Processing the Results of Electroencephalography for Patients

Suffering from Depression after Neuro-electrostimulation Course:

Case Study

Vladimir Kublanov and Anton Dolganov

Research Medical and Biological Engineering Centre of High Technologies, Ural Federal University,

Mira 19, 620002, Yekaterinburg, Russian Federation

Keywords: Neuro-electrostimulation, Neurovisualisation, Electroencephalography, Case Study.

Abstract: The article presented the results of electroencephalography (EEG) signal processing in a case study of

neuro-electrostimulation application for patients suffering from depression. Neuro-electrostimulation was

performed by the SYMPATHOCOR-01 device in two modes - multichannel and single-channel stimulation.

The analysis of changes in the EEG activity maps during neuro-electrostimulation course was carried out.

The common conclusion for all patients is an increase in the homogeneity for the distribution of spectral

power density for EEG signals. A quantitative method for estimating the level of the brain zones activation

was proposed. For patients from the multichannel stimulation group, an increase in the activation level was

observed. It was noted that for patients from the single-channel stimulation group there were zones in which

a significant decrease in the level of activation was observed.

1 INTRODUCTION

The depressive disorder is rapidly spreading of

among the able-bodied population in the developed

countries. This determines the relevance for search

of an effective ways of treatment and rehabilitation

approaches to mitigate this disease. Unfortunetley

the depressive disorder can occur at any age,

resulting in sharp limitation of a person’s adaptation

to constantly changing environmental conditions

(Culpepper et al., 2015).

The problem of depression is primarily

determined by the lack of knowledge about the

pathophysiological mechanisms of this disease.

Recently, in developed countries, number work has

emerged on the use of neuroimaging techniques to

solve this problem. For example, when analyzing

brain activity using fMRI, it was found that,

compared with practically healthy patients, patients

with depression experience different patterns of

impairment of the cerebral cortex during the patient's

life (Schmaal et al., 2017).

The most common approach to normalize and

strengthen the physiological activity of brain tissue

is neuroprotective therapy (Kupfer et al., 2012). This

therapy is mainly implies application of drugs. The

use of drugs does not always exclude side effects.

To a lesser extent, this refers to physiotherapeutic

methods, especially methods that use low-intensity

electric current for stimulation (Cook et al., 2016).

Promising for solving the problems of

neurorehabilitation are technologies in which multi-

electrode stimulation systems are used. This

direction is actively developed in the works of

research teams headed by Y. Danilov and V.S.

Kublanov. There, for neurorehabilitation, a spatially

distributed field of monopolar low-frequency current

pulses is used, the characteristics of which are

similar to endogenous processes in neural networks.

In the known technical implementations of such

devices, either branches of the cranial nerves (PoNS

device (Danilov et al., 2015)) or cervical ganglia of

the sympathetic nervous system (SYMPATHOCOR-

01 device (Kublanov et al., 2017)) are used as

targets for stimulation.

The SYMPATHOCOR -01 device implements

the technology of multichannel neuro-

electrostimulation. This technology allows physician

to manage the activities of conductive formations

and performs the process of neuromodulation.

Medical use of the device SYMPATHOCOR -01 is

implemented as a method of DCASNS - a dynamic

Kublanov, V. and Dolganov, A.

Processing the Results of Electroencephalography for Patients Suffering from Depression after Neuro-electrostimulation Course: Case Study.

DOI: 10.5220/0007695605770582

In Proceedings of the 12th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2019), pages 577-582

ISBN: 978-989-758-353-7

577

correction of the activity of the sympathetic nervous

system. The DCASNS method provides correction

of autonomic balance, determined by the ratio

between the activity of the parasympathetic and

sympathetic divisions of the autonomic nervous

system (Petrenko et al., 2015).

The purpose of this work is to process the results

of electroencephalography (EEG) in assessing the

effectiveness of electrical neurostimulation device

SYMPATHOCOR-01 for patients suffering from

depression.

2 MATERIALS AND METHODS

2.1 Case Study Group

The case study involved 6 subjects diagnosed with a

depression. The study was approved by the ethical

committee of the State Scientific-Research Institute

of Physiology & Basic Medicine (Protocol No. 13 of

November 16, 2017).

The subjects were divided into two groups. In the

first group, the neuro-electrostimulation device

SYMPATHOCOR -01 was used in the multichannel

stimulation mode. The upper and middle ganglia of

the sympathetic nervous system were selected as

targets.

