Emotion can reflect the psychological and physiological health of human beings, and the main expression of human emotion is voice and facial expression. How to extract and effectively integrate the two modes of emotion information is one of the main challenges faced by emotion recognition. In this paper, a multi-branch bidirectional multi-scale time perception model is proposed, which can detect the forward and reverse speech Mel-frequency spectrum coefficients in the time dimension. At the same time, the model uses causal convolution to obtain temporal correlation information between different scale features, and assigns attention maps to them according to the information, so as to obtain multi-scale fusion of speech emotion features. Secondly, this paper proposes a two-modal feature dynamic fusion algorithm, which combines the advantages of AlexNet and uses overlapping maximum pooling layers to obtain richer fusion features from different modal feature mosaic matrices. Experimental results show that the accuracy of the multi-branch bidirectional multi-scale time sensing dual-modal emotion recognition model proposed in this paper reaches 97.67% and 90.14% respectively on the two public audio and video emotion data sets, which is superior to other common methods, indicating that the proposed emotion recognition model can effectively capture emotion feature information and improve the accuracy of emotion recognition.
Exploring the functional network during the interaction between emotion and cognition is an important way to reveal the underlying neural connections in the brain. Sparse Bayesian network (SBN) has been used to analyze causal characteristics of brain regions and has gradually been applied to the research of brain network. In this study, we got theta band and alpha band from emotion electroencephalogram (EEG) of 22 subjects, constructed effective networks of different arousal, and analyzed measurements of complex network including degree, average clustering coefficient and characteristic path length. We found that: ① compared with EEG signal of low arousal, left middle temporal extensively interacted with other regions in high arousal, while right superior frontal interacted less; ② average clustering coefficient was higher in high arousal and characteristic path length was shorter in low arousal.
Objective
To investigate the status quo and influencing factors of depression and anxiety in postoperative patients with thoracic neoplasms.
Methods
The general information questionnaire and Huaxi emotional-distress index scale (HEI) were adopted to survey 70 patients after surgery of thoracic neoplasms at the thoracic nursing outpatients from September to November 2016. There were 43 males and 27 females with age of 18-78 (56.20±11.34) years.
Results
The prevalence rate of depression and anxiety among postoperative patients with thoracic neoplasms was 50.0%, and moderate to severe negative emotions predominated. There was significant difference in educational levels, postoperative hospitalization and postoperative complications (P<0.05), while no significant difference in age, gender, disease types, complicated diseases, surgical procedures, pathological stages and hospitalization expenditures between patients with unhealthy emotions and normal emotions (P>0.05).
Conclusion
There is a high prevalence rate of negative emotion among postoperative patients with thoracic neoplasms. Educational levels, postoperative hospitalization and postoperative complications are important factors for negative emotion.
Focused on the world-wide issue of improving the accuracy of emotion recognition, this paper proposes an electroencephalogram (EEG) signal feature extraction algorithm based on wavelet packet energy entropy and auto-regressive (AR) model. The auto-regressive process can be approached to EEG signal as much as possible, and provide a wealth of spectral information with few parameters. The wavelet packet entropy reflects the spectral energy distribution of the signal in each frequency band. Combination of them gives a better reflect of the energy characteristics of EEG signals. Feature extraction and fusion are implemented based on kernel principal component analysis. Six emotional states from a public multimodal database for emotion analysis using physiological signals (DEAP) are recognized. The results show that the recognition accuracy of the proposed algorithm is more than 90%, and the highest recognition accuracy is 99.33%. It indicates that this algorithm can extract the feature of EEG emotion well, and it is a kind of effective emotion feature extraction algorithm, providing support to emotion recognition.
ObjectiveTo investigate the negative emotions of patients before cardiac surgery in West China Hospital in order to analyze the related factors.MethodsThe Huaxi emotional-distress index (HEI), a screening tool for mood disorders developed by the Mental Health Center of West China Hospital, was used for preoperative psychological evaluation of 1 968 adult patients hospitalized in cardiac surgery from March 2016 to July 2014. There were 835 males and 1 133 females at age of 49±13 years.Results Fifty-one patients (2.6%) had negative emotions, among whom 6 patients were screened for suicide risk. After intervention, none of them had serious consequences caused by adverse emotions, such as automatic discharge from hospital, avoidance of surgery and suicide.ConclusionThis study found that most of the cardiac surgery patients in West China Hospital have good psychological status before surgery, and a few suffered from negative emotions. “Huaxi emotional-distress index” is simple, effective and worth promoting.
As an important component of the event related potential (ERP), late positive potential (LPP) is an ideal component for studying emotion regulation. This study was focused on processing and analysing the LPP component of the emotional cognitive reappraisal electroencephalogram (EEG) signal. Firstly, we used independent component analysis (ICA) algorithm to remove electrooculogram, electromyogram and some other artifacts based on 16 subjects' EEG data by using EGI 64-channal EEG acquisition system. Secondly, we processed feature extraction of the EEG signal at Pz electrode by using one versus the rest common spatial patterns (OVR-CSP) algorithm. Finally, the extracted LPP component was analysed both in time domain and spatial domain. The results indicated that ① From the perspective of amplitude comparison, the LPP amplitude, which was induced by cognitive reappraisal, was much higher than the amplitude under the condition of watching neural stimuli, but lower than the amplitude under condition of watching negative stimuli; ② from the perspective of time process, the difference between cognitive reappraisal and watching after processing with OVR-CSP algorithm was in the process of range between 0.3 s and 1.5 s; but the difference between cognitive reappraisal and watching after processing with averaging method was during the process between 0.3 s and 1.25 s. The results suggested that OVR-CSP algorithm could not only accurately extract the LPP component with fewer trials compared with averaging method so that it provided a better method for the follow-up study of cognitive reappraisal strategy, but also provide neurophysiological basis for cognitive reappraisal in emotional regulation.
