Научная статья на тему 'Нейросети мозга и их моделирование с помощью нейроэмуляторов'

Нейросети мозга и их моделирование с помощью нейроэмуляторов Текст научной статьи по специальности «Медицинские технологии»

CC BY
240
49
i Надоели баннеры? Вы всегда можете отключить рекламу.
Ключевые слова
хаос / нейросети / эффект Еськова-Зинченко / реверберации / chaos / neural networks / the effect of Eskov-Zinchenko / reverberation.

Аннотация научной статьи по медицинским технологиям, автор научной работы — А. Н. Инюшкин, М. А. Филатов, С. В. Григорьева, И. Б. Булатов

Традиционно динамика биоэлектрической активности мозга, как результат работы нейросетей мозга, описывалась в рамках статистических методов. Однако, последние 25-30 лет начинает реализовываться гипотеза Н.А. Бернштейна о «повторении без повторений», что нашло доказательство в виде эффекта Еськова-Зинченко. Целью настоящих исследований является доказательство неустойчивости в работе нейросетей мозга по всем параметрам электроэнцефалограмм. В рамках этого эффекта доказывается непрерывный хаос и реверберации в работе нейросетей мозга. Эти два базовых свойства переносятся на работу нейроэмуляторов. Показано, что задание хаотических начальных значений весов wi0 диагностических признаков xi и многочисленные реверберации (повторение настройки нейросети) обеспечивают системный синтез, т.е. моделируют эвристическую деятельность мозга человека в условиях недостатка информации.

i Надоели баннеры? Вы всегда можете отключить рекламу.
iНе можете найти то, что вам нужно? Попробуйте сервис подбора литературы.
i Надоели баннеры? Вы всегда можете отключить рекламу.

NEURAL BRAIN NETWORKS AND THEIR MODELING THROUGH NEURAL EMULATORS

Traditionally, the dynamics of the bioelectrical activity of the brain as a result of the operation of the neural networks of the brain, have been described within the framework of statistical methods. However, during 25-30 years the hypothesis of N.А. Bernstein on "repetition without repetition" began to be realized. The proof of this hypothesis is the effect of Eskov-Zinchenko. The purpose of this investigation is to prove instability in the operation of the neural networks of the brain in all parameters of electroencephalograms. As part of this effect, continuous chaos and reverberation in the work of the neural networks of the brain is proved. These two basic properties are transferred to the work of neural emulators. It is shown that the assignment of chaotic initial values of the weights wi0 of diagnostic signs xi and numerous reverberations (repetition of the neural network setting) provide system synthesis, i.e. simulate the heuristic activity of the human brain in conditions of lack of information.

Текст научной работы на тему «Нейросети мозга и их моделирование с помощью нейроэмуляторов»

JOURNAL OF NEW MEDICAL TECHNOLOGIES - 2018 - V. 25, № 4 - P. 304-314

УДК: 611.018.834

НЕЙРОСЕТИ МОЗГА И ИХ МОДЕЛИРОВАНИЕ С ПОМОЩЬЮ НЕЙРОЭМУЛЯТОРОВ А.Н. ИНЮШКИН*, М.А. ФИЛАТОВ", С.В. ГРИГОРЬЕВА**, И.Б. БУЛАТОВ**

*ФГБОУВПО «Самарский национальный исследовательский университет им. акад. С.П. Королева»,

Московское ш., д. 34, Самара, 443086, Россия **БУВО «Сургутский государственный университет», ул. Ленина, д. 1, Сургут, 628400, Россия, e-mail: firing.squad@mail.ru

Аннотация. Традиционно динамика биоэлектрической активности мозга, как результат работы нейросетей мозга, описывалась в рамках статистических методов. Однако, последние 25-30 лет начинает реализовываться гипотеза Н.А. Бернштейна о «повторении без повторений», что нашло доказательство в виде эффекта Еськова-Зинченко. Целью настоящих исследований является доказательство неустойчивости в работе нейросетей мозга по всем параметрам электроэнцефалограмм. В рамках этого эффекта доказывается непрерывный хаос и реверберации в работе нейросетей мозга. Эти два базовых свойства переносятся на работу нейроэмуляторов. Показано, что задание хаотических начальных значений весов wi0 диагностических признаков xi и многочисленные реверберации (повторение настройки нейросети) обеспечивают системный синтез, т.е. моделируют эвристическую деятельность мозга человека в условиях недостатка информации.

Ключевые слова: хаос, нейросети, эффект Еськова-Зинченко, реверберации.

Введение. Длительное время все науки о мозге использовали при обработке базовых параметров нейросетей (активность отдельных нейронов или электроэнцефалограмм - ЭЭГ) традиционные статистические методы. При этом рассчитывались статистические функции распределения для параметров Xi ЭЭГ, спектральные плотности сигналов - СПС (для ЭЭГ, например), автокорреляции A(t) и т.д. Все эти характеристики при этом считались повторяемыми и самовоспроизводимыми в наблюдениях [2-8].

Однако, в 1947 г. Н.А. Бернштейн выдвинул гипотезу о «повторении без повторений» в биомеханике, которая сейчас нашла подтверждение в эффекте Еськова-Зинченко (ЭЕЗ). Этот ЭЕЗ доказывает отсутствие статистической устойчивости для подряд получаемых выборок треморо-грамм (ТМГ) у одного испытуемого в неизменном гомеостазе. Если выборки ТМГ статистически не повторяются, то можно было бы предположить, что и работа центрального регулятора любых движений (нейросетей мозга - НСМ человека) не может быть произвольно и статистически повторена. Проверке этого утверждения и посвящается настоящая работа [1-7].