In the second group, a neuro-electrostimulator

was used in the single-channel stimulation mode;

descending nerve fibers to the stellate ganglion were

selected as targets.

Table 1 summarize data on the case study group.

Table 1: Case study group data.

Case history Mode

794 multichannel

798 single-channel

864 single-channel

862 multichannel

863 single-channel

865 multichannel

2.2 Study Description

The study used a 126-channel EEG recording

system. The sampling rate was 1000 Hz.

Registration of EEG took place simultaneously with

fMRI studies (Sokolov et al., 2017). During the

study, the subjects lay at rest (Rest State). In the

study the electrode location system 10–5, which is a

more dense version of the system 10–20 (Oostenveld

and Praamstra, 2001). An example of the location of

the electrodes in three-dimensional space is

presented in Figure 1.

Figure 1: 3-D representation of the electrode location.

In the present work, the features of EEG signals

recorded during primary studies (prior to the neuro-

electrostimulation course) and after stimulation

procedures.

The time of registration of EEG signals was

about 10 minutes. The first and last minute of the

EEG signals were excluded from the analysis due to

the presence of motion artifacts.

2.3 Data Analysis Methods

To process the EEG signals, is an open source

software toolkit for MATLAB – EEGLAB – was

used. This toolkit is a practical implementation of

the functions and graphic interface used in the

processing and visualization of electrophysiological

signals (Delorme and Makeig, 2004).

After importing the “raw” EEG signals, spectral

powers were evaluated. In this wour, we investigated

four spectral ranges:

• Delta rhythm - from 3 to 4 Hz;

• Theta rhythm - from 4 to 7 Hz;

• Alpha rhythm - from 8 to 15 Hz;

• Beta rhythm - from 16 to 31 Hz;

It should be noted that in the delta rhythm

frequencies below 3 Hz were not analyzed, due to

the presence of noise.

Evaluation of the spectral components in the

EEGLAB is carried out using the function

pop_spectopo(EEG, TIME), where the EEG variable

is a matrix of EEG signals from time for each of 126

channels. The TIME variable contains information

about the beginning and end of the time interval,

within which the spectral component is assessed. In

the present work, the evaluation of the spectral

NNSNT 2019 - Special Session on Non-invaisive Neuro-stimulation in Neurorehabilitation Tasks

578

components was carried out in 10-second windows,

with an overlap of 5 seconds. In total, spectral

indices were obtained for each subject in 80 epochs.

The result of using the pop_spectopo function is the

Spectral Power Density, estimated by the Welch

method, for each channel for all frequencies. The

spectral power estimates are presented on a

logarithmic scale (Welch, 1967).

The power estimation in the studied frequency

ranges was carried out by summation over the

corresponding frequencies. Thus, for each channel,

for each subject, four spectrograms were obtained,

describing the change in spectral power densities by

epochs.

For evaluation of the localization in different

activation zones, it was proposed to group 126 EEG

channels into 11 zones - frontal (left and right - F_L,

F_R), temporal (left and right - T_L, T_R), central

(left and right - C_L, C_R), occipital (left and right -

P_L, P_R), parietal (left and right - O_L, O_R).

Separately zone Z was considered, which took into

account the channels located on the central axis. The

division of the EEG channels into zones is presented

in Table 2.

Table 2: Distribution of channels by zones.

Zone EEG channel

Z 'FPZ' 'FZ' 'CPZ' 'CZ' 'IZ' 'PZ' 'OZ' 'POZ'

F_R

'AF3' 'AF7' 'AFF1H' 'AFF5H' 'AFP1' 'F1' 'F3' 'F5' 'F7'

'F9' 'FFC1H' 'FFC3H' 'FFC5H' 'FP1'

F_L

'AF4' 'AF8' 'AFF2H' 'AFF6H' 'AFP2' 'F10' 'F2' 'F4' 'F6'

'F8' 'FFC2H' 'FFC4H' 'FFC6H' 'FP2'

T_R

'FFT7H' 'FFT9H' 'FT7' 'FT9' 'FTT7H' 'T7' 'TP7' 'TP9'

'TPP7H' 'TPP9H' 'TTP7H'

T_L

'FFT10H' 'FFT8H' 'FT10' 'FT8' 'FTT8H' 'T8' 'TP10'

'TP8' 'TPP10H' 'TPP8H' 'TTP8H'

