Search

Electroencephalography (EEG) is widely used for mental stress classification, but effective feature extraction and transfer across subjects remain challenging due to its variability. In this paper, a novel deep neural network combining convolutional neural network (CNN) and adversarial theory, named symmetric deep convolutional adversarial network (SDCAN), is proposed for stress classification based on EEG. The adversarial inference is introduced to automatically capture invariant and discriminative features from raw EEG, which aims to improve the classification accuracy and generalization ability across subjects. Experiments were conducted with 22 human subjects, where each participant’s stress was induced by the Trier Social Stress Test paradigm while EEG was collected. Stress states were then calibrated into four or five stages according to the changing trend of salivary cortisol concentration. The results show that the proposed network achieves improved accuracies of 87.62% and 81.45% on the classification of four and five stages, respectively, compared to conventional CNN methods. Euclidean space data alignment approach (EA) was applied and the improved generalization ability of EA-SDCAN across subjects was also validated via the leave-one-subject-out-cross-validation, with the accuracies of four and five stages being 60.52% and 48.17%, respectively. These findings indicate that the proposed SDCAN network is more feasible and effective for classifying the stages of mental stress based on EEG compared with other conventional methods.

Depression disorder has become one of the major psychological diseases endangering human health. Researcher in the affective computing community is supporting the development of reliable depression severity estimation system, from multiple modalities (speech, face, text), to assist doctors in their diagnosis. However, the limited amount of annotated data has become the main bottleneck restricting the study on depression screening, especially when deep learning models are used. To alleviate this issue, in this work we propose to use Deep Convolutional Generative Adversarial Network (DCGAN) for features augmentation to improve depression severity estimation from speech. To the best of our knowledge, this approach is the first attempt to apply the Generative Adversarial Network for depression severity estimation from speech. Besides, to measure the quality of the augmented features, we propose three different measurement criteria, characterizing the spatial, frequency and representation learning of the augmented features. Finally, the augmented features are used to train depression estimation models. Experiments are carried out on speech signals from the Audio Visual Emotion Challenge (AVEC2016) depression dataset, and the relationship between the model performance and data size is explored. Our experimental results show that: 1) The combination of the three proposed evaluation criteria can effectively and comprehensively evaluate the quality of the augmented features. 2) When increasing the size of the augmented data, the performance of depression severity estimation gradually improves and the model converges to a certain stable state. 3) The proposed DCGAN based data augmentation approach effectively improves the performance of depression severity estimation, with the root mean square error (RMSE) reduced to 5.520 and mean absolute error (MAE) reduced to 4.634, which is better than most of the state of the art results on AVEC 2016.

In the temporal process of expressing the emotions, some intervals embed more salient emotion information than others. In this paper, by introducing an attentive temporal pooling module into the deep neural network (DNN) architecture, we present a simple but effective speech emotion recognition (SER) framework, which is able to automatically highlight the emotionally salient segments while suppressing the influence of less relevant ones. For an input speech utterance, the extracted feature sequence of hand-crafted low-level descriptors (LLDs) are evenly split into several overlapping temporal segments, and the segment-level features are computed by performing functionals on the LLDs of each segment. These segment-level features are then input into a DNN model outputting the emotion probabilities as well as the more condensed representation of each segment. An attentive temporal pooling module, consisting of an auxiliary DNN and a Gaussian Mixture Model (GMM), is proposed to learn the emotional saliency weights of different temporal segments from the condensed representations, which are then assigned to the segment-level emotion probabilities for the final utterance-level prediction. Notably, the attentive temporal pooling module and the DNN architecture for feature abstraction can be jointly trained using only the utterance-level labels, while without any frame-level or segment-level supervisory information. Experimental results on the three public released emotion datasets RML, EMO-DB, and IEMOCAP show that the proposed framework obtains state-of-the-art performance on SER.

Sparse representation has exhibited excellent performance in face recognition. However, this method requires some areas for improvement, especially on insufficient face samples. We aim to design a simple and efficient method to improve sparse representation to solve problems with a small sample size. This paper provides two primary contributions that are very effective in small sample face recognition. First, in order to enhance recognition robustness, we designed an intuitive and mathematically controllable transfer learning method of sparse representation by introducing labeled samples. Second, to obtain high recognition accuracy, we developed a weighted fusion scheme to integrate the sparse representation results generated from original and labeled samples. In the ORL dataset, our algorithm’s highest accuracy rate is 95%. In the FERET dataset, our highest classification accuracy rate is 95%. In the more complex LFW dataset, our highest classification accuracy rate has also reached 83.33%. This shows that our experimental results demonstrate that the proposed method can obtain sufficient performance, whereas the weighted fusion scheme can take advantage of sparse representation on the basis of original and labeled samples. This paper will be very useful for identification based on the Internet-of-Medical-Things.

There is a general agreement in the brain computer interface community that the feature extracting method called the common spatial pattern (CSP) combined with nonlinear classifiers can provide excellent results in some cases. However, CSP is also known to be very sensitive to noise and prone to over fitting, and the performance of this spatial filter is closely related to the operational frequency band of electroencephalogram data. To address this issue, we propose a new method FB-TRCSP+RF based on CSP and random forest. The FB-TRCSP is combined by the 8th-order Butterworth bandpass-filters and the CSP with Tikhonov regularization, which is a more robust feature extraction method compared to the CSP. Then, the model is applied to an experimental data set collected from 14 subjects and is compared with the non-regularization method FB-CSP+RF. The results show that the method we proposed yields relatively higher median classification accuracies and shows a stronger ability in subject-to-subject learning compared to prevailing approaches.