Predictive Modeling for Heart Rate: A Comparative Analysis of LSTM, XGBoost, and LightGBM DOI

Jørgen Robin Våge,

Saptanil Ghose,

Rudra Pratap Deb Nath

et al.

Published: Dec. 13, 2023

This study evaluates the effectiveness of Long Short-Term Memory (LSTM) networks, XGBoost, and LightGBM in predicting heart rates across diverse time windows. Leveraging a dataset from 22 users multiple observation windows, this research significantly broadens current understanding rate prediction field personalized health monitoring sports science. Performance metrics such as Mean Absolute Error (MAE), Squared (MSE), Root (RMSE), Scatter Index (SI) were employed for assessment. Over intervals 30, 60, 180 seconds, both XGBoost outperformed LSTM terms MAE, MSE, RMSE, SI. These results suggest that are superior options high-accuracy prediction, underscoring their potential utility healthcare athletic performance applications.

Language: Английский

A machine-learning approach for stress detection using wearable sensors in free-living environments DOI

Mohamed Abd Al-Alim,

Roaa I. Mubarak, Nancy M. Salem

et al.

Computers in Biology and Medicine, Journal Year: 2024, Volume and Issue: 179, P. 108918 - 108918

Published: July 18, 2024

Language: Английский

Citations

8
Elicitation of Anxiety Without Time Pressure and Its Detection Using Physiological Signals and Artificial Intelligence: A Proof of Concept DOI Creative Commons
Antonio Di Tecco, Francesco Pistolesi, Beatrice Lazzerini

et al.

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 22376 - 22393

Published: Jan. 1, 2024

Stress can be defined as a state of anxiety (or mental tension) caused by particular situation. Everybody experiences stress to some level, but how we respond significantly affects our well-being. Various events generate that leads stress. For example, not having enough time complete task or being late are situations where (and stress) depends on temporal factor: the scarcity time. But people also slide into they live in condition causes them tense, independently The studies eliciting laboratory settings have less widely considered this variant. This paper presents proof concept (PoC) investigated possibility stimulating without pressure through purposely edited horror movie trailer, giving new insights emotional evoked controlled audiovisual stimuli. PoC comprised an AI-based classifier detect person's emotion among anxiety , xmlns:xlink="http://www.w3.org/1999/xlink">relaxation and xmlns:xlink="http://www.w3.org/1999/xlink">none two based galvanic skin response (GSR), photoplethysmogram (PPG), heart rate (HR), achieving accuracy higher than 96%. Key application areas include media marketing, psychology. Media producers could improve their content capture audience better; psychologists create tailored exposure promote gradual desensitization triggers.

Language: Английский

Citations

4
Stress management with HRV following AI, semantic ontology, genetic algorithm and tree explainer DOI Creative Commons
Ayan Chatterjee, Michael A. Riegler,

K Ganesh

et al.

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Feb. 17, 2025

Heart Rate Variability (HRV) serves as a vital marker of stress levels, with lower HRV indicating higher stress. It measures the variation in time between heartbeats and offers insights into health. Artificial intelligence (AI) research aims to use data for accurate level classification, aiding early detection well-being approaches. This study's objective is create semantic model features knowledge graph develop an accurate, reliable, explainable, ethical AI predictive analysis. The SWELL-KW dataset, containing labeled conditions, examined. Various techniques like feature selection dimensionality reduction are explored improve classification accuracy while minimizing bias. Different machine learning (ML) algorithms, including traditional ensemble methods, employed analyzing both imbalanced balanced datasets. To address imbalances, various formats oversampling such SMOTE ADASYN experimented with. Additionally, Tree-Explainer, specifically SHAP, used interpret explain models' classifications. combination genetic algorithm-based using Random Forest Classifier yields effective results datasets, especially non-linear features. These optimized play crucial role developing management system within Semantic framework. Introducing domain ontology enhances representation acquisition. consistency reliability Ontology assessed Hermit reasoners, reasoning performance measure. significant indicator stress, offering its correlation mental well-being. While non-invasive, interpretation must integrate other assessments holistic understanding individual's response. Monitoring can help evaluate strategies interventions, individuals maintaining

Language: Английский

Citations

0
Intelligent Stress Detection Using ECG Signals: Power Spectrum Imaging with Continuous Wavelet Transform and CNN DOI Open Access
Rodrigo Mateo-Reyes, Irving A. Cruz‐Albarran, Luis A. Morales-Hernández

et al.