Цель исследования - доказательство статистической неустойчивости для подряд получаемых выборок ЭЭГ у одного испытуемого (в неизменном гомеостазе) и использование та-

кого хаоса в работе нейроэмуляторов (нейро-ЭВМ или НЭВМ).

Объекты и методы исследования. Согласно Хельсинской декларации производилась регистрация ЭЭГ у 12-ти человек (в одном и том же отведении). При этом для каждого испытуемого регистрация ЭЭГ производилась за интервал 5 сек. по 15 раз (с одной и той же области мозга), а полученные аналоговые сигналы x(t) квантовались с периодом квантования т=10 мсек и полученные дискретные значения ЭЭГ (в виде дискретных файлов, в цифре) записывались как выборки таких сигналов (ЭЭГ) [7-14].

Таким образом, для каждого испытуемого было получено 15 выборок ЭЭГ, в каждой выборке находилось по 500 точек (период регистрации ЭЭГ Т=5 сек). Эти 15 выборок ЭЭГ обрабатывались статистически путем сравнения и построения матриц таких парных сравнений выборок ЭЭГ (в виде Xl(t)). Для каждого испытуемого строилась своя матрица парных сравнений выборок (по критерию Вилкоксона), которая содержала 105 независимых (разных) пар сравнения ЭЭГ [7,20-28].

В таких матрицах парных сравнений выборок ЭЭГ находились пары выборок, для которых критерий Вилкоксона Р>0,05. В этом случае эти две выборки (эту пару) можно было отнести к одной (общей) генеральной совокупности. Определялись числа k для каждой такой мат-

10иККЛЬ ОБ ОТШ МЕБТСЛЬ ТЕСЫК0ШЫЕ8 - 2018 - V. 25, № 4 - Р. 304-314

рицы, т.е. находилось число пар к, для которых Р>0,05. Одновременно рассчитывалось для этих же выборок ЭЭГ по 15 выборок СПС и по 15 выборок А(Ь) и для каждого испытуемого [1722,28-32]. В этом случае, еще строились (отдельно) матрицы для СПС и матрицы для А(Ь).

Таким образом, было получено 12 матриц для ЭЭГ, 12 матриц для СПС и 12 матриц для А(Ь) для всех 12-ти испытуемых. В этих матрицах находились кг (для ЭЭГ), к2 (для СПС) и кз (для А(ф. Производилось сравнение этих кг, к2, кз для каждого из 12-ти испытуемых.

Результаты и их обсуждение. Сразу отметим, что все 12 полученных значений кг для ЭЭГ имели разные значения (иногда кг совпадали для отдельных испытуемых), но все эти кг не превышали кг<40% от всех 105 пар (разных) сравнения выборок ЭЭГ. Для примера мы представляем характерную матрицу (табл. 1), где число кг=37. Это довольно большое число (обычно кг^25), но это число менее 40%.

Таблица г

Матрица парного сравнения ЭЭГ одного и того же здорового человека (число повторов N=15), использовался критерий Вилкоксона (критическое значение р<0,05, число совпадений к/=37)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1 0.00 0.00 0.00 0.00 0.00 0.19 0.00 0.00 0.00 0.01 0.00 0.01 0.00 0.04

2 0.00 0.99 0.00 0.06 0.93 0.02 0.25 0.33 0.57 0.00 0.03 0.04 0.00 0.00

3 0.00 0.99 0.00 0.09 0.75 0.03 0.21 0.50 0.95 0.00 0.10 0.10 0.00 0.00

4 0.00 0.00 0.00 0.00 0.00 0.00 0.02 0.00 0.03 0.00 0.51 0.00 0.15 0.00

5 0.00 0.06 0.09 0.00 0.28 0.29 0.00 0.88 0.11 0.00 0.00 0.65 0.00 0.00

6 0.00 0.93 0.75 0.00 0.28 0.11 0.07 0.57 0.39 0.00 0.09 0.40 0.00 0.00

7 0.19 0.02 0.03 0.00 0.29 0.11 0.00 0.09 0.10 0.00 0.00 0.58 0.00 0.01

8 0.00 0.25 0.21 0.02 0.00 0.07 0.00 0.05 0.71 0.00 0.43 0.07 0.00 0.00

9 0.00 0.33 0.50 0.00 0.88 0.57 0.09 0.05 0.08 0.00 0.00 0.60 0.00 0.00

10 0.00 0.57 0.95 0.03 0.11 0.39 0.10 0.71 0.08 0.00 0.18 0.60 0.00 0.00

11 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.30

12 0.00 0.03 0.10 0.51 0.00 0.09 0.00 0.43 0.00 0.18 0.00 0.00 0.01 0.00

13 0.01 0.04 0.10 0.00 0.65 0.40 0.58 0.07 0.60 0.60 0.00 0.00 0.00 0.00

14 0.00 0.00 0.00 0.15 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00

15 0.04 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.30 0.00 0.00 0.00

В целом, числа кг при сравнении выборок ЭЭГ для всех испытуемых невелики (менее 40 %) и это говорит о статистической неустойчивости выборок Х{ для ЭЭГ у любого испытуемого, находящегося в неизменном гомеостазе (человек спокойно сидел, релаксация). При этом сама ЭЭГ демонстрирует непрерывные и хаотические колебания биопотенциалов мозга, что нами характеризуется как реверберации. Непрерывный хаос ЭЭГ (табл. 1) и хаотические колебания (реверберации) являются главными характеристиками любой биоэлектрической ак-

тивности НСМ (у нас в виде ЭЭГ). Возникает вопрос: имеет ли это принципиальное значение для работы мозга и как это можно моделировать, если это базовое свойство НСМ?

