Exploring the Intrinsic Features of EEG Signals via Empirical Mode Decomposition for Depression Recognition

Abstract

Depression is a severe psychiatric illness that causes emotional and cognitive impairment and has a considerable impact on patients' thoughts, behaviors, feelings and well-being. Moreover, methods for recognizing and treating depression are lacking in clinical practice. Electroencephalogram (EEG) signals, which objectively reflect the internal workings of the brain, is a promising and objective tool for recognizing and diagnosing of depression and enhancing clinical effects. However, previous EEG feature extraction methods have not performed well when exploring the intrinsic characteristics of highly complex and nonstationary EEG signals. To address this issue, we propose a regularization parameter-based improved intrinsic feature extraction method of EEG signals via empirical mode decomposition (EMD), which mines the intrinsic patterns in EEG signals, for depression recognition. Furthermore, our method can effectively solve the problem that EMD fails to extract intrinsic features.In this method, we first select an appropriate regularization parameter to generate the regularization matrix. Next, we calculate the sum of the matrix products of the IMFs and the regularization matrix and leverage the inverse of this matrix to extract the intrinsic features. The classification results of our method on four EEG datasets reached 0.8750, 0.8850, 0.8485 and 0.7768, respectively. In addition, compared with the iEMD method, our method requires less computational costs. These results support our claim that our method can effectively strengthen the depression recognition performance,

A Random Forest Weights and 4-Dimensional Convolutional Recurrent Neural Network for EEG Based Emotion Recognition

“…The EEG data of each subject are treated as an individual dataset and subsequently partitioned into training and test subsets. One-tenth of the subjects' data were selected as the validation set during training and the other nine-tenths of the data were used for training [28], [29]. The overall classification performance of the emotion recognition model was then determined by computing the average accuracy across all test subsets.…”

Laser self-mixing interference displacement signal filtering method based on empirical mode decomposition and wavelet threshold

“…But the approach is not without its pitfalls, such as the risk of mode mixing issues. In recent years, variational mode decomposition (VMD) has garnered attention for its robustness against noise and ability to prevent mode mixing, a limitation seen in EMD [9,10]. Yet another alternative is Kalman Filtering, a recursive state estimation method [11].…”

Decoding Algorithm of Motor Imagery Electroencephalogram Signal Based on CLRNet Network Model

“…Additionally, since there are individual differences in the collected MI-EEG signals, each participant requires model training from the beginning, resulting in substantial computational costs [ 12 ]. Since the features of MI-EEG signals vary over time and can produce large differences among individuals, the selection of reliable and stable feature extraction methods is currently an important research direction [ 13 ].…”