Journal of Experimental and Theoretical Analyses, Journal Year: 2025, Volume and Issue: 3(1), P. 6 - 6

Published: Feb. 26, 2025

Stress is a natural response of the organism to challenging situations, but its accurate detection due subjective nature. This study proposes model based on depth-separable convolutional neural networks (DSCNN) analyze heart rate variability (HRV) and detect stress. Electrocardiogram (ECG) signals are pre-processed remove noise ensure data quality. The then transformed into two-dimensional images using continuous wavelet transform (CWT) identify pattern recognition in time–frequency domain. These representations classified DSCNN determine presence methodology has been validated SWELL-KW dataset, achieving an accuracy 99.9% by analyzing three states (neutral, time pressure, interruptions) 25 samples experiment, scanning acquired signal every 5 s for 45 min per state. proposed approach characterized ability ECG means short duration sampling, classification stress without need complex feature extraction processes. efficient tool analysis from biomedical signals.

Language: Английский

Citations

0
Breaking Barriers in Stress Detection: An Inter-Subject Approach Using ECG Signals DOI
Shahane Tigranyan,

A. V. Martirosyan

2022 IEEE 46th Annual Computers, Software, and Applications Conference (COMPSAC), Journal Year: 2024, Volume and Issue: unknown, P. 1850 - 1855

Published: July 2, 2024

Language: Английский

Citations

0
Explaining Predicted Stress Levels in employed Individuals DOI

Latika Meelu,

Gebremariam Assres

Published: May 24, 2024

Language: Английский

Citations

0
Mental Stress Assessment in Working Environment for an Individual Using Wearable Sensor of EEG and Pulse Signal Measured with Help of Deep Learning Algorithm DOI
Karthikeyan Venkatesan Munivel,

S. Bhuvaneshwar,

V.A. Nishanth

et al.

IFIP advances in information and communication technology, Journal Year: 2024, Volume and Issue: unknown, P. 80 - 92

Published: Dec. 19, 2024

Language: Английский

Citations

0
Meta-Heuristic Feature Optimization for Predictive Analysis on HRV Dataset and Semantic Knowledge Representation for Stress Management: A Case-Study Towards Ethical AI DOI Creative Commons
Ayan Chatterjee,

K Ganesh,

Michael Reigler

et al.

Research Square (Research Square), Journal Year: 2023, Volume and Issue: unknown

Published: Sept. 11, 2023

Abstract Background: Heart Rate Variability (HRV) is intimately associated with stress and can serve as a valuable indicator of the individual’s level. HRV variation in length time between heartbeats. Lower higher levels, while indicates better resilience adaptability. The parameters be classified − time-domain, frequency-domain, non-linear. Parameters employed to assess individuals’ health observe effects interventions such exercise, reduction, medication. Research field artificial intelligence (AI) ongoing that attempts classify based on data. HRV, which physiological health, has received attention potential component incorporate into models predict levels accurately. Monitoring offer insights interplay mental aiding early detection holistic approaches well-being. Objective: primary goal this study perform semantic modeling vital features knowledge graph, followed by, developing an accurate, reliable, explainable, ethical AI model pipeline predictive analysis Methods: In regard, we have considered well-known multimodal SWELL work (SWELL−KW) dataset case represents following conditions no stress, pressure, interruption. selected shows labeled relationship deemed suitable for study. We explored different feature selection dimensionality reduction techniques extract relevant from enhance classification accuracy reduced bias. used various machine learning (ML) algorithms (e.g., traditional ensemble) imbalanced balanced datasets. data formats scaled, normalized, standardized) oversampling Synthetic Minority Oversampling Technique (SMOTE) Adaptive (ADASYN)) generate synthetic samples minority class. Tree-Explainer Shapley Additive Explanations (SHAP)) explain classifications. Results: As are non-linear, genetic algorithm-based Random Forest Classifier produced highest result both optimized set been beneficial design develop management system Semantic framework. Therefore, introduced concept domain ontology represent obtained knowledge. consistency Ontology evaluated Hermit reasoners reasoning time. Conclusions: Overall, serves marker provide health. It’s crucial recognize non-invasive its interpretation should combined other subjective objective measures order comprehensively understand response. monitoring may help individuals effectiveness interventions.

