On the Use of Machine Learning Techniques and Non-Invasive Indicators for Classifying and Predicting Cardiac Disorders DOI Creative Commons
Raydonal Ospina, A. Ferreira, H. M. de Oliveira

et al.

Biomedicines, Journal Year: 2023, Volume and Issue: 11(10), P. 2604 - 2604

Published: Sept. 22, 2023

This research aims to enhance the classification and prediction of ischemic heart diseases using machine learning techniques, with a focus on resource efficiency clinical applicability. Specifically, we introduce novel non-invasive indicators known as Campello de Souza features, which require only tensiometer clock for data collection. These features were evaluated comprehensive dataset disease cases from repository. Our findings highlight ability algorithms not streamline diagnostic procedures but also reduce errors dependency extensive testing. Three key features—mean arterial pressure, pulsatile blood pressure index, resistance-compliance indicator—were found significantly improve accuracy in binary classification. Logistic regression achieved highest average among examined classifiers when utilizing these features. While such contribute substantially process, they should be integrated into broader framework that includes patient evaluations medical expertise. Therefore, present study offers valuable insights leveraging science techniques diagnosis management cardiovascular diseases.

Language: Английский

Advanced series decomposition with a gated recurrent unit and graph convolutional neural network for non-stationary data patterns DOI Creative Commons

Huimin Han,

Harold Neira-Molina, Asad Khan

et al.

Journal of Cloud Computing Advances Systems and Applications, Journal Year: 2024, Volume and Issue: 13(1)

Published: Jan. 18, 2024

Abstract In this study, we present the EEG-GCN, a novel hybrid model for prediction of time series data, adept at addressing inherent challenges posed by data's complex, non-linear, and periodic nature, as well noise that frequently accompanies it. This synergizes signal decomposition techniques with graph convolutional neural network (GCN) enhanced analytical precision. The EEG-GCN approaches data one-dimensional temporal signal, applying dual-layered using both Ensemble Empirical Mode Decomposition (EEMD) GRU. two-pronged process effectively eliminates interference distills complex into more tractable sub-signals. These sub-signals facilitate straightforward feature analysis learning process. To capitalize on decomposed is employed to discern intricate interplay within map interdependencies among points. predictive then synthesizes weighted outputs GCN yield final forecast. A key component our approach integration Gated Recurrent Unit (GRU) EEMD framework, referred EEMD-GRU-GCN. combination leverages strengths GRU in capturing dependencies EEMD's capability handling non-stationary thereby enriching set available enhancing overall accuracy stability model. evaluations demonstrate achieves superior performance metrics. Compared baseline model, shows an average R2 improvement 60% 90%, outperforming other methods. results substantiate advanced proposed underscoring its potential robust accurate forecasting.

Language: Английский

Citations

7
Comprehensive Analysis of Air Quality Trends in India Using Machine Learning and Deep Learning Models DOI
Isha Ganguli,

Meet Nakum,

B K Das

et al.

Published: Jan. 2, 2025

Language: Английский

Citations

1
Autoregressive integrated moving average with semantic information: An efficient technique for intelligent prediction of dengue cases DOI
Wanarat Juraphanthong, Kraisak Kesorn

Engineering Applications of Artificial Intelligence, Journal Year: 2025, Volume and Issue: 143, P. 109985 - 109985

Published: Jan. 13, 2025

Language: Английский

Citations

1
The impact of lifting COVID-19 restrictions on influenza transmission across countries DOI Creative Commons
Wenjuan Du, Zheng Feng, Yi Zhao

et al.

Advances in Continuous and Discrete Models, Journal Year: 2025, Volume and Issue: 2025(1)

Published: Jan. 31, 2025

Language: Английский

Citations

1
Multi-strain COVID-19 dynamics with vaccination strategies: Mathematical modeling and case study DOI Creative Commons

Venkatesh Ambalarajan,

Ankamma Rao Mallela,

Prasantha Bharathi Dhandapani

et al.

Alexandria Engineering Journal, Journal Year: 2025, Volume and Issue: 119, P. 665 - 684

Published: Feb. 12, 2025

Language: Английский

Citations

1
Riding into the Future: Transforming Jordan’s Public Transportation with Predictive Analytics and Real-Time Data DOI
Anber Abraheem Shlash Mohammad, Sulieman Ibraheem Shelash Al-Hawary,

Khaleel Ibrahim Al‐ Daoud

et al.

