On Some Possibilities of Organizing a Mobile Hardware-information

System for Polyfactorial Neuro-electrostimulation

Vladimir Kublanov, Mikhail Babich and Anton Dolganov

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

Mira 19, 620002, Yekaterinburg, Russian Federation

Keywords: Neuro-electrostimulator, Current Pulses Field, Neuroplasticity, Information System, Machine Learning.

Abstract: In paper the organizational principles of the mobile hardware-information system for polyfactorial neuro-

electrostimulation were considered. It was shown that the system can be implemented by three functionally

separate blocks, one of which ensures the formation of a spatially distributed field of current pulses, the

second is the specialized interface for the patient, and the third is the specialized interface for the doctor.

The exchange of information between the blocks is provided by a telemetric communication channel or via

the global network using mobile wearable computers (which can include a personal computer, tablet or

smartphone). A personalized patient information system can be implemented on the basis of the neuro-

electrostimulation system. In this case patient data can be placed on the server of the medical institution.

The prospects for using artificial intelligence and machine learning to control the treatment process were

discussed.

1 INTRODUCTION

Medical rehabilitation today is the relevant area of

medicine, which is associated with its great social

significance (Gunn, 2017; Khan et al., 2015; Voinea

et al., 2017).

It is known that the state of health and illness

differ in the level of the adaptive mechanisms in the

body (Pulakat et al., 2011). Increasing the adaptation

of a healthy person to constantly changing

environmental conditions is the main measure of

health and is achieved through temporal adjustment.

Adaptation of a sick person to the conditions of

existence is carried out by compensating for

impaired functions. At the same time, adaptation and

compensatory mechanisms are based on functional,

biochemical and morphological properties and

reactions of the body that are similar in nature and

are based on increasing of the functional capabilities

for existing structures and functions (Ram-Tiktin,

2011).

In order to launch these mechanisms, particular

conditions are necessary that can be initialized with

the help of biotechnical systems of restorative

medicine, in which physical fields are used for

stimulation. At the same time regulatory systems

that stimulate the processes of adaptation and

compensation are used as stimuli targets.

This article discusses some of the possibilities

for organizing such system in which polyfactorial

neuro-electrostimulation is used to control the

processes of adaptation and compensation.

2 MATERIALS AND METHODS

Stimulation of the Neck Neural Formations and

Organization of the Neuro-electrostimulation

Process Management

In general, modern biotechnical systems, which are

focused on solving the problems of medical

rehabilitation, should provide:

1. Formation in the problem area of the body by

means of the external physical field for targeted

physiological changes aimed at restoring the health

of persons with disabilities.

2. Regulation of the structure of the external

physical field and its biotropic parameters, as well as

the choice of targets for stimulation, which form the

‘targeted’ physiological changes.

572

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

On Some Possibilities of Organizing a Mobile Hardware-information System for Polyfactorial Neuro-electrostimulation.

DOI: 10.5220/0007695505720576

In Proceedings of the 12th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2019), pages 572-576

ISBN: 978-989-758-353-7

3. Measuring the patient's response to targeted

stimulation by monitoring functional changes in the

central and autonomic nervous systems, as well as

mental and behavioral functions.

With that in mind, the first task is conditionally

“tied” to the patient, the second to the doctor. To

implement the third task, one can use both

embedded and stationary systems of functional

diagnostics. Recently, video surveillance systems are

also in high demand (Klompmaker et al., 2010).

An individual multidisciplinary approach to

medical rehabilitation organization is promising.

The physician must take into account the patient's

age, gender, main and concomitant diseases, the

degree of adaptation-compensatory mechanisms

training and the biorhythmic activity of the vital

body functions. This means that during

rehabilitation, data on clinical, functional and mental

changes should be observed during the treatment

process. Using this data allows physician to organize

an information system to support the treatment

process.

According to WHO, impaired cerebral

circulation and diseases for which treatment requires

neurorehabilitation are the most common causes of

disability and mortality among the population

(WHO, 2018). As noted earlier in our papers

(Kublanov, 2008), technologies that use a spatially

distributed field of monopolar low-frequency current

pulses, which characteristics are similar to

endogenous processes in neural networks, are

promising for solving such rehabilitation problems.

In this case, neural formations of the neck are used

as targets for stimulation: segmental control centers

of vital functions (cervical sympathetic ganglia) and

pathways of suprasegmental homeostasis regulation

centers (glossopharyngeal and vagus nerves and

their branches, as well as the cervical plexus of

spinal nerves).

At the same time, during the electrical

stimulation of the cervical spinal plexus, branches of

the vagus, accessory and glossopharyngeal nerves,

the gray matter of the brainstem can be stimulated

along afferent pathways. Through the reticular

formation, the effect in this case extends to the

thalamic structures and the cerebral cortex.

