Performance evaluation of multi-channel electroencephalogram signal (EEG) based time frequency analysis for human emotion recognition

Wagh, Kalyani P.; Vasanth, K.

doi:10.1016/j.bspc.2022.103966

Cited by 26 publications

(11 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ACO–CNN–LSTM proposed in this paper was compared with the existing references, and the results of the comparison were shown in Table 9 . As shown in Table 9 , when both extracted DE as feature data for classification, the average accuracy of emotion classification of the proposed ACO–CNN–LSTM method in this paper was higher than references [ 22 , 26 , 49 , 52 , 53 ]; When multiple features were used as classification data, the average accuracy of emotion classification by ACO–CNN–LSTM was still higher than references [ 18 , 54 , 55 ]. In terms of channels, the proposed ACO–CNN–LSTM was advanced in channel selection, and obtained the effective channels and the higher classification accuracy.…”

Section: Discussionmentioning

confidence: 73%

See 1 more Smart Citation

Optimal channel dynamic selection for Constructing lightweight Data EEG-based emotion recognition

Zhang,

Xu,

Zhang

et al. 2024

Heliyon

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 73%

“…It ultimately achieved 89 % accuracy on the SEED dataset. Wagh K P et al [ 18 ]used wavelet transform and extracted features such as Power Spectral density (PSD), Energy, Standard Deviation, and Variance, to classify the emotional states. Using three classifiers and the final decision tree had the highest accuracy of 71.52 %.…”

Section: Introductionmentioning

confidence: 99%

Optimal channel dynamic selection for Constructing lightweight Data EEG-based emotion recognition

Zhang,

Xu,

Zhang

et al. 2024

Heliyon

View full text Add to dashboard Cite

“…By creating a comprehensive method for gathering physiological emotional databases, Chanel et al [29] proved that employing EEG signals for arousal evaluation was feasible. Wagh and Vasanth [30] achieved maximum classi cation rates of 71.52% and 60.19% with decision tree and k nearest neighbour techniques, respectively, using a range of classi ers and a discrete wavelet transform to identify distinct emotions in EEG signals.…”

Section: An Improved Radial Basis Function Neural Network Methods For...mentioning

confidence: 99%

Towards non-invasive PTSD diagnosis: Utilising EEG based Emotion Recognition with the DEAP Database

Sudha,

2024

Preprint

View full text Add to dashboard Cite

Post-Traumatic Stress Disorder (PTSD) poses a significant challenge in mental health diagnosis, necessitating innovative and non-invasive approaches. This paper explores the efficacy of emotion recognition through electroencephalography (EEG) as a potential diagnostic tool for PTSD. Leveraging the rich resource of the DEAP EEG database, this study focuses on employing statistical features, namely mean, standard deviation, kurtosis, and Hjorth parameters, to ascertain emotional states associated with PTSD. This work outlines the pressing need for effective and non-invasive PTSD diagnosis methods, emphasizing the potential of emotion recognition as a groundbreaking approach. EEG, with its ability to capture neural activity in real-time, emerges as a promising biomarker for decoding emotional responses associated with PTSD. The paper employs a 1D Convolutional Neural Network (1D CNN) as the classifier algorithm, demonstrating its efficacy in discriminating between valence, arousal, and liking associated with PTSD-related emotional responses. Results indicate a remarkable classification accuracy of 97.18%, highlighting the potential of the proposed approach for PTSD diagnosis. This research contributes a non-invasive diagnostic method, bridging the gap between neuroscience, emotion recognition, and mental health, ultimately paving the way for more effective and accessible PTSD assessment tools.

show abstract

“…Wagh et al [6] employed Support Vector Machine (SVM), k-Nearest Neighbor (kNN), and Decision Tree (DT) algorithms to classify positive, neutral, and negative emotions using time and time-frequency domain features extracted from various channels of electroencephalogram (EEG) data. The classifiers were trained on the SJTU emotion EEG dataset (SEED), resulting in an accuracy of 72.46% for DT and 60.19% for kNN.…”

Section: Related Workmentioning

confidence: 99%

Deep Learning Models for Stress Analysis in University Students: A Sudoku-Based Study

Chen

Lee

2023

Sensors

View full text Add to dashboard Cite

Due to the phenomenon of “involution” in China, the current generation of college and university students are experiencing escalating levels of stress, both academically and within their families. Extensive research has shown a strong correlation between heightened stress levels and overall well-being decline. Therefore, monitoring students’ stress levels is crucial for improving their well-being in educational institutions and at home. Previous studies have primarily focused on recognizing emotions and detecting stress using physiological signals like ECG and EEG. However, these studies often relied on video clips to induce various emotional states, which may not be suitable for university students who already face additional stress to excel academically. In this study, a series of experiments were conducted to evaluate students’ stress levels by engaging them in playing Sudoku games under different distracting conditions. The collected physiological signals, including PPG, ECG, and EEG, were analyzed using enhanced models such as LRCN and self-supervised CNN to assess stress levels. The outcomes were compared with participants’ self-reported stress levels after the experiments. The findings demonstrate that the enhanced models presented in this study exhibit a high level of proficiency in assessing stress levels. Notably, when subjects were presented with Sudoku-solving tasks accompanied by noisy or discordant audio, the models achieved an impressive accuracy rate of 95.13% and an F1-score of 93.72%. Additionally, when subjects engaged in Sudoku-solving activities with another individual monitoring the process, the models achieved a commendable accuracy rate of 97.76% and an F1-score of 96.67%. Finally, under comforting conditions, the models achieved an exceptional accuracy rate of 98.78% with an F1-score of 95.39%.

show abstract

Performance evaluation of multi-channel electroencephalogram signal (EEG) based time frequency analysis for human emotion recognition

Cited by 26 publications

References 28 publications

Optimal channel dynamic selection for Constructing lightweight Data EEG-based emotion recognition

Optimal channel dynamic selection for Constructing lightweight Data EEG-based emotion recognition

Towards non-invasive PTSD diagnosis: Utilising EEG based Emotion Recognition with the DEAP Database

Deep Learning Models for Stress Analysis in University Students: A Sudoku-Based Study

Contact Info

Product

Resources

About