In order to analyze the cerebral cortex activated by the multi-modality consciousness task with additional Chinese character read silently, spatial source location analysis was used to analyze the brain electrical signals related to consciousness tasks. First, an independent component analysis algorithm is used to decompose the EEG signals of all electrodes into their own independent components. Each component can be considered as an equivalent dipole, and its position and direction in the cerebral cortex are determined; Analyze and determine dipoles related to consciousness tasks; finally analyze and contrast dipole differences between multimodal consciousness tasks and single modality consciousness tasks. Ten subjects were tested and the results showed that the average number of dipoles associated with the two multimodal consciousness tasks was 5 and 4.3, respectively, which were more than single-mode tasks, and the order of dipoles was The average classification accuracy of the four tasks is consistent, indicating that more activation of the cerebral cortex is conducive to improving the classification accuracy, and the validity and rationality of the experiment are verified.

0 Preface

Human-computer interaction technology has been widely studied in recent years, especially the development of human bioelectricity as an interactive medium. The Brain Computer Interface (BCI) is a human-computer interaction technology that uses the brain's signal to control ancillary equipment. It provides new rehabilitation methods for those with motor impairments, giving users new ways of communication and interaction. The BCI system has been successfully used for the control of electric wheelchairs, text input, robots and smart home control.

There are many ways to detect brain activity. Electroencephalography (EEG) has the advantages of high security, low cost, and so on. Therefore, this method is most popular in BCI. There are multiple experimental paradigms that produce distinguishable EEG signals. Different algorithms are used to analyze the EEG of these experimental paradigms. P300 evoked potentials often use time domain analysis algorithms such as analysis of variance, coherence averaging, and correlation analysis. The EEG signals generated by steady-state visual evoked and motion imaging have certain frequency characteristics and can be analyzed by time-frequency analysis algorithms such as short-time Fourier transform and wavelet transform [5]. When studying multi-channel EEG signals, space-time analysis algorithms such as Sparse Common Spatial Pattern (SCSP), which introduces channel selection, can also be used [6].

In the research work that considered increasing the BCI operation dimension and classification accuracy through experimental design, the task group proposed a multimodal conscious task experimental paradigm with additional related Chinese character silent readings, and comprehensively verified the time, space, and frequency. The validity and rationality of the experimental paradigm. Compared to a single thinking activity, the task of multimodal awareness with the addition of Chinese characters for reading is a combination of multiple higher-order thinking activities, which involves more cerebral cortex. In this paper, spatial source location analysis is used to study EEG signals, analyze and compare the cerebral cortex activated by different consciousness tasks to study the physiological mechanism of multimodal consciousness tasks, and help BCI research and design.

1 experimental design

In the experiment, 10 subjects (22 males and 3 females) aged 22 to 28 years participated and were identified as S1~S10. Six subjects participated in similar experiments. Others participated in the brain electricity experiment for the first time. Subjects were right-handed and had good physical condition and had enough rest before the experiment. The subject sat down to complete the entire experiment and received a message by looking at the 22-inch LCD screen 1 m ahead. Subjects signed the Informed Consent after understanding the purpose of the experiment and the precautions.

In order to compare the effect of conscious tasks with Chinese characters read silently, the whole experiment includes two-step experiments. The first-step experiment is a multi-modality experiment paradigm, and the second-step experiment is a single-modality experiment paradigm. This experiment is no feedback experiment. The specific content is shown in Figure 1. The two characters “Left” and “壹” appear randomly as prompt messages. During the idle period, the screen will show an asterisk "*" and the subject will remain relaxed for 2 s. A small cross “+” of 1 s is then displayed to prepare the subject. A Chinese character will appear afterwards for 1 s. Then a 4 s black screen is displayed as an imaginary period. If the Chinese character is “left”, the subject imagines the body as left-handed and silently read “left” in the first-step experiment. In the second-step experiment, only imagine the body is left-handed. If the Chinese character is “壹”, the subject imagines “壹” when reading the “壹” stroke in the first-step experiment. In the second-step experiment, the subject only imagines writing a stroke. After the 4 s consciousness task is over, one experiment is completed and it can relax for 0.5~1.5 s. Each tip appears 15 times in a set of experiments and each subject completes 5 sets of experiments on the same day. After one set of experiments, subjects can rest for 5 min.

2 Data Acquisition

The silent reading of Chinese characters is a language activity. Imagining that the body rotates to the left involves spatial imagination. Imagining and writing Chinese characters requires the coordination of language cortex and sports cortex. The experimental electrodes are shown in Fig. 2. They conform to the International 10/20 System Lead Setup and cover the 35-lead electrodes of the cerebral cortex in the Wilnick, Broca, and motor sensory cortices. The instrument used in the experiment was SynAmps 2 system from Neuroscan, USA. Two bipolar electrodes are placed to record the eye from the horizontal and vertical directions, respectively. In order to reduce the 50 Hz power frequency interference, a ground electrode should be provided on the forehead of the head. In addition, when the EEG signal is collected, the reference electrode needs to be placed on the top of the head. In order to reduce the signal distortion during the experiment, the contact resistance of all electrodes needs to be maintained below 5 kΩ. The system's band-pass filter is set to 0.1 to 100 Hz and the signal sampling rate is set to 250 Hz.

3 analysis methods

3.1 Pretreatment

After data acquisition, the Ocular Artifact Reduction function of the SCAN 4.5 software included in the system is first used to remove the electrooculogram signal. The useful information in adult EEG signals is mainly in alpha and beta waves. Therefore, the signal is filtered through a 4-45 Hz band-pass filter to remove noise.