Language: Английский

Citations

1
Meta-Heuristic Feature Optimization for Predictive Analysis on HRV Dataset and Semantic Knowledge Representation for Stress Management: A Case-Study Towards Ethical AI DOI Creative Commons

K Ganesh,

Ayan Chatterjee, Michael A. Riegler

et al.

Research Square (Research Square), Journal Year: 2023, Volume and Issue: unknown

Published: Sept. 13, 2023

Abstract Background: Heart Rate Variability (HRV) is intimately associated with stress and can serve as a valuable indicator of the individual’s level. HRV variation in length time between heartbeats. Lower higher levels, while indicates better resilience adaptability. The parameters be classified − time-domain, frequency-domain, non-linear. Parameters employed to assess individuals’ health observe effects interventions such exercise, reduction, medication. Research field artificial intelligence (AI) ongoing that attempts classify based on data. HRV, which physiological health, has received attention potential component incorporate into models predict levels accurately. Monitoring offer insights interplay mental aiding early detection holistic approaches well-being. Objective: primary goal this study perform semantic modeling vital features knowledge graph, followed by, developing an accurate, reliable, explainable, ethical AI model pipeline predictive analysis Methods: In regard, we have considered well-known multimodal SWELL work (SWELL−KW) dataset case represents following conditions no stress, pressure, interruption. selected shows labeled relationship deemed suitable for study. We explored different feature selection dimensionality reduction techniques extract relevant from enhance classification accuracy reduced bias. used various machine learning (ML) algorithms (e.g., traditional ensemble) imbalanced balanced datasets. data formats scaled, normalized, standardized) oversampling Synthetic Minority Oversampling Technique (SMOTE) Adaptive (ADASYN)) generate synthetic samples minority class. Tree-Explainer Shapley Additive Explanations (SHAP)) explain classifications. Results: As are non-linear, genetic algorithm-based Random Forest Classifier produced highest result both optimized set been beneficial design develop management system Semantic framework. Therefore, introduced concept domain ontology represent obtained knowledge. consistency Ontology evaluated Hermit reasoners reasoning time. Conclusions: Overall, serves marker provide health. It’s crucial recognize non-invasive its interpretation should combined other subjective objective measures order comprehensively understand response. monitoring may help individuals effectiveness interventions.

Language: Английский

Citations

1
Chronic Stress Recognition Based on Time-Slot Analysis of Ambulatory Electrocardiogram and Tri-Axial Acceleration DOI
Jiayu Li, Manman Wang, Feifei Zhang

et al.

IEEE Transactions on Affective Computing, Journal Year: 2023, Volume and Issue: 15(3), P. 1178 - 1189

Published: Oct. 23, 2023

Stress, especially chronic stress, is a high risk factor of many physical and mental health problems. This work acquired 702 days full-day ambulatory electrocardiogram (ECG) Tri-axial acceleration (T-ACC) data from 104 healthy college students realized stress recognition through signal processing, statistical test machine learning. We divided the 24 hours day into 153 time slots, calculated 30 features ECG T-ACC in each slot. Statistical above subjects with no labels showed that altered autonomic nervous control heart not only daily activity but also rest at night, leading to smaller rate variability, faster less complexity heartbeat rhythm. More specifically, parasympathetic night was weakened by stress. expressed as 30×153 matrix, applied four-layer fully connected neural network classify labels, obtained 88.17% detection accuracy leave-one-subject-out cross test.

Language: Английский

Citations

1