Stimulation of the sympathetic trunk nodes provides

an effect on both the vascular tone of the cerebral

arteries and the vegetative nuclei of the spinal cord.

As a result of these actions, it is possible to influence

various functional processes in the brain tissues,

modulate the autonomic processes, and influence the

motor control and cognitive functions (Michailov et

al., 2013).

An analysis of the anatomical features for the

neural formations in the neck made it possible to

establish that, for neuro-electrostimulation tasks, 6

targets on the neck surface are promising, the

projections of which correspond to the location of

the upper and middle cervical ganglia of the

sympathetic nervous system (target 1), the

sympathetic trunk (target 3) ), spinal plexus (target

4), vagus nerve (target 5), accessory nerve and

branches of the glossopharyngeal nerve (target 6)

(Orlov and Nozdrachev, 2010).

The formation of these stimulation targets is

ensured by selecting the electrodes of a neuro-

electrostimulator, which allow the formation of an

adequate field of current pulses in the projection of

the corresponding nervous formations. Example of

the electrodes formation, which allows stimulation

of the aforementioned targets is presented on Figure

Figure 1: Electrodes combinations for different targets.

Living organisms are complex and consist of

many interconnected systems, its functional state is

determined by a large number of biophysical

variables (factors). The more factors that can be used

in rehabilitation, the more likely it will be to achieve

the desired therapeutic effect. In addition, each of

the methods used in rehabilitation should

complement and not duplicate the others, be

independent of them and not create discomfort due

to the inconvenience of operation. Therefore, a

promising neuro-electrostimulator should, on the

one hand, be polyfactorial, and on the other, be

compact and mobile, since the large mass-

dimensional characteristics of the neuro-

electrostimulator may be the reason for the

discomfort.

3 RESULTS

Organization of Hardware-information System

Based on the tasks to be solved, the neuro-

electrostimulator of the mobile hardware-

information system can be implemented from three

functionally separate blocks:

• the first block ensures the formation of a

monopolar rectangular current pulses field;

On Some Possibilities of Organizing a Mobile Hardware-information System for Polyfactorial Neuro-electrostimulation

573

• the second block is a specialized patient interface

and provides a solution of two tasks:

- changing the structure of current pulses the field,

setting the values of biotropic parameters of pulses

(amplitude, duration and repetition rate of a

sequence of consecutive pulses) and choosing the

stimulation target;

- collecting information about the patient, its clinical

data and functional parameters of the central and

autonomic nervous systems, as well as monitoring

data for the neurorehabilitation process;

• the third block is a specialized doctor interface and

provides:

- data input into the second block on the structure of

the current pulse field, the value of the biotropic

parameters of the pulses and the stimulation target;

- provides the second block with data for managing

the treatment process (turning on / off the first block

and parameters of the cyclogram of the stimulation

procedure), as well as comments about the patient

and the course of the treatment process.

The schematic of the mobile hardware-

information system for the polyfactorial neuro-

electrostimulation is presented on Figure 2.

Figure 2: Mobile hardware-information system schematic.

In order to ensure the mobility and compactness

of the first block, in its implementation it is

necessary to use high-level system integration

electronic radio devices and microcontrollers. It

consists of two multi-element electrodes, between

which a spatially distributed field of current pulses is

formed; a multichannel source of pulsed current, the

functions of which are realized by two multiplexers

and a controlled current source; battery; transceiver

telemetry communication channel; flash memory

that implements the functions of a persistent storage

device; microcontroller. Flash memory allows

physician to save individual patient data, as well as

data on the structure of the field of current pulses

and the characteristics of partial pulses in each

treatment procedure. It is essentially one of the main

carriers of information about the patient’s treatment

process, provides the ability to use its personalized

medicine system and save these data in the service

available for global computer networks. The

technical implementation of the first unit made it

possible to ensure its compactness and mobility: the

prototype of the product has a mass of not more than

200 g, and its overall dimensions are 90x50x18 mm.

Mobile wearable computers can be used as the

second and third units, which include a personal

computer, tablet or a smartphone. The exchange of

information between the first and second blocks is

provided by a telemetric communication channel (in

particular Bluetooth Low Energy), and between the

second and third blocks via the Internet.

With the help of wearable computers, it is

possible to organize the interaction of a doctor and a

patient remotely using telemedicine technical means

via the Internet.

The approach described above made it possible

to implement a system for neuro-electrostimulation

using mobile and compact devices.

4 DISCUSSION

Information System of the Treatment Process

Support in Neurorehabilitation

The information support system for the treatment

process should be formed using personalized data

from patients who have undergone a course of

neurorehabilitation. In general, regardless of the

pathology that is treated during medical

rehabilitation, for the implementation of such a

system it is necessary to solve the following tasks:

1. The choice of biomedical signal, which is a

reliable indicator of changes in the functioning of

the patient's state in case of particular pathology.