C_R

'FC1' 'FC3' 'FC5' 'FCC1H' 'FCC3H' 'FCC5H' 'C1' 'C3'

'C5' 'CCP1H' 'CCP3H' 'CCP5H' 'CP1' 'CP3' 'CP5'

C_L

'FC2' 'FC4' 'FC6' 'FCC2H' 'FCC4H' 'FCC6H' 'C2' 'C4'

'C6' 'CCP2H' 'CCP4H' 'CCP6H' 'CP2' 'CP4' 'CP6'

P_R

'CPP1H' 'CPP3H' 'CPP5H' 'P1' 'P3' 'P5' 'P7' 'P9'

'PPO1H' 'PPO5H' 'PPO9H'

P_L

'CPP2H' 'CPP4H' 'CPP6H' 'P10' 'P2' 'P4' 'P6' 'P8'

'PPO10H' 'PPO2H' 'PPO6H'

O_R 'O1' 'O9' 'OI1H' 'PO3' 'PO7' 'PO9' 'POO1' 'POO9H'

O_L 'O10' 'O2' 'OI2H' 'PO10' 'PO4' 'PO8' 'POO10H' 'POO2'

The estimation of changes in the level of

activation in the zone is proposed. Initial data were

used to estimate the average level of the spectrogram

value in the zone over the entire time interval for all

channels (activation threshold). Then, for each time

epoch, the proportion of channels that exceed the

activation threshold was estimated. The same

activation threshold was used in the analysis of

signals recorded after neuroelectrostimulation

procedures.

3 RESULTS

3.1 Power Spectral Density

Visualization

Figures 2-3 show the visualization of the spectral

power density of EEG signals before and after

neuro-electrostimulation for a patient CH794.After

the procedures of neuro-electrostimulation for

patient CH794, normalization of activity is observed

— after the stimulation procedures, the distribution

of spectral power is more uniform. At the same time,

it is worth noting a statistically significant increase

in the power level in the frontal and occipital zones

by more than 10 dB, for delta and beta rhythms, and

by 7 dB for theta and alpha rhythms.

Figure 2: Visualization of power density for patient

CH794 before stimulation.

Figure 3: Visualization of power density for patient

CH794 after stimulation.

Data for patient CH862 indicate a slight change

in the activity of spectral powers. It is worth noting

the decrease in power of EEG signals for all four

rhythms by 10 dB in the central zone of the brain

(electrodes Cz, C1, C2, FCZ, CPZ).

The most significant changes for patient CH865

are observed for the beta rhythm in the right

occipital region (about 7 dB). It is worth noting the

Processing the Results of Electroencephalography for Patients Suffering from Depression after Neuro-electrostimulation Course: Case Study

579

decrease in EEG activity in the central zone for the

delta, theta and alpha rhythms (8 dB), the increase in

the activity of the theta rhythm in the frontal zone (4

dB), as well as the increase in the activity of the

delta rhythm in the right temporal zone (6 dB).

Figures 4-5 show the visualization of the power

spectral density of EEG signals before and after

neuro-electrostimulation for patient CH798.Data for

patient CH798 indicates that the changes are

asymmetrical: for the left side (electrodes F5, F3,

FC5, FC3, C5, C3, CP5, Cp3, P5, P3), an increase in

the EEG power for theta and alpha rhythms by 8 dB

is observed, for delta and beta, the spectral power

increase by 4 dB. At the same time, for the right side

(electrodes F4, F2, FC4, FC2, C4, C2, CP4, CP2,

P4, P2), in delta, theta and alpha rhythms, a decrease

in EEG power is observed by 8 dB and by 6 dB in

the beta rhythm.

Analysis of the results for patient CH863,

indicates a local change in the power of the EEG.

For the delta, theta, and alpha rhythms, an increase

in the spectral power is observed by 7 dB in the

region of the T8, FC6, FC4, FC2, and FC1

electrodes. For the beta rhythm, there is a decrease

in EEG power by 8 dB in the area of the electrodes

FCZ, Cz and CPZ. It should be noted that after the

neuro-electrostimulation procedures, the activity of

the EEG rhythms became more homogeneous - local

heterogeneities in the center disappeared.

For patient CH864, it is worth noting the

decrease in EEG activity for all rhythms in the area

of the Cz electrode. At the same time, there is a

slight increase in the power of the left half of the

brain for the delta and theta rhythms by 4 dB.

