necessary prerequisites partial differential equations (PDEs) ensure development suitable mathematical frameworks for momentum, energy, concentration. These appropriate nonlinear PDEs are frequently transmuted into ordinary (ODEs) by implemented similarity transformation. results these ODEs have significant impact on BVP4C approach from MATLAB package computational structures. graphs tabular data provided various values pertinent parameters non-dimensional velocity temperature, concentration profiles, numerical skin friction, Nusselt number, Sherwood number were found discussed in detail. A novel aspect effort was effective incorporation multiple linear regression (MLR) employing machine learning (ML), statistical technique forecast physical quantities present accuracy 95%. Finally, response predicted variables verified using regression. potential benefit outcomes develop therapeutic diagnostic strategies cancer treatment, as well better understanding medical problems designing more drug delivery systems. In particular, developments computer technology resources, enormous cost simulations still keeps them becoming clinical tool. An additional showed acceptable congruence tangible findings recent enlargements future investigators.

Language: Английский

---

TimeSQL: Improving multivariate time series forecasting with multi-scale patching and smooth quadratic loss

Information Sciences, Journal Year: 2024, Volume and Issue: 671, P. 120652 - 120652

Published: April 25, 2024

Language: Английский

---

MPSTAN: Metapopulation-Based Spatio–Temporal Attention Network for Epidemic Forecasting

Entropy, Journal Year: 2024, Volume and Issue: 26(4), P. 278 - 278

Published: March 25, 2024

Accurate epidemic forecasting plays a vital role for governments to develop effective prevention measures suppressing epidemics. Most of the present spatio–temporal models cannot provide general framework stable and accurate epidemics with diverse evolutionary trends. Incorporating epidemiological domain knowledge ranging from single-patch multi-patch into neural networks is expected improve accuracy. However, relying solely on neglects inter-patch interactions, while constructing challenging without population mobility data. To address aforementioned problems, we propose novel hybrid model called metapopulation-based attention network (MPSTAN). This aims accuracy by incorporating adaptively defining interactions. Moreover, incorporate both construction loss function help learn transmission dynamics. Extensive experiments conducted two representative datasets different evolution trends demonstrate that our proposed outperforms baselines provides more short- long-term forecasting. We confirm effectiveness in learning investigate impact ways integrating observe using leads efficient forecasting, selecting appropriate can further.

Language: Английский

---

Combining the dynamic model and deep neural networks to identify the intensity of interventions during COVID-19 pandemic

PLoS Computational Biology, Journal Year: 2023, Volume and Issue: 19(10), P. e1011535 - e1011535

Published: Oct. 18, 2023

During the COVID-19 pandemic, control measures, especially massive contact tracing following prompt quarantine and isolation, play an important role in mitigating disease spread, quantifying dynamic rate estimate their impacts remain challenging. To precisely quantify intensity of interventions, we develop mechanism physics-informed neural network (PINN) to propose extended transmission-dynamics-informed (TDINN) algorithm by combining scattered observational data with deep learning epidemic models. The TDINN can not only avoid assuming specific functions advance but also make networks follow rules systems process learning. We show that proposed fit multi-source Xi'an, Guangzhou Yangzhou cities well, moreover reconstruct development trend Hainan Xinjiang incomplete reported data. inferred temporal evolution patterns contact/quarantine rates, selected best combination from family accurately simulate time series learned algorithm, consequently reconstructed process. based on have epidemiologically reasonable meanings. In addition, has been verified multiple waves Liaoning province shows good performance. find significant fluctuations estimated a feedback loop between strengthening/relaxation intervention strategies recurrence outbreaks. Moreover, findings there is diversity shape curves rates considered regions, which indicates variation adopted various regions.

Language: Английский

---

Analyzing COVID-19 Spread Mechanisms in Japan Using Time Series Decomposition, Clustering, and Regression

COVID, Journal Year: 2025, Volume and Issue: 5(2), P. 24 - 24

Published: Feb. 17, 2025

In this study, we analyzed daily time series data on newly confirmed COVID-19 cases in each prefecture of Japan to investigate the mechanisms driving virus’s spread. The dataset spans from 20 January 2020 7 May 2023, covering 1204 days and providing insights into case variations across prefectures. First, for were decomposed trend, weekly variation, short-term components. Using trend components, estimated lag infection spread between prefectures, revealing that Okinawa Tokyo consistently led compared other regions. Factors influencing these values also analyzed. Through a cluster analysis, categorized all prefectures 13 groups conducted detailed investigation dynamics within group. results highlighted regions centered around Kanto area acted as primary epicenter, nationwide through Osaka Kyoto. Additionally, examined effects holidays seasonal components using regression analysis. findings showed initially had negative effect numbers, followed by significant positive one week later. Regarding effects, November exhibited highest impact, while March demonstrated impact during period.

Language: Английский

---

A new hybrid approach based on machine learning for more efficient time series forecasting

Data & Metadata, Journal Year: 2025, Volume and Issue: 4, P. 589 - 589

Published: April 9, 2025

Introduction: Forecasting new student enrollment in bachelor's degree programs has emerged as a critical need for higher education institutions. Accurate predictions are essential improving the student-teacher ratio and optimizing resource allocation.Methods: A hybrid approach combining statistical machine learning techniques was proposed to develop accurate forecasting models. The study utilized historical database of Ibn Zohr University, which included data from over twenty institutions dating back 2003. This dataset used train validate models.Results: demonstrated superior accuracy compared standalone results indicated that method effectively captured trends provided reliable forecasts.Conclusions: concluded offers robust solution education. It highlighted potential improve prediction accuracy, thereby aiding better planning management..

Language: Английский

---

State of health prediction of lithium-ion batteries with adaptive loss-based graph neural network

Journal of Energy Storage, Journal Year: 2025, Volume and Issue: 124, P. 116637 - 116637

Published: May 2, 2025

Language: Английский

---