2. Determination of the diagnostically significant

features complexes for this biomedical signal to

assess the current functional state of the patient.

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

574

3. Prediction of the functional state for the patient

during the treatment process.

Let us consider a case study of information

support for the treatment process using the example

of developing a decision support system for a

physician in the treatment of arterial hypertension.

Analysis of the pathophysiological factors of

arterial hypertension indicates a special role of the

autonomic nervous system in their formation (da

Silva et al., 2014). It is known that blood pressure is

supported by physiologically several regulatory

mechanisms, including neuronal and humoral. The

exclusive role here belongs to the vegetative nervous

system. Therefore, its functional disorders can be

considered as the most important pathophysiological

factor in the development of arterial hypertension.

For example, in the pathogenesis of hypertension,

the role of increased activity of the sympathetic

division of the autonomic nervous system is noted,

which can lead to impaired neurogenic regulation of

the circulatory system (Cardinali, 2017).

The works of physiologists have shown that

heart rate variability (HRV) can be a reliable

indicator of changes in the autonomic nervous

system (Malik, 1996). The HRV reflects the work of

the cardiovascular system and mechanisms of

regulation of the whole organism, including the

overall activity of regulatory mechanisms, neuro-

humoral regulation of the heart, the relationship

between the sympathetic and parasympathetic

divisions of the autonomic nervous system.

To reduce the effect of situational deviations on

the results of diagnostics, it is advisable to use

functional tests (or loads) in studies. In this case,

when choosing the load that activates the reaction of

the problem function, one can get information that

most fully reflects the pathophysiological state of the

patient. In case of the cardiovascular system

disorders rehabilitation, a tilt-test study is often used

as such a load, in which the patient is transferred

from a horizontal position to vertical head up and

back to horizontal position.

The study involved two groups of subjects -

relatively healthy volunteers and patients suffering

from arterial hypertension of 2-3 degrees (Dolganov

et al., 2017). It is known that the HRV signal can be

described by a variety of features in the time and

frequency domains, as well as using nonlinear

dynamics methods. After reduction of the HRV

features numbers, it is possible to form a sets of

diagnostically significant features using the methods

of machine learning in solving the problem of binary

classification. The best solution was obtained using

a search strategy based on evolutionary

programming. In that case, quadratic discriminant

analysis was used as a classification method. The

obtained sets of diagnostically significant features

consisted of 12–15 features of HRV (out of 192

features total) recorded in each of the functional

states when performing tilt-test studies:

If one apply quadratic discriminant analysis to

predict the class of the subject (“healthy” or “patient

suffering from arterial hypertension”), then the result

of this operation will be the intended class of the

subject and the probability of belonging to this class.

As there are only two classes of subjects, the use of

quadratic discriminant analysis allows to reduce the

multidimensional space of diagnostically significant

parameters to the one-dimensional space of the

decision rule metrics. When training a classifier, a

hyperplane is formed that separates the two classes

of subjects. In the decision rule space, this

hyperplane defines the origin. In our case, the

positive values of the metric in the space of decision

rules correspond to the class of subjects “healthy”,

the negative values of the metric correspond to the

class of subjects “patients suffering from arterial

hypertension”. In our case the accuracy of the

proposed approach can reach 98% (Dolganov and

Kublanov, 2018).

Further, changes in the metrics in the space of

decision rules and the dynamics of changes in blood

pressure during the treatment process were analyzed.

It was shown that the dynamics of changes in the

metrics obtained on the basis of sets of

diagnostically significant features had a rather high

degree of consistency with changes in blood

pressure. This indicates that the obtained sets of

diagnostically significant features can be used as

additional markers of change during the

rehabilitation process.

5 CONCLUSION

The organization principles of the hardware-

information system proposed in this work allowed to

create a mobile and compact system for

neurorehabilitation. It was shown that the

application of artificial intelligence and machine

learning makes it possible to ensure the management

of the rehabilitation process taking into account the

requirements of personalized medicine.

At the moment, the neuro-electrorehabilitation

polyfactorial system has been clinically tested in the

treatment of depressive anxiety disorders, children

with attention deficit disorder, rehabilitation of

patients after traumatic brain injuries (Kublanov,

On Some Possibilities of Organizing a Mobile Hardware-information System for Polyfactorial Neuro-electrostimulation

575

Retyunskii, et al., 2016; Kublanov, Petrenko, et al.,

2016; Kublanov et al., 2017; Petrenko et al., 2015,

2017). Clinical studies have shown that, compared

with the state-of-art, a higher efficiency of treatment

is achieved by involving the regulatory process in

addition to the autonomic nervous system, brain

structures responsible for cognitive, motor, visual,

auditory, vestibular and other brain functions.

Finally, the rehabilitation process is becoming more

adapted to the problems of a particular patient.

