What is the clinical evidence from this study?

Study design: Other. Population: Stress (n=25). Intervention: Depth-separable convolutional neural networks (DSCNN) with continuous wavelet transform (CWT). Primary outcome: Accuracy of stress state classification.

What does this research mean for the field?

A depth-separable convolutional neural network model using continuous wavelet transform of ECG signals achieves 99.9% accuracy in classifying stress states without requiring complex feature extraction. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

February 26, 2025Open Access

Intelligent Stress Detection Using ECG Signals: Power Spectrum Imaging with Continuous Wavelet Transform and CNN

RMRodrigo Mateo-ReyesAutonomous University of Queretaro ICIrving A. Cruz-AlbarránAutonomous University of Queretaro LMLuis A. Morales‐HernandezAutonomous University of Queretaro

Key Result

A depth-separable convolutional neural network model using continuous wavelet transform of ECG signals achieved 99.9% accuracy in classifying stress states.

Structured PICO

Does a depth-separable convolutional neural network (DSCNN) model using continuous wavelet transform (CWT) on ECG signals accurately detect stress?

Population

25 samples from the SWELL-KW dataset evaluated for stress detection across neutral, time pressure, and interruption states.

Intervention

Depth-separable convolutional neural networks (DSCNN) model using continuous wavelet transform (CWT) on ECG signals

Outcome

Accuracy of stress detection across three states (neutral, time pressure, and interruptions)surrogate

A novel DSCNN model using continuous wavelet transform on ECG signals can highly accurately detect stress states without complex feature extraction.

Abstract

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

AI에게 질문

Bookmark

View Full Paper