3.2 Independent Component Analysis

Independent Component Analysis (ICA) is an important research method in the signal processing field and belongs to the blind source separation algorithm [8]. After ICA decomposition, useful signals and noise signals can be separated to perform feature extraction or brain electricity mapping.

3.3 Time-frequency analysis

After the ICA is decomposed into its own independent components, the EEG signal can be analyzed by time-frequency analysis to determine whether the independent component is related to thinking activity. Event-Related Spectral Perturbation (ERSP) is an improvement in the conventional time-frequency analysis algorithm, which can highlight the amount of change in EEG energy relative to idleness when the task is conscious [9]. The ERSP analyzes the relative changes in the energy spectrum over the time axis and the baseline energy spectrum that appears.

If the energy spectrum is increased or decreased after entering the imagination period, it means that the independent component is related to thinking activity, and vice versa. Take subject S6 imagine left-handedness while reading "left" silently as an example. The ERSP diagram of his first and fourth independent components after 4-45 Hz band-pass filtering is shown in Fig.3. The Chinese character prompt appears in the position of the ERSP diagram 0 s in Fig. 3. After 1 s, the subject starts the corresponding imaginary activity. The energy of the 4th independent component of subject S6 in the 9-15 Hz range increases with the progress of thinking, and his first independent component has no similar features. Therefore, it can be judged that the thinking activity is related to the fourth component and not related to the first component.

3.4 Equivalent Dipole Source Location Analysis

After obtaining independent components related to consciousness tasks by time-frequency analysis, spatial source location analysis can be used to inversely derive important information such as position, intensity, and direction of neural activity sources in the brain, so that the cerebral cortex activated by the consciousness task can be judged.

Equivalent dipole analysis was chosen in this paper. The algorithm considers event-related neurons as one or several current dipoles that are active in the brain [10]. The direction and position of these dipoles are determined by minimizing the residual error:

Where L is the lead field matrix, j is the transition dipole moment, and U is the ICA decomposed independent component matrix.

The detailed operation flow is:

(1) ICA decomposition of EEG signals using EEGLAB toolbox [11], in which decomposition algorithm adopts extended Infomax method;

(2) Select the BEM for the head model, then calculate the corresponding dipole through the DIPFIT 2.2 plug-in in the toolbox for the independent component;

(3) Scan the approximate position of the dipole with a three-dimensional coarse network;

(4) Taking the position obtained in the previous step as the initial value, the non-linear cross-fitting algorithm is used to calculate the exact position of the dipole.

4 Results and Analysis

The purpose of this paper is to compare the intrinsic differences between multimodal consciousness tasks and single consciousness tasks by spatially locating EEG signals related to consciousness tasks, and finally to prove multimodality from the perspective of time, space, and frequency. The feasibility and rationality of the consciousness task.

Take the subject S6 imagining the body as left-handed and silently reading the “left” as an example. Figure 3 shows the process of spatial source location. In the two-step experiment, 35 electrodes were used to record EEG signals. Therefore, after ICA decomposition, 35 components were firstly obtained, and component 1 and component 4 were two of them. All components were obtained by equivalent dipole analysis and their respective corresponding dipoles were obtained, and their positions and directions in the brain were marked. Finally, ERSP is used to resolve the independent components and dipoles associated with consciousness tasks. The ERSP graph can determine that the component 4 is related to thinking. Figure 3 finally shows the position and orientation of the five dipoles associated with the idea of ​​reading the “Left” when they are left-handed. They are located at the top of the right cerebral cortex and temporal lobe.

Two types of imaginations for the two-step experiment of 10 subjects were analyzed. The results of dipoles related to consciousness tasks are shown in Figure 4. According to the average result of 10 subjects, the number of dipoles who read “Bian” silently when writing “壹” was the highest, reaching 5, indicating that this type of thinking activity is more complex and involves more cerebral cortex. The results of other conscious tasks were: Imagining the body to read "left" (4.3), only imagining strokes (3.6), and only imagining the body (2.5). This coincides with the ranking of the average classification accuracy of the four individuals. (The classification accuracy comes from the literature [7]). Compared to single modality, multimodal conscious tasks involve more complex thinking activities and therefore more cerebral cortical involvement, which makes EEG signals more separable. The results of this paper have integrated the results of EEG signal feature extraction and classification [7], and found the internal reason for the improvement of the classification accuracy of EEG signals, and finally fully verified the feasibility and rationality of the multimodal experimental paradigm.

5 Conclusion

The equivalent dipole analysis method can inversely calculate the EEG signal to locate the spatial position of the signal source. This method is relatively inexpensive and convenient. In this paper, the spatial source location analysis of the multi-modal and single-mode experiments is performed. The number and distribution of dipoles in the two-step experiments are compared. It is found that the multi-modal sub-experiments have more couples related to thinking activities. Polar. More dipoles also mean that more cerebral cortex is activated, which also validates the intrinsic reasons for the increased classification accuracy of multimodal consciousness tasks. Therefore, the rationality and feasibility of multimodal experiments are verified from the perspective of spatial source location.

9V Switching Wall Charger

9V Switching Wall Charger

9V Switching Wall Charger


About this item

  • 9V Switching wall charger
  • 110V input voltage / 9VDC 1A/2A/3A... output voltage
  • For use with Arduino Uno, Mega and MB102 Power supply boards
  • Connector size: 5.5 x 2.1mm/5.5*2.5mm...
  • Center or Tip is positive, sleeve is negative



9v wall charger,AC Power Supply Wall Plug,Wall Adapter Power Supply,9V Power Adapter,ac 50/60hz power adapter,Wall Adapter Power Supply - 9VDC,100-240v converter switching power adapter

Shenzhen Waweis Technology Co., Ltd. , https://www.laptopsasdapter.com