Automatic stress detection in car drivers based on non-invasive physiological signals using machine learning techniques

Neural Computing and Applications, Journal Year: 2023, Volume and Issue: 35(17), P. 12891 - 12904

Published: March 9, 2023

Abstract Stress is now thought to be a major cause wide range of human health issues. However, many people may ignore their stress feelings and disregard take action before serious physiological mental disorders place. The heart rate (HR) blood pressure (BP) are the most markers used in various studies detect for human, because they captured non-invasively using wearable sensors, these recommended provide information on person’s state. Most assessment have been undertaken laboratory-based controlled environment. This paper proposes an approach identify automotive drivers based selected biosignals, namely, ECG, EMG, GSR, respiration rate. In this study, six different machine learning models (KNN, SVM, DT, LR, RF, MLP) classify between stressed relaxation states. Such system can integrated with Driver Assistance System (DAS). proposed detection technique (SDT) consists three main phases: (1) Biosignal Pre-processing, which signal segmented filtered. (2) Feature Extraction, some discriminate features extracted from each biosignal describe state driver. (3) Classification. results show that RF classifier outperforms other techniques classification accuracy 98.2%, sensitivity 97%, specificity 100% drivedb dataset.

Language: Английский

---

A Novel Human Stress Level Detection Technique Using EEG

Published: Sept. 1, 2023

In the 21st century, a significant portion of world's population is plagued by stress. Stress harmful to humans and can cause various physical mental illnesses, such as headaches, anxiety, depression, heart diseases. The aim this study design machine learning-based model measure stress level person using electroencephalography (EEG) signals frontal lobe. proposed assessment technique classify into four categories: no stress, low moderate high our study, first EEG subjects have been recorded while they solve mathematical question sets with different complexity levels. Subsequently, handcrafted feature extraction techniques employed for extracting eight features, namely Skewness, Kurtosis, Maximum, Mean, Mean Absolute Value, Minimum, Standard Deviation, Power Spectral Density from pre-processed signals. A majority voting-based ensemble classifier has designed combining predictions three classifiers, i.e., Support Vector Machine (SVM), K-Nearest neighbour (KNN), Naive Bayes, predict person's level. obtained classification accuracy 93.85 %.

Language: Английский

---

A deep learning-based approach for distinguishing different stress levels of human brain using EEG and pulse rate

Computer Methods in Biomechanics & Biomedical Engineering, Journal Year: 2023, Volume and Issue: 27(16), P. 2303 - 2324

Published: Nov. 6, 2023

In today's world, people suffer from many fatal maladies, and stress is one of them. Excessive can have deleterious effects on the health, brain, mind, nervous system humans. The goal this paper to design a deep learningbased human level measurement technique using electroencephalogram (EEG), pulse rate. research, EEG signals rate healthy subjects are recorded while they solve four different question sets increasing complexity. It assumed that undergo through levels, i.e., 'no stress', 'low 'medium 'high solving these sets. An attention mechanism-based CNN-TLSTM (convolutional neural network-tanh long short-term memory) model proposed detect mental person. layer incorporated into designed TLSTM network increase classification accuracy model. CNN used for automated extraction intricate features Then classify person categories CNNextracted features. obtained average 97.86%. Experimentally, it found highly effective outperforms most existing state-of-the-art techniques. future, functional Near-Infrared Spectroscopy (fNIRS), ECG, Galvanic Skin Response (GSR) be employed with effectiveness technique.

Language: Английский

---

Learning-based estimation of operators’ psycho-physiological state

Expert Systems with Applications, Journal Year: 2025, Volume and Issue: unknown, P. 127097 - 127097

Published: March 1, 2025

Language: Английский

---

Exploring the Efficacy of Piezoelectric-Based Sensory Systems for Heart Rate Monitoring in Differentiating Stress vs. Relax Conditions

IIUM Engineering Journal, Journal Year: 2024, Volume and Issue: 25(2), P. 325 - 337