Эти же выборки ЭЭГ быстрым преобразованием Фурье обрабатывались с целью получения их (выборок) СПС. Для каждого испытуемого таким образом было получено по 15 выборок его СПС (для каждой ЭЭГ) и эти СПС тоже обрабатывались до расчета матриц парных сравнений выборок. Находились тоже 12 матриц парных статистических сравнений (по критерию Вилкоксона, для каждого испытуемого одна, своя матрица СПС), и в них находились к2. Характерная такая матрица представлена в табл. 2, где число совпадений к2=48, т.е. несколько превышает табл. 1 для кг.

Здесь уже к2>40%, т.е. больше чем кг, но в любом случае к2 для СПС не превышает 50% от всех (разных) 105 пар сравнения выборок СПС. Таким образом, и спектральные плотности сигнала не могут демонстрировать статистические совпадения выборок (это неустойчивый процесс). Однако, для СПС мы имеем несколько большую долю стохастики, т.е. к2 доходит до 50%. Напомним, что в статистике выборки совпадают, если вероятность Р^0,95. В нашем случае для табл. 1 и 2 мы тогда должны иметь не менее 97 пар совпадений. При этом подчеркнем, что в наших таблицах каждая пара имеет свою (особую) генеральную совокупность, т.е. даже если мы будем иметь к2=97, то эти пары в основном будут различны, у них будут разные генеральные совокупности, хотя в табл. 1 мы имеем примеры, когда 3-я выборка совпадает с 8-й, 9-й и 10-й, например (но это редкие явления).

Все это доказывает глобальность ЭЕЗ, когда не только к2<50%, но и все эти пары имеют разные генеральные совокупности. Подобная закономерность у нас получилась и для автокорреляций А(Ь), что представлено в табл. 3. Здесь число пар выборок А(Ь) кз тоже невелико (кз=35). Обычно к3^40% для всех 12-ти матриц парных сравнений выборок А(Ь). Таким образом и А(Ь) тоже демон-

10иККЛЬ ОБ ОТШ МЕБТСЛЬ ТЕСЫК0ШЫЕ8 - 2018 - V. 25, № 4 - Р. 304-314

стрирует статистическую неустойчивость выборок, что полностью согласуется с ЭЕЗ. Все это показывает, что НСМ работают хаотично, но в режиме непрерывных ревербераций [7,27-32].

Матрица парного сравнения СПС ЭЭГ одного и того же здорового человека (число повторов N=15), использовался критерий Вилкоксона (критическое значение р<0,05, число совпадений кг=48)

Хаос НСМ по параметрам ЭЭГ (их статистических функций х)), их СПС и А(Ь) требует аналогичного хаоса и в моделях работы НСМ. Мы сейчас это реализуем на базе нейроэмуля-торов, когда многократно заставляем нейроэмулятор решать одну и ту же задачу

Таблица 2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1 0,00 0,00 0,11 0,00 0,00 0,16 0,00 0,00 0,00 0,00 0,93 0,01 0,00 0,01

2 0,00 0,47 0,00 0,85 0,71 0,00 0,00 0,06 0,86 0,00 0,00 0,25 0,00 0,59

3 0,00 0,47 0,01 0,42 0,91 0,00 0,01 0,09 0,37 0,53 0,00 0,27 0,00 0,27

4 0,11 0,00 0,01 0,02 0,01 0,12 0,31 0,96 0,00 0,02 0,05 0,13 0,00 0,10

5 0,00 0,85 0,42 0,02 0,18 0,00 0,02 0,06 0,98 0,14 0,00 0,03 0,00 0,06

6 0,00 0,71 0,91 0,01 0,18 0,00 0,06 0,01 0,64 0,23 0,00 0,08 0,00 0,34

7 0,16 0,00 0,00 0,12 0,00 0,00 0,00 0,12 0,00 0,00 0,53 0,00 0,02 0,01

8 0,00 0,00 0,01 0,31 0,02 0,06 0,00 0,73 0,02 0,00 0,16 0,22 0,00 0,42

9 0,00 0,06 0,09 0,96 0,06 0,01 0,12 0,73 0,15 0,37 0,14 0,10 0,00 0,49

10 0,00 0,86 0,37 0,00 0,98 0,64 0,00 0,02 0,15 0,46 0,00 0,13 0,00 0,38

11 0,00 0,00 0,53 0,02 0,14 0,23 0,00 0,00 0,37 0,46 0,00 0,53 0,00 0,49

12 0,93 0,00 0,00 0,05 0,00 0,00 0,53 0,16 0,14 0,00 0,00 0,00 0,00 0,00

13 0,01 0,25 0,27 0,13 0,03 0,08 0,00 0,22 0,10 0,13 0,53 0,00 0,00 0,00

14 0,00 0,00 0,00 0,00 0,00 0,00 0,02 0,00 0,00 0,00 0,00 0,00 0,00 0,00

15 0,01 0,59 0,27 0,10 0,06 0,34 0,01 0,42 0,49 0,38 0,49 0,00 0,00 0,00

Таблица 3

Матрица парного сравнения Л(Ь) ЭЭГ одного и того же здорового человека (число повторов N=15), использовался критерий Вилкоксона (критическое значение р<0,05, число совпадений кз=35)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1 0,95 0,81 0,30 0,39 0,00 0,00 0,20 0,00 0,00 0,00 0,32 0,08 0,00 0,38

2 0,95 0,75 0,80 0,10 0,00 0,00 0,29 0,00 0,00 0,00 0,93 0,04 0,00 0,43

3 0,81 0,75 0,75 0,58 0,00 0,00 0,38 0,01 0,00 0,00 0,36 0,31 0,00 0,40

4 0,30 0,80 0,75 0,00 0,00 0,00 0,11 0,00 0,00 0,00 0,69 0,02 0,00 0,34

5 0,39 0,10 0,58 0,00 0,00 0,00 0,52 0,14 0,30 0,00 0,01 0,65 0,00 0,12

6 0,00 0,00 0,00 0,00 0,00 0,07 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00