Figure 4: Visualization of power density for patient

CH798 before stimulation.

Summarizing the analysis of visualization of the

spectral power density of EEG signals, it can be

concluded that the most significant changes are

observed for patients CH794 and CH798. At the

same time, changes for patient CH794 (from the

group of multichannel stimulation) were more

pronounced, while for patient CH798 (from the

group of single-channel stimulation) the changes

were asymmetric. In general, for all patients, an

increase in spectral power uniformity is observed.

Figure 5: Visualization of power density for patient

CH798 after stimulation.

3.2 Evaluation of the Activation Level

Figures 6-8 show the bar graphs of the level of

activation of the EEG channels for 11 zones, before

(blue) and after stimulation (yellow) for the

multichannel stimulation group. Activation of 100%

is obtained when spectral power of all channels in

zone is higher than threshold.

Figure 6: CH794; estimation of activation level, %.

Figure 7: CH862; estimation of activation level, %.

According to Figure 6, it can be concluded that

the activation level increases for all EEG rhythms

NNSNT 2019 - Special Session on Non-invaisive Neuro-stimulation in Neurorehabilitation Tasks

580

for patient CH794. The most significant changes are

in all zones for the delta rhythm, as well as the

occipital zones (P and O) for theta, alpha and beta

rhythms. It should be noted that the level of

activation varies slightly during the study.

Figure 8: CH865; estimation of activation level, %.

According to Figure 7 for the patient CH862, the

level of activation basically remained the same. The

increase in activation in the temporal zones (T_R

and T_L) for the alpha rhythm is statistically

significant. At the same time, it is worth noting the

increase in the degree of scatter of the level of

activation during the study after the neuro-

electrostimulation procedures.

Based on the data of Figure 8, it can be

concluded that for patient CH865 is the most

significant increase in the level of activation in the

occipital zones (O_R and O_L) for alpha and beta

rhythms. At the same time, it is worth noting the

increase in the activation level for theta, alpha and

beta rhythms for the right frontal zone (F_R).

Figures 9-11 show the bar graphs of the level of

activation of the EEG channels for 11 zones, before

(blue) and after stimulation (yellow) for a single-

channel stimulation group.

Figure 9: CH798; estimation of activation level, %.

The data on Figure 9 indicate the diversified

nature of changes in the level of activation in patient

CH798. For all rhythms, an increase in the level of

activation is observed for the right frontal zone, as

well as a decrease in the level of activation for the

occipital zone and the left central zone. For theta,

alpha and beta rhythms it is worth noting a

significant increase in the level of activation in the

temporal zones, as well as in the left parietal zone.

Figure 10: CH863; estimation of activation level, %.

Figure 11: CH864; estimation of activation level, %.

Based on the graphs shown in Figure 10, it can

be concluded that for patient CH863, there is a

decrease in the level of activation for theta and beta

rhythms. For the alpha rhythm, an increase in the

activation level is observed for the left frontal, right

temporal and left parietal.

The bar graphs shown in Figure 11 indicate that

patient CH864 experienced an increase in the

activation level for most zones in the delta and theta

rhythm. For the alpha rhythm, a decrease in the level

of activation in the parietal and occipital zones is

observed.

Summing up the evaluation of the brain areas

activation level, it can be concluded that an increase

in the level of activation is observed for patients

from the group of multichannel stimulation, this is

especially pronounced for patient CH794. It is noted

that for patients from the single-channel stimulation

group there were zones in which a significant

decrease in the level of activation is observed.

4 CONCLUSIONS

The article has presented the results of EEG signal

processing in a case study of neuro-

Processing the Results of Electroencephalography for Patients Suffering from Depression after Neuro-electrostimulation Course: Case Study

581

electrostimulation therapy application for patients

suffering from depression. The neuro-

electrostimulation procedures were tested in two

modes – multichannel and single-channel

stimulation.

Summarizing the analysis of visualization of the

spectral power density of EEG, it can be concluded

that the most significant changes have been observed

for patients CH794 and CH798. At the same time,

changes for patient CH794 (from the group of

multichannel stimulation) had been more

pronounced, while for patient CH798 (from the

group of single-channel stimulation), the changes

had been asymmetric. In general, for all patients, an

increase in EEG power uniformity has been

observed.