Objective To evaluate the changes in depressive symptoms and emotional responses in obstructive sleep apnea (OSA) patients after six months of continuous positive airway pressure (CPAP) therapy. Methods From June 2021 to December 2023, adult patients diagnosed with OSA at our hospital who were recommended for CPAP therapy as a first-line treatment were recruited. Demographic data (age, body mass index, gender), oxygen desaturation index, maximum duration of apnea and maximum duration of apnea were recorded. The patients were divided into a CPAP group and a non-CPAP group according to whether they were compliant to CPAP treatment. All patients completed questionnaires (including CES-D, DERS, ERS, and ESS) at 0, 1, 2, 4, and 6 months. Differences in general data and questionnaire results were compared between the two groups. Results The patients in the CPAP group showed significantly lower levels of depression and daytime sleepiness at 1, 2, 4, and 6 months compared with the non-CPAP group, with statistically significant differences (all P<0.05). Additionally, the CPAP group exhibited significantly lower scores in emotional responses and difficulties in emotion regulation across the same time points, with statistically significant differences (all P<0.05). In the non-CPAP group, increases in the apnea hypopnea index (AHI) and worsening emotional responses were key factors contributing to the exacerbation of depressive symptoms in OSA patients, with statistically significant differences (P<0.05). Conclusions CPAP therapy significantly improves depressive symptoms, emotional responses, and emotional regulation in OSA patients. Increases in the AHI and worsening emotional responses are primary factors leading to the worsening of depressive symptoms in OSA patients.
Analyzing the influence of mixed emotional factors on false memory through brain function network is helpful to further explore the nature of brain memory. In this study, Deese-Roediger-Mc-Dermott (DRM) paradigm electroencephalogram (EEG) experiment was designed with mixed emotional memory materials, and different kinds of music were used to induce positive, calm and negative emotions of three groups of subjects. For the obtained false memory EEG signals, standardized low resolution brain electromagnetic tomography algorithm (sLORETA) was applied in the source localization, and then the functional network of cerebral cortex was built and analyzed. The results show that the positive group has the most false memories [(83.3 ± 6.8)%], the prefrontal lobe and left temporal lobe are activated, and the degree of activation and the density of brain network are significantly larger than those of the calm group and the negative group. In the calm group, the posterior prefrontal lobe and temporal lobe are activated, and the collectivization degree and the information transmission rate of brain network are larger than those of the positive and negative groups. The negative group has the least false memories [(73.3 ± 2.2)%], and the prefrontal lobe and right temporal lobe are activated. The brain network is the sparsest in the negative group, the degree of centralization is significantly larger than that of the calm group, but the collectivization degree and the information transmission rate of brain network are smaller than the positive group. The results show that the brain is stimulated by positive emotions, so more brain resources are used to memorize and associate words, which increases false memory. The activity of the brain is inhibited by negative emotions, which hinders the brain’s memory and association of words and reduces false memory.
ObjectiveTo investigate the job satisfaction, emotional state and related factors of medical staff participating in online consultation of West China Internet Hospital during the COVID-19 epidemic.MethodsThrough literature review and expert consultation (Delphi method), the questionnaire was developed, and the online consulting medical staff of West China Hospital of Sichuan University were invited to conduct the questionnaire survey from 26 January to 19 June 2020, and finally the statistical analysis was summarized.ResultsA total of 132 valid questionnaires were retrieved. Of the 132 subjects, 127 people (96.2%) expressed satisfaction or special satisfaction with the online consulting office format; 103 respondents (78.0%) said that online consulting did not affect or completely did not affect the work and life; 81 people (61.4%) consulted online more than 5 days a week, and 108 people (81.8%) worked within 2 hours a day; the vast majority (97.7%) of the research subjects were satisfied with the content of the training materials and the related support work of the coordination group. Only 29 (22.0%) of the study participants believed that the epidemic caused negative emotions, mainly due to the severity of the epidemic.ConclusionThe online consulting medical staff are satisfied with the office form, training materials and coordination work group of the COVID-19 epidemic, and think that it does not affect their work and life. 22.0% of medical staff have negative emotions, and the severity of the epidemic is the main reason.
The result of the emotional state induced by music may provide theoretical support and help for assisted music therapy. The key to assessing the state of emotion is feature extraction of the emotional electroencephalogram (EEG). In this paper, we study the performance optimization of the feature extraction algorithm. A public multimodal database for emotion analysis using physiological signals (DEAP) proposed by Koelstra et al. was applied. Eight kinds of positive and negative emotions were extracted from the dataset, representing the data of fourteen channels from the different regions of brain. Based on wavelet transform, δ, θ, α and β rhythms were extracted. This paper analyzed and compared the performances of three kinds of EEG features for emotion classification, namely wavelet features (wavelet coefficients energy and wavelet entropy), approximate entropy and Hurst exponent. On this basis, an EEG feature fusion algorithm based on principal component analysis (PCA) was proposed. The principal component with a cumulative contribution rate more than 85% was retained, and the parameters which greatly varied in characteristic root were selected. The support vector machine was used to assess the state of emotion. The results showed that the average accuracy rates of emotional classification with wavelet features, approximate entropy and Hurst exponent were respectively 73.15%, 50.00% and 45.54%. By combining these three methods, the features fused with PCA possessed an accuracy of about 85%. The obtained classification accuracy by using the proposed fusion algorithm based on PCA was improved at least 12% than that by using single feature, providing assistance for emotional EEG feature extraction and music therapy.