7 0,00 0,00 0,00 0,00 0,00 0,07 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00

8 0,20 0,29 0,38 0,11 0,52 0,00 0,00 0,04 0,09 0,00 0,03 0,46 0,00 0,03

9 0,00 0,00 0,01 0,00 0,14 0,00 0,00 0,04 0,78 0,00 0,00 0,28 0,00 0,00

10 0,00 0,00 0,00 0,00 0,30 0,00 0,00 0,09 0,78 0,00 0,00 0,07 0,00 0,00

11 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00

12 0,32 0,93 0,36 0,69 0,01 0,00 0,00 0,03 0,00 0,00 0,00 0,00 0,03 0,72

13 0,08 0,04 0,31 0,02 0,65 0,00 0,00 0,46 0,28 0,07 0,00 0,00 0,00 0,03

14 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,03 0,00 0,00

15 0,38 0,43 0,40 0,34 0,12 0,00 0,00 0,03 0,00 0,00 0,00 0,72 0,03 0,00

Таблица 4

Расчёт весовых ш для шести признаков - параметров х, ССС школьников при широтных перемещениях (мальчики) в режиме многих п итераций (п=1000)

Расчеты итераций по выборкам (№1000) Веса признаков х- после Л=1000 итераций НЭВМ

п=1000 ¡=(1,.., 1000) х1 - БШ х2 - РАЯ хз - БББ хл - БЮММ х; - 1ЛВ хв - Бр02

1 и 2 0,22 0,09 0,08 0,12 0,39 1,00

1 и 3 0,55 0,33 0,42 0,62 0,55 0,87

разделения выборок х, описывающих гомеостаз одной и той же группы испытуемых (в неизменном гомео-стазе). Тогда на каждой итерации (настройке НЭВМ) мы задаем одни и те же выборки х-, но при этом хаотически (из равномерного распределения ^юЕ(0,1)) задаем начальные веса диагностических признаков х- для данной группы испытуемых [7,8,12,13].

В качестве примера мы приводим диаграмму расчёта весовых коэффициентов (при настройке НЭВМ) по 6-ти признакам х-, описывающих состояние сердечнососудистой системы (ССС) учащихся ХМАО-Югры при их широтных перемещениях. Использовались параметры х-, описывающие ССС, которые в каждой итерации имели свои (особые) значения весов признаков х-, что и показывает рис. 1. Здесь вертикальные столбцы (для каждого х-) показывают их значение М), которое от итерации к итерации меняется хаотически. Рисунок демонстрирует хаос Ш- при хаосе Шю, но в итоге статистически появляется различие между весами признаков, хотя статистика не дает таких различий. В итоге, после п итераций (обычно п>1000) мы получаем существенные различия для х-, т.е. можно найти параметры порядка - главные диагностические признаки х-*, что представлено в табл. 4. Их число 2 всегда будет небольшим (для табл. 4 имеем 2=п=3 в виде хгБШ-

ВЕСТНИК НОВЫХ МЕДИЦИНСКИХ ТЕХНОЛОГИЙ - 2018 - Т. 25, № 4 - С. 304-314 JOURNAL OF NEW MEDICAL TECHNOLOGIES - 2018 - V. 25, № 4 - P. 304-314

^=0,55; X6-Sp02=087; X4-SDNN=0,62), что фактически моделирует эвристическую работу мозга человека. НЭВМ в режиме п итераций (с хаосом Wio) дает нам модели эвристической работы мозга человека (врача при постановке правильного диагноза), когда статистика вообще не показывает различий (неопределенность 1-го типа) и мы имеем большой дефицит такой информации об объекте (у нас это параметры ССС при переезде с Севера РФ на Юг).

Выводы:

1. Генерация биопотенциалов мозга (в виде ЭЭГ) происходит хаотически. Полученные подряд выборки ЭЭГ, их СПС и А^) статистически не совпадают. Доля стохастики не превышает 50%, но и получаемые параметры между собой тоже статистически не совпадают. Мы имеем множество разных генеральных совокупностей для подряд получаемых выборок, которые все будут различны. Это доказывает ЭЕЗ для ЭЭГ (при этом испытуемые находятся в неизменном гомеостазе).

Если ввести хаос и реверберации в работу НСМ человека (в виде нейроэмуляторов, как моделей НСМ), то при многих реверберациях (п>1000) получим разделение весов диагностических признаков xi. Это имитирует работу мозга человека (в режиме эвристики), что находит применение в клинической медицине (при системном синтезе). Очевидны перспективы применения таких НЭВМ (в режиме хаоса и ревербераций) в клинической медицине при постановке правильных диагнозов.

Рис. 1. Диаграмма распределения весов Wi признаков Xi состояния ССС мальчиков в 1-ом и 2-ом измерениях для каждого )-того обучения (метод градиентного спуска) искусственной нейронной сети (/=1,.., 1000, (как пример п=50)), при хаотическом начальном «встряхивании» wюE(0,l) из равномерного распределения

NEURAL BRAIN NETWORKS AND THEIR MODELING THROUGH NEURAL EMULATORS

AN. INIUSHKIN*, М.А. FILATOV", S.V. GRIGORIEVA**, I.B. BULATOV**

*Samara State Aerospace University of academician C.P. Korolev, Moscow highway 34, city of Samara, 443086, Russia **Surgut State University, Lenina pr., 1, Surgut, 628400, Russia, e-mail: firing.squad@mail.ru

Abstract. Traditionally, the dynamics of the bioelectrical activity of the brain as a result of the operation of the neural networks of the brain, have been described within the framework of statistical methods. However, during 25-30 years the hypothesis of N A. Bernstein on "repetition without repetition" began to be realized. The proof of this hypothesis is the effect of Eskov-Zinchenko. The purpose of this investigation is to prove instability in the operation of the neural networks of the brain in all parameters of electroencephalograms. As part of this effect, continuous chaos and reverberation in the work of the neural networks of the brain is proved. These two basic properties are transferred to the work of neural emulators. It is shown that the assignment of chaotic initial values of the weights wi0 of diagnostic signs xi and numerous reverberations (repetition of the neural network setting) provide system synthesis, i.e. simulate the heuristic activity of the human brain in conditions of lack of information.