Published: July 14, 2024

Stress has diverse effects on human physiological reactions, and one such effect is heart rate (HR). The established methods to acquire HR by electrocardiogram (ECG) photoplethysmogram (PPG). ECG electrodes need be placed the chest, which can cause inconvenience not practical in daily life, while PPG signals are known contain more noise than ECG. Thus, this work aims investigate efficacy of a piezoelectric-based sensory system measuring using signal differentiate stressed relaxed conditions means statistical analysis. Two activities were conducted achieve goal. first experiment involved collecting analysing piezoelectric measure pulse (bpm) compare with from PPG. For second experiment, was calculated 20 subjects (male female, age ranging between 25) conditions. stress condition triggered two stressors: Stroop Colour Word Test Digit Span Test. Statistical analyses reveal strong positive correlation (HR) oximeter readings (r(12) = 0.993, p < 0.001), despite fact that values precisely identical. In addition, findings also indicate there significant mental states (stressed relaxed) (p<0.05). Employing within-subject design condition, results further illustrated elevated during (Mean±SD 72.395±0.097) diminished 71.615±0.126). Therefore, suggested been validated as an effective categorizing relaxation based signals. ABSTRAK: Tekanan mempunyai pelbagai kesan terhadap reaksi fisiologi manusia dan satu daripadanya adalah kadar denyut jantung Kaedah biasa bagi mengetahui melalui elektrokardiogram fotofetismogram Elektrod perlu dipasang pada dada di mana boleh menyebabkan ketidakselesaan tidak praktikal dalam kehidupan seharian, manakala isyarat diketahui mengandungi lebih banyak bunyi berbanding Oleh itu, kajian ini bertujuan mengkaji kecekapan ystem deria berdasarkan piezoeletrik mengukur menggunakan membezakan keadaan tertekan atau tenang cara analisis ystemic. Bagi mencapai tujuan ini, dua aktiviti dijalankan. Pertama eksperimen melibatkan pengumpulan nadi membandingkan daripada Eksperimen kedua, piezoelektrik dikira dari subjek (lelaki ystemic, berumur antara tertekan. Keadaan dibuat ystemi: Ujian Patah Warna Rentang Digit. Analisis ystemic mendedahkan hubungkait yang kuat bacaan oksimeter walaupun benar-benar serupa. Tambahan, penemuan menunjukkan terdapat penting tahap (tertekan tenang) Dengan mengaplikasi reka bentuk subjek, dapatan meningkat kurang telah diuji berkesan mengkategori jantung.

Language: Английский

---

Enhancing Human Stress Detection from Optimized and Selected Features Using Feed Forward Neural Network

Lecture notes in networks and systems, Journal Year: 2024, Volume and Issue: unknown, P. 251 - 260

Published: Jan. 1, 2024

Language: Английский

---

Predicting and Classifying Heart Rates Using Instantaneous Video Data

2022 IEEE 46th Annual Computers, Software, and Applications Conference (COMPSAC), Journal Year: 2023, Volume and Issue: unknown, P. 1076 - 1083

Published: June 1, 2023

Heart Rate (HR) and Variability (HRV) is an essential measurement to know the heart's cardiovascular condition. Many works have been done for measuring HR-HRV based on facial video non-invasively. In this paper, our previous work experience of by Remote photoplethysmography signals (rPPG) analysis, we built a prediction model from 10-second time series data extracted video. work, used instantaneous public dataset with several models predict HR-HRV, stress levels exclusively dataset. We here some popular algorithms appropriate task. also analyzed level classification gender subject using same videos 16 different classifiers resulting in almost perfect accuracy classifiers.

Language: Английский

---

Stress Responses of Examiners during Ophthalmic Examination Practices in Healthy Young Students

Applied Sciences, Journal Year: 2023, Volume and Issue: 13(18), P. 10250 - 10250

Published: Sept. 13, 2023

The stress reaction of trainees is an issue in the practices medical-related examinations that involve real-time decision making based on examiner–subject interactions. Goldmann perimetry (GP) test one these examinations. To evaluate students’ reactions practice GP test, stress-related substances and heart rate variability were measured forty students enrolled practice. While there was no significant increase during practice, significantly increased sympathetic activities observed at beginning tests. Moreover, plasma cortisol before tests showed a positive correlation to self-confidence scores, indicating students, especially those with higher anxious for upcoming unfamiliar subjects. Once began, they felt relieved procedures had learned repeatedly. On other hand, while average ACTH decreased secretion correlated positively duration, skillful participants less test. In practices, pre-training how deal subjects may be helpful reducing trainees, addition procedure itself.

Language: Английский

---