A quantitative method for estimating the level of

activation of brain zones has been proposed. Initial

data (prior to the neuro-electrostimulation

procedures application) were used to estimate the

average level of the EEG spectrogram value in the

zone over the entire time interval for all channels

(activation threshold). Then, for each time epoch,

the proportion of channels that exceed the activation

threshold was estimated. The same activation

threshold was used in the analysis of signals

recorded after neuro-electrostimulation procedures.

Considering the evaluation of the brain areas

activation level, it could be concluded that an

increase in the level of activation was observed for

patients from the group of multichannel stimulation,

this is especially pronounced for patient CH794. It

was noted that for patients from the single-channel

stimulation group there were zones in which a

significant decrease in the level of activation was

observed.

ACKNOWLEDGEMENTS

The EEG signals data acquisition within the study

(Chapter 2) was supported by the Act 211 of the

Government of the Russian Federation (contract no.

02.A03.21.0006). The EEG signals data processing

(Chapter 3) was funded by RFBR (project no, 18-29-

02052).

REFERENCES

Cook, I.A., Abrams, M., and Leuchter, A.F., 2016.

Trigeminal nerve stimulation for comorbid

posttraumatic stress disorder and major depressive

disorder. Neuromodulation: Technology at the Neural

Interface, 19 (3), 299–305.

Culpepper, L., Muskin, P.R., and Stahl, S.M., 2015. Major

depressive disorder: understanding the significance of

residual symptoms and balancing efficacy with

tolerability. The American journal of medicine, 128

(9), S1–S15.

Danilov, Y., Kaczmarek, K., Skinner, K., and Tyler, M.,

2015. Cranial Nerve Noninvasive Neuromodulation:

New Approach to Neurorehabilitation. In: F.H.

Kobeissy, ed. Brain Neurotrauma: Molecular,

Neuropsychological, and Rehabilitation Aspects. Boca

Raton (FL): CRC Press/Taylor & Francis.

Delorme, A. and Makeig, S., 2004. EEGLAB: an open

source toolbox for analysis of single-trial EEG

dynamics including independent component analysis.

Journal of neuroscience methods, 134 (1), 9–21.

Kublanov, V., Babich, M., and Dolganov, A., 2017.

Principles of Organization and Control of the New

Implementation of the “SYMPATHOCOR-01” Neuro-

electrostimulation Device. Presented at the Special

Session on Neuro-electrostimulation in

Neurorehabilitation Tasks, 276–282.

Kupfer, D.J., Frank, E., and Phillips, M.L., 2012. Major

depressive disorder: new clinical, neurobiological, and

treatment perspectives. The Lancet, 379 (9820), 1045–

1055.

Oostenveld, R. and Praamstra, P., 2001. The five percent

electrode system for high-resolution EEG and ERP

measurements. Clinical neurophysiology, 112 (4),

713–719.

Petrenko, T.S., Kublanov, V.S., and Retiunskiy, K.Y.,

2015. Dynamic Correction of the Activity Sympathetic

Nervous System (Dcasns) to Restore Cognitive

Functions. European Psychiatry, 30, 843.

Schmaal, L., Hibar, D.P., Sämann, P.G., Hall, G.B.,

Baune, B.T., Jahanshad, N., Cheung, J.W., van Erp,

T.G.M., Bos, D., and Ikram, M.A., 2017. Cortical

abnormalities in adults and adolescents with major

depression based on brain scans from 20 cohorts

worldwide in the ENIGMA Major Depressive

Disorder Working Group. Molecular psychiatry, 22

(6), 900–909.

Sokolov, A.V., Vorobyev, S.V., Efimtcev, A.Y., Dekan,

V.S., Trufanov, G.E., Lobzin, V.Y., and Fokin, V.A.,

2017. fMRI and Voxel-based Morphometry in

Detection of Early Stages of Alzheimer’s Disease: In:

Proceedings of the 10th International Joint

Conference on Biomedical Engineering Systems and

Technologies. Presented at the 4th International

Conference on Bioimaging, Porto, Portugal:

SCITEPRESS - Science and Technology Publications,

67–71.

Welch, P., 1967. The use of fast Fourier transform for the

estimation of power spectra: a method based on time

averaging over short, modified periodograms. IEEE

Transactions on audio and electroacoustics, 15 (2),

70–73.

NNSNT 2019 - Special Session on Non-invaisive Neuro-stimulation in Neurorehabilitation Tasks

582