JOURNAL OF NEW MEDICAL TECHNOLOGIES - 2018 - V. 25, № 4 - P. 304-314 Key words: chaos, neural networks, the effect of Eskov-Zinchenko, reverberation.

Introduction. For a long time, all the sciences of the brain used traditional statistical methods to process the basic parameters of neural networks (the activity of individual neurons or electroencephalograms - EEG). The statistical distribution functions f(xi) were calculated for the Xi parameters of the EEG, the spectral signals densities of the— the SSD (for the EEG, for example), the autocorrelation A(t), etc. At the same time, all these characteristics were considered repeatable and self-reproducible in observations [2-8].

However, in 1947 N.A. Bernstein put forward the hypothesis of «repetition without repetition» in bio-mechanics, which has now been confirmed in the effect of Eskov-Zinchenko (EEZ). This EEZ proves the absence of statistical stability for successively obtained samples of tremorograms (TMG) in one test subject in unchanged homeostasis. If the TMG samples are not statistically repeated, then it can be assumed that the work of the central regulator of any movements (neural networks of the brain - human NNB) cannot be arbitrarily and statistically repeated. This paper aims to verify this statement [1-7].

iНе можете найти то, что вам нужно? Попробуйте сервис подбора литературы.

Purpose of the research: proof of statistical instability for successively obtained EEG samples from one test subject (in unchanged homeostasis) and the use of such chaos in the work of neural emulators (neuro-computer).

The object and methods of research. According to the Helsinki Declaration, an EEG was recorded from 12 people (in the same lead). At the same time, for each test subject, the EEG recording was performed in an interval of 5 seconds. 15 times (from the same brain area), and the received analog signals x(t) were quantized with a quantization period t=10 ms and the obtained discrete EEG values (as discrete files, in digital) were recorded as samples x(t) such signals (EEG) [7-14].

Thus, 15 EEG samples were obtained for each subject, each sample contained 500 points (EEG recording period T=5 sec). These 15 EEG samples were processed statistically by comparing and constructing matrices of such paired comparisons of EEG samples (as x1(t)). For each test subject, their own matrix of pairwise comparisons of samples was built (according to the Wilcoxon criterion), which contained 105 independent (different) EEG comparison pairs [7,20-28].

In such matrices of paired comparisons of EEG samples there were pairs of samples for which the Wilcoxon P criterion is P>0,05. In this case, these two

samples (this pair) could be attributed to one (common) general population. The numbers k were determined for each such matrix, i.e. there was a number of pairs k for which P>0,05. At the same time, the same EEG samples were calculated for 15 SSD samples and 15 A(t) samples for each subject [17-22,2832]. In this case, the matrices for the SSD and the matrices for A(t) were still (separately) constructed.

Thus, 12 matrices for EEG, 12 matrices for SSD and 12 matrices for A(t) for all 12 subjects were obtained. In these matrices were ki (for EEG), k2 (for SSD) and k3 (for A(t)). A comparison of these k1, k2, k3 was made for each of the 12 subjects.

Results of research and discussion. Note that all 12 obtained k1 values for EEG had different values (sometimes k1 coincided for individual subjects), but all k1 did not exceed k1<40% of all 105 pairs of (different) comparisons of EEG samples. For example, we present the characteristic matrix (table 1), where the number ki=37. This is a fairly large number (usually ki^25), but this number is less than 40%.

In general, the numbers of k1 when comparing EEG samples for all subjects are small (less than 40%). This fact is evidence of the statistical instability of the xi samples for the EEG of any test subject in unchanged homeostasis (human was sitting quietly, relaxation). At the same time, the EEG demonstrates continuous and chaotic oscillations of the brain biopotentials (EEG). From our point of view, this is a reverb. Continuous EEG chaos (table 1) and chaotic oscillations (reverberations) are the main characteristics of any bioelectric activity of the NNB (in our case, in the form of EEG). The question arises: is it of fundamental importance for the work of the brain and how can it be modeled if it is the basic property of NNB?

The same EEG samples were processed using the Fourier transform to obtain their (samples) signal spectral densities. For each subject, 15 samples of his SSD were obtained (for each EEG). These SSD were also processed prior to the calculation of the matrix of pairwise comparisons of samples. There were 12 matrices of paired statistical comparisons (according to the Wilcoxon criterion, one for each subject, his own SSD matrix), and they calculated k2. The characteristic matrix is presented in table 2, where the number of coincidences is k2=48, i.e. somewhat higher than table 1 for k1.

In this case, k2040%, i.e. more than ki. But, in any case, k2 for the SSD does not exceed 50% of all (different) 105 pairs of comparison of SSD sam-

JOURNAL OF NEW MEDICAL TECHNOLOGIES - 2018 - V. 25, № 4 - P. 304-314

ples. Thus, signal spectral densities cannot demonstrate statistical coincidences of samples (this is an unstable process). However, for SSD, we have a somewhat larger share of stochastics, i.e. k2 reaches 50%. Recall that in statistics, the samples are the same if the probability P>0,95. In our case, for the table 1 and table 2 then we must have at least 97 pairs of matches. In this case, we emphasize that in our tables, each pair has its own (special) general population. For example, if we have k2=97, then these pairs will be basically different, they will have different population sets, although in table 1 we have examples when the 3-rd sample coincides with the 8-th, 9-th and 10-th, for example (these are rare phenomena).

