Accurate Mental Stress Detection Using Sequential Backward Selection and Adaptive Synthetic Methods

IEEE Transactions on Neural Systems and Rehabilitation Engineering, Journal Year: 2024, Volume and Issue: 32, P. 3095 - 3103

Published: Jan. 1, 2024

The daily experience of mental stress profoundly influences our health and work performance while concurrently triggering alterations in brain electrical activity. Electroencephalogram (EEG) is a widely adopted method for assessing cognitive affective states. This study delves into the EEG correlates potential use resting evaluating levels. Over 13 weeks, longitudinal focuses on real-life experiences college students, collecting data from each 18 participants across multiple days classroom settings. To tackle complexity arising multitude features imbalance samples levels, we sequential backward selection (SBS) feature adaptive synthetic (ADASYN) sampling algorithm imbalanced data. Our findings unveil that delta theta account approximately 50% selected through SBS process. In leave-one-out (LOO) cross-validation, combination band power pair-wise coherence (COH) achieves maximum balanced accuracy 94.8% stress-level detection above dataset. Notably, using ADASYN borderline synthesized minority over-sampling technique (borderline-SMOTE) methods enhances model compared to traditional SMOTE approach. These results provide valuable insights signals levels scenarios, shedding light strategies managing more effectively.

Language: Английский

---

IoT-Based Assessment of a Driver’s Stress Level

Sensors, Journal Year: 2024, Volume and Issue: 24(17), P. 5479 - 5479

Published: Aug. 23, 2024

Driver Monitoring Systems (DMSs) play a key role in preventing hazardous events (e.g., road accidents) by providing prompt assistance when anomalies are detected while driving. Different factors, such as traffic and conditions, might alter the psycho-physiological status of driver increasing stress workload levels. This motivates development advanced monitoring architectures taking into account aspects. In this work, we propose novel

Language: Английский

---

Revolutionizing cardiovascular health: integrating deep learning techniques for predictive analysis of personal key indicators in heart disease

Neural Computing and Applications, Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 15, 2024

Abstract Cardiovascular diseases (CVDs) remain a global burden, highlighting the need for innovative approaches early detection and intervention. This study investigates potential of deep learning, specifically convolutional neural networks (CNNs), to improve prediction heart disease risk using key personal health markers. Our approach revolutionizes traditional healthcare predictive modeling by integrating CNNs, which excel at uncovering subtle patterns hidden interactions among various indicators such as blood pressure, cholesterol levels, lifestyle factors. To achieve this, we leverage advanced network architectures. The model utilizes embedding layers transform categorical data into numerical representations, extract spatial features, dense complex predict CVD risk. Regularization techniques like dropout batch normalization, along with hyperparameter optimization, enhance generalizability performance. Rigorous validation against conventional methods demonstrates model’s superiority, significantly higher R 2 value 0.994. achievement underscores valuable tool clinicians in prevention management. also emphasizes interpretability learning models addresses ethical considerations ensure responsible implementation clinical practice.

Language: Английский

---

Multimodal driver emotion recognition using motor activity and facial expressions

Frontiers in Artificial Intelligence, Journal Year: 2024, Volume and Issue: 7

Published: Nov. 27, 2024

Driving performance can be significantly impacted when a person experiences intense emotions behind the wheel. Research shows that such as anger, sadness, agitation, and joy increase risk of traffic accidents. This study introduces methodology to recognize four specific using an intelligent model processes analyzes signals from motor activity driver behavior, which are generated by interactions with basic driving elements, along facial geometry images captured during emotion induction. The research applies machine learning identify most relevant for recognition. Furthermore, pre-trained Convolutional Neural Network (CNN) is employed extract probability vectors corresponding under investigation. These data sources integrated through unidimensional network classification. main proposal this was develop multimodal combines accurately (anger, joy) in drivers, achieving 96.0% accuracy simulated environment. confirmed significant relationship between drivers' activity, geometry, induced emotions.

Language: Английский

---

Winds of Power: Data Analysis for the Relationship between Wind Speed, Gust, and Power Output

Journal of Engineering Research - Egypt/Journal of Engineering Research, Journal Year: 2023, Volume and Issue: 7(5), P. 189 - 194

Published: Nov. 1, 2023

Wind turbines are the most cost-effective and quickly evolving renewable energy technology. Benefits of this technology include no carbon emissions, resource conservation, job creation, flexible applications, modularity, fast installation, rural power grid improvement, potential for agricultural or industrial use.

Language: Английский

---