Data & Metadata, Journal Year: 2025, Volume and Issue: 4, P. 887 - 887

Published: April 4, 2025

Introduction: This study explores how predictive analytics and real-time data integration can improve efficiency in Jordan’s public transportation network. By addressing scheduling, route optimization, congestion management, it responds to growing urban transit demands the region.Methods: Data were collected over three months from official ridership logs, GPS-enabled buses, traffic APIs. ARIMA-based time-series forecasting captured historical trends, while a Random Forest model incorporated index, average wait times, other operational variables. Metadata management protocols (JSON/XML) facilitated cross-agency sharing.Results: ARIMA proved effective for short-term passenger demand projections, although occasionally underpredicted sudden peaks. The approach yielded stronger overall accuracy, explaining roughly 85% of variation when combining with records. Real-time streams further supported dynamic scheduling adjustments.Conclusion: Combining models IoT-based enhance reliability user satisfaction system. Although limited by timeframe scope, findings underscore importance multi-agency collaboration ongoing policy support sustain data-driven innovations.

Language: Английский

Citations

1
An IoT-fuzzy intelligent approach for holistic management of COVID-19 patients DOI Creative Commons

Muhammad Zia Ur Rahman,

Muhammad Azeem Akbar, Víctor Leiva

et al.

Heliyon, Journal Year: 2023, Volume and Issue: 10(1), P. e22454 - e22454

Published: Nov. 20, 2023

In this study, an internet of things (IoT)-enabled fuzzy intelligent system is introduced for the remote monitoring, diagnosis, and prescription treatment patients with COVID-19. The main objective present study to develop integrated tool that combines IoT logic provide timely healthcare diagnosis within a smart framework. This tracks patients' health by utilizing Arduino microcontroller, small affordable computer reads data from various sensors, gather data. Once collected, are processed, analyzed, transmitted web page access via IoT-compatible Wi-Fi module. cases emergencies, such as abnormal blood pressure, cardiac issues, glucose levels, or temperature, immediate action can be taken monitor critical COVID-19 in isolation. employs recommend medical treatments patients. Sudden changes these conditions remotely reported through providers, relatives, friends. assists professionals making informed decisions based on patient's condition.

Language: Английский

Citations

12
PVS-GEN: Systematic Approach for Universal Synthetic Data Generation Involving Parameterization, Verification, and Segmentation DOI Creative Commons
Kyung Min Kim, Jong Wook Kwak

Sensors, Journal Year: 2024, Volume and Issue: 24(1), P. 266 - 266

Published: Jan. 2, 2024

Synthetic data generation addresses the challenges of obtaining extensive empirical datasets, offering benefits such as cost-effectiveness, time efficiency, and robust model development. Nonetheless, synthetic data-generation methodologies still encounter significant difficulties, including a lack standardized metrics for modeling different types comparing generated results. This study introduces PVS-GEN, an automated, general-purpose process verification. The PVS-GEN method parameterizes time-series with minimal human intervention verifies construction using specific metric derived from extracted parameters. For complex data, iteratively segments dataset until parameter can reproduce that reflects characteristics, irrespective sensor type. Moreover, we introduce PoR to quantify quality by evaluating its characteristics. Consequently, proposed automatically generate diverse covers wide range types. We compared existing methodologies, demonstrated superior performance. It similarity up 37.1% across multiple 19.6% on average metric,

Language: Английский

Citations

3
Campgrounds and climate change: An extreme weather event study for nature-based entrepreneurship DOI
Christopher A. Craig, Leiza Nochebuena-Evans, Robert D. Evans

et al.

Journal of Business Venturing Insights, Journal Year: 2024, Volume and Issue: 22, P. e00477 - e00477

Published: May 23, 2024

Language: Английский

Citations

3
A six-compartment model for COVID-19 with transmission dynamics and public health strategies DOI Creative Commons

Venkatesh Ambalarajan,

Ankamma Rao Mallela,

Vinoth Sivakumar

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Sept. 27, 2024

Language: Английский

Citations

3