The matrix pairwise comparison of EEG of the same healthy human (number of repetitions N=15), Wilcoxon test was used (critical value p<0,05, number of coincidences ki=37)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1 0.00 0.00 0.00 0.00 0.00 0.19 0.00 0.00 0.00 0.01 0.00 0.01 0.00 0.04

2 0.00 0.99 0.00 0.06 0.93 0.02 0.25 0.33 0.57 0.00 0.03 0.04 0.00 0.00

3 0.00 0.99 0.00 0.09 0.75 0.03 0.21 0.50 0.95 0.00 0.10 0.10 0.00 0.00

4 0.00 0.00 0.00 0.00 0.00 0.00 0.02 0.00 0.03 0.00 0.51 0.00 0.15 0.00

5 0.00 0.06 0.09 0.00 0.28 0.29 0.00 0.88 0.11 0.00 0.00 0.65 0.00 0.00

6 0.00 0.93 0.75 0.00 0.28 0.11 0.07 0.57 0.39 0.00 0.09 0.40 0.00 0.00

7 0.19 0.02 0.03 0.00 0.29 0.11 0.00 0.09 0.10 0.00 0.00 0.58 0.00 0.01

8 0.00 0.25 0.21 0.02 0.00 0.07 0.00 0.05 0.71 0.00 0.43 0.07 0.00 0.00

9 0.00 0.33 0.50 0.00 0.88 0.57 0.09 0.05 0.08 0.00 0.00 0.60 0.00 0.00

10 0.00 0.57 0.95 0.03 0.11 0.39 0.10 0.71 0.08 0.00 0.18 0.60 0.00 0.00

11 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.30

12 0.00 0.03 0.10 0.51 0.00 0.09 0.00 0.43 0.00 0.18 0.00 0.00 0.01 0.00

13 0.01 0.04 0.10 0.00 0.65 0.40 0.58 0.07 0.60 0.60 0.00 0.00 0.00 0.00

14 0.00 0.00 0.00 0.15 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00

15 0.04 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00 0.30 0.00 0.00 0.00

when not only k2<50%, but all these couples have different general population. A similar pattern is true for autocorrelations A(t), which is presented in table 3. Here the number of pairs of samples A(t) k3 is also small (k3= 35). Usually k3^40% for all 12 matrices of paired comparisons of samples A(t). Thus, A(t) also demonstrates the statistical instability of the samples, which is fully consistent with EEZ. All this shows that the NNB work chaotically, but in the mode of continuous reverberations [7,27-32].

Chaos NNB according to EEG parameters (their statistical functions f(x)), their SSD and A(t) requires similar chaos in the models of the operation of the NNB. Now we are implementing this on the basis of neural emulators (or neuro-computer), Table 1 when we repeatedly force the neural emulator to solve the same problem of separating samples xi, describing the ho-meostasis of the same group of subjects (in constant homeos-tasis). Then at each iteration (settings of the computer) we set the same samples xi, but randomly (from a uniform distribution WioE (0,1)) we set the initial weights wi0 of diagnostic signs xi for this group of subjects [7,8,12,13].

As an example, we present a chart for calculating weights Wi (when setting up a comput-

Table 2

The matrix pairwise comparison of SSD of EEG of the same healthy human (number of repetitions N=15), Wilcoxon test was used (critical value p<0,05, number of coincidences fe=48)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1 0,00 0,00 0,11 0,00 0,00 0,16 0,00 0,00 0,00 0,00 0,93 0,01 0,00 0,01

2 0,00 0,47 0,00 0,85 0,71 0,00 0,00 0,06 0,86 0,00 0,00 0,25 0,00 0,59

3 0,00 0,47 0,01 0,42 0,91 0,00 0,01 0,09 0,37 0,53 0,00 0,27 0,00 0,27

4 0,11 0,00 0,01 0,02 0,01 0,12 0,31 0,96 0,00 0,02 0,05 0,13 0,00 0,10

5 0,00 0,85 0,42 0,02 0,18 0,00 0,02 0,06 0,98 0,14 0,00 0,03 0,00 0,06

6 0,00 0,71 0,91 0,01 0,18 0,00 0,06 0,01 0,64 0,23 0,00 0,08 0,00 0,34

7 0,16 0,00 0,00 0,12 0,00 0,00 0,00 0,12 0,00 0,00 0,53 0,00 0,02 0,01

8 0,00 0,00 0,01 0,31 0,02 0,06 0,00 0,73 0,02 0,00 0,16 0,22 0,00 0,42

9 0,00 0,06 0,09 0,96 0,06 0,01 0,12 0,73 0,15 0,37 0,14 0,10 0,00 0,49

10 0,00 0,86 0,37 0,00 0,98 0,64 0,00 0,02 0,15 0,46 0,00 0,13 0,00 0,38

11 0,00 0,00 0,53 0,02 0,14 0,23 0,00 0,00 0,37 0,46 0,00 0,53 0,00 0,49

12 0,93 0,00 0,00 0,05 0,00 0,00 0,53 0,16 0,14 0,00 0,00 0,00 0,00 0,00

13 0,01 0,25 0,27 0,13 0,03 0,08 0,00 0,22 0,10 0,13 0,53 0,00 0,00 0,00

14 0,00 0,00 0,00 0,00 0,00 0,00 0,02 0,00 0,00 0,00 0,00 0,00 0,00 0,00

15 0,01 0,59 0,27 0,10 0,06 0,34 0,01 0,42 0,49 0,38 0,49 0,00 0,00 0,00

er) according to 6 signs Xi, describing the state of the cardiovascular system of pupils of KHMAO-Ugra during their latitudinal movements. Parameters Xi were used, describing the cardiovascular system, which in each iteration had their own (special) values of weights of signs xi, as shown in figure 1. Here the vertical columns (for each xi) show their value (w), which varies chaotically from iteration to iteration.

All these facts prove the global nature of EEZ,

ВЕСТНИК НОВЫХ МЕДИЦИНСКИХ ТЕХНОЛОГИЙ - 2018 - Т. 25, № 4 - С. 304-314 JOURNAL OF NEW MEDICAL TECHNOLOGIES - 2018 - V. 25, № 4 - P. 304-314

The matrix pairwise comparison of A(t) of EEG of the same healthy human (number of repetitions N=15), Wilcoxon test was used (critical value p<0,05, number of coincidences fcs=35)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1 0,95 0,81 0,30 0,39 0,00 0,00 0,20 0,00 0,00 0,00 0,32 0,08 0,00 0,38

2 0,95 0,75 0,80 0,10 0,00 0,00 0,29 0,00 0,00 0,00 0,93 0,04 0,00 0,43

3 0,81 0,75 0,75 0,58 0,00 0,00 0,38 0,01 0,00 0,00 0,36 0,31 0,00 0,40

4 0,30 0,80 0,75 0,00 0,00 0,00 0,11 0,00 0,00 0,00 0,69 0,02 0,00 0,34

5 0,39 0,10 0,58 0,00 0,00 0,00 0,52 0,14 0,30 0,00 0,01 0,65 0,00 0,12

6 0,00 0,00 0,00 0,00 0,00 0,07 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00

7 0,00 0,00 0,00 0,00 0,00 0,07 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00

8 0,20 0,29 0,38 0,11 0,52 0,00 0,00 0,04 0,09 0,00 0,03 0,46 0,00 0,03

9 0,00 0,00 0,01 0,00 0,14 0,00 0,00 0,04 0,78 0,00 0,00 0,28 0,00 0,00

10 0,00 0,00 0,00 0,00 0,30 0,00 0,00 0,09 0,78 0,00 0,00 0,07 0,00 0,00

11 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00

12 0,32 0,93 0,36 0,69 0,01 0,00 0,00 0,03 0,00 0,00 0,00 0,00 0,03 0,72

13 0,08 0,04 0,31 0,02 0,65 0,00 0,00 0,46 0,28 0,07 0,00 0,00 0,00 0,03

14 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,03 0,00 0,00

15 0,38 0,43 0,40 0,34 0,12 0,00 0,00 0,03 0,00 0,00 0,00 0,72 0,03 0,00

Table 4

Calculation of weight w for six signs - parameters Xi of cardiovascular system for schoolchildren at latitudinal movements (boys) in the mode of many n iterations (n=1000)

Calculations iterations by samples (№1000) Weights wt signs xi after N=1000 iterations of the neural computer

n=1000 j=(l,..., 1000) xi - SIM X2 - PAR Хз - SSS X4 - SDNN xs - INB X6 - SpO2

1и 2 0,22 0,09 0,08 0,12 0,39 1,00

1 и 3 0,55 0,33 0,42 0,62 0,55 0,87

Fig. 1. The distribution diagram of weights Wi for signs xi of the state of cardiovascular system of boys in the 1-st and 2-nd dimensions for each j-th training (gradient descent method) of an artificial neural network (j=1, .., 1000, (as an example n=50 )), with random initial "shaking" wmE (0,1) from a uniform distribution

Table 3 The figure demonstrates the chaos of Wt in the chaos of Wi0, but as a result, there is a statistically different difference between the weights of the signs, although the statistics do not give such differences. As a result, after n iterations (usually nM000), we get significant differences for xi, i.e. you can find the order parameters - the main diagnostic features of xi*, which are presented in table 4. Their number Z will always be small (for table 4 we have Z=n=3 in the form X1-SIM-W,=0,55; X6-Sp02=0,87; X4-SDNN=0,62), which actually simulates the heuristic human brain work. Neuro-computer in n-iteration mode (with Wi0 chaos) gives us models of the heuristic work of the human brain (the doctor when making the correct diagnosis), when the statistics do not show differences at all (type 1 uncertainty) and we have a large deficit of such information about the object (we have these are parameters of the cardiovascular system when moving from the North of the Russian Federation to the South).

Conclusions:

1. The generation of brain biopotentials (in the form of EEG) occurs randomly. The consecutive EEG samples, their SSD and A(t) statistically do not match. The share of stochastics does not exceed 50%, but the parameters obtained are also not statistically the same. We have a lot of different general sets for successively received samples, which will all be different. This proves the EEZ for the EEG (while the subjects are in a constant ho-meostasis).

2. If chaos and reverberations are introduced into the work of a human's NNB (in the form of neural emulators, as NNB models), then with many reverberations (n5T000) we obtain the separation of weights of diagnostic signs xi. This simulates the work of the human brain (in the mode of heuristics), which finds application in clinical medicine (in system synthesis). Obvious are the prospects for the use of such neural computers (in the mode of chaos and reverberations) in clinical medicine when making correct diagnoses.

JOURNAL OF NEW MEDICAL TECHNOLOGIES - 2018 - V. 25, № 4 - P. 304-314

Литература

1. Белощенко Д.В., Якунин В.Е., Потетюрина Е.С., Королев Ю.Ю. Оценка параметров электромио-грамм у женщин при разном статическом усилии в режиме повторения // Клиническая медицина и фармакология. 2017. Т. 3, № 1. С. 26-31.