ACKNOWLEDGEMENTS

The work was supported by Act 211 Government of

the Russian Federation, contract № 02.A03.21.0006.

REFERENCES

Cardinali DP (2017) Autonomic Nervous System: Basic

and Clinical Aspects. Springer.

da Silva NT, Giacon TR, Vanderlei FM, et al. (2014)

Hypertension and Autonomic Control. American

Journal of Medical Sciences and Medicine 2(2): 48–

53.

Dolganov A and Kublanov V (2018) Towards a Decision

Support System for Disorders of the Cardiovascular

System - Diagnosing and Evaluation of the Treatment

Efficiency. In: Proceedings of the 11th International

Joint Conference on Biomedical Engineering Systems

and Technologies - Volume 5: AI4Health (BIOSTEC

2018), 22 May 2018, pp. 727–733. DOI:

10.5220/0006753407270733.

Dolganov AY, Kublanov VS, Belo D, et al. (2017)

Comparison of Machine Learning Methods for the

Arterial Hypertension Diagnostics. Applied Bionics

and Biomechanics 2017(5985479). DOI:

10.1155/2017/5985479.

Gunn AE (2017) Cancer rehabilitation. Phys Med Rehabil

Clin N Am 28(1): 1–17.

Khan F, Amatya B, Gosney J, et al. (2015) Medical

rehabilitation in natural disasters: a review. Archives of

physical medicine and rehabilitation 96(9): 1709–

1727.

Klompmaker F, Busch C, Nebe K, et al. (2010) Designing

a Telemedical System for Cardiac Exercise

Rehabilitation. In: Biomedical Engineering Systems

and Technologies, 20 January 2010, pp. 111–122.

Communications in Computer and Information

Science. Springer, Berlin, Heidelberg. DOI:

10.1007/978-3-642-18472-7_9.

Kublanov V, Dolganov A, Shalyagin M, et al. (2017)

Efficiency of dynamic correction of sympathetic

nervous system activity in patients with panic

disorder. In: Proceedings - 2017 International Multi-

Conference on Engineering, Computer and

Information Sciences, SIBIRCON 2017, Novosibirsk

Akademgorodok, Russia, 2017, pp. 571–574. DOI:

10.1109/SIBIRCON.2017.8109956.

Kublanov VS (2008) A hardware-software system for

diagnosis and correction of autonomic dysfunctions.

Biomedical Engineering 42(4): 206–212. DOI:

10.1007/s10527-008-9047-7.

Kublanov VS, Retyunskii KY and Petrenko TS (2016) A

New Method for the Treatment of Korsakoff’s

(amnestic) Psychosis: Neurostimulation Correction of

the Sympathetic Nervous System. Neuroscience and

Behavioral Physiology 46(7): 748–753.

Kublanov VS, Petrenko TS, Petrenko AA, et al. (2016)

The recovery of cognitive functions for patients with

the organic amnestic syndrome by means of the non-

invasive adaptive neuro-electrostimulation device. In:

Cognitive Sciences, Genomics and Bioinformatics

(CSGB), 2016, pp. 1–3. IEEE.

Malik M (1996) Heart rate variability: Standards of

measurement, physiological interpretation, and clinical

use. Circulation 93(5): 1043–1065.

Michailov SS, Chukbar AV and Tsybul’kin AG (2013)

Human Anatomy [in Russian]. M.: GEOTAR-Media.

Orlov RS and Nozdrachev AD (2010) Normal Physiology.

A. Textbook [in Russian]. M.: GEOTAR-Media.

Petrenko T, Kublanov V and Retyunskiy K (2017) The

role of neuroplasticity in the treatment of cognitive

impairments by means multifactor neuro-

electrostimulation of the segmental level of the

autonomic nervous system. European Psychiatry 41:

S770.

Petrenko TS, Kublanov VS and Retiunskiy KY (2015)

Dynamic Correction of the Activity Sympathetic

Nervous System (Dcasns) to Restore Cognitive

Functions. European Psychiatry 30: 843.

Pulakat L, DeMarco VG, Ardhanari S, et al. (2011)

Adaptive mechanisms to compensate for

overnutrition-induced cardiovascular abnormalities.

American Journal of Physiology-Regulatory,

Integrative and Comparative Physiology 301(4):

R885–R895.

Ram-Tiktin E (2011) A decent minimum for everyone as a

sufficiency of basic human functional capabilities. The

American Journal of Bioethics 11(7): 24–25.

Voinea G-D, Butnariu S and Mogan G (2017)

Measurement and geometric modelling of human

spine posture for medical rehabilitation purposes using

a wearable monitoring system based on inertial

sensors. Sensors 17(1): 3.

WHO (2018) World Health Statistics 2018: Monitoring

Health for the SDGs Sustainable Development Goals.

World Health Organization.

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

576