References

1. Beloshchenko DV, Yakunin VE, Potetyurina ES, Korolev YuYu. Otsenka parametrov elektromiogramm u zhenshchin pri raznom staticheskom usilii v rez-hime povtoreniya [Evaluation of the parameters of electromyograms in women with different static forces in the repetition mode]. Klinicheskaya medit-sina i farmakologiya. 2017;3(1):26-1. Russian.

2. Гавриленко Т.В., Горбунов Д.В., Чертищев А.А., Булатов И.Б. Оценка параметров электромио-грамм с позиции термодинамики неравновесных систем И.Р. Пригожина // Вестник кибернетики. 2017. Т. 26, № 2. С. 109-114.

2. Gavrilenko TV, Gorbunov DV, CHertishchev AA, Bulatov IB. Ocenka parametrov ehlektromiogramm s pozicii termodinamiki neravnovesnyh sistem I.R. Pri-gozhina [Estimation of electromyogram parameters from the position of thermodynamics of nonequili-brium systems I.R. Prigozhin]. Vestnik kibernetiki. 2017;26(2):109-14. Russian.

3. Денисова Л.А., Белощенко Д.В., Башкато-ва Ю.В., Горбунов Д.В. Особенности регуляции двигательных функций у женщин // Клиническая медицина и фармакология. 2017. Т. 3, № 4. С. 1116.

3. Denisova LA, Beloshchenko DV, Bashkato-va YUV, Gorbunov DV. Osobennosti regulyacii dviga-tel'nyh funkcij u zhenshchin [Features of regulation of motor functions in women]. Klinicheskaya medicina i farmakologiya. 2017;3(4):11-6. Russian.

4. Денисова Л.А., Прохоров С.А., Шакирова Л.С., Филатова Д.Ю. Хаос параметров сердечнососудистой системы школьников в условиях широтных перемещений // Вестник новых медицинских технологий. 2018. №1. С. 133-142. БОТ: 10.24411/1609-2163-2018-15989.

5. Еськов В.В. Возможности термодинамического подхода в электромиографии // Вестник кибернетики. 2017. Т. 28, № 4. С. 109-115.

6. Еськов В.В., Еськов В.М., Вохмина Ю.В. Гипотеза Н.А. Бернштейна и статистическая неустойчивость выборок параметров треморограмм // Вестник кибернетики. 2018. Т. 29, № 1. С. 33-38.

4. Denisova LA, Prohorov SA, SHakirova LS, Filato-va DYU. Haos parametrov serdechno-sosudistoj sis-temy shkol'nikov v usloviyah shirotnyh peremesh-chenij [Chaos of parameters of cardiovascular system of schoolchildren in conditions of latitudinal movements]. Vestnik novyh medicinskih tekhnologij. 2018;25(1):133-42. Russian. DOI: 10.24411/16092163-2018-15989.

5. Es'kov VV. Vozmozhnosti termodinamicheskogo podhoda v ehlektromiografii [Possible thermodynamic approach to electromyography]. Vestnik kibernetiki. 2017;28(4):109-15. Russian.

6. Es'kov VV, Es'kov VM, Vohmina YUV. Gipoteza N.A. Bernshtejna i statisticheskaya neustojchivost' vyborok parametrov tremorogramm [Hypothesis N.. Bernstein and statistical instability of samples parameters of tremorogramm]. Vestnik kibernetiki. 2018;29(1):33-8. Russian.

7. Еськов В.В., Пятин В.Ф., Клюс Л.Г., Миллер А.В. Гомеостатичность нейросетей мозга // Вестник новых медицинских технологий. 2018. №1. C. 102-113. DOI: 0.24411/1609-2163-2018-15985.

7. Es'kov VV, Pyatin VF, Klyus LG, Miller AV. Go-meostatichnost' nejrosetej mozga [Homeostaticity of brain neural networks]. Vestnik novyh medicinskih tekhnologij. 2018;1:102-13. DOI: 0.24411/1609-21632018-15985. Russian.

8. Еськов В.В. Проблема статистической неустой-чивоси в биомеханике и в биофизике в целом // Вестник новых медицинских технологий. 2018. №2. С. 166-175. БО1: 10.24411/1609-2163-201816071.

9. Еськов В.В., Пятин В.Ф., Еськов В.М., Григорьева С.В. Особенности регуляции сердечно-

iНе можете найти то, что вам нужно? Попробуйте сервис подбора литературы.

8. Es'kov VV. Problema statisticheskoj neustojchi-vosi v biomekhanike i v biofizike v celom [The problem of statistical instability in biomechanics and Biophysics in General]. Vestnik novyh medicinskih tekh-nologij. 2018;2:166-75. DOI: 10.24411/1609-21632018-16071. Russian.

9. Es'kov VV, Pyatin VF, Es'kov VM, Grigor'eva SV. Osobennosti regulyacii serdechno-sosudistoj sistemy

JOURNAL OF NEW MEDICAL TECHNOLOGIES - 2018 - V. 25, № 4 - P. 304-314

сосудистой системы организма человека нейросе-тями мозга // Вестник новых медицинских технологий. 2018. №2. С. 188-199. Б01: 10.24411/16092163-2018-16073.

10. Еськов В.В., Белощенко Д.В., Баженова А.Е., Живаева Н.В. Влияние локального холодового воздействия на параметры электромиограмм у женщин // Экология человека. 2018. № 9. С. 42-47.

11. Еськов В.М., Томчук А.Г., Широков В.А., Ураева Я.И. Стохастический и хаотический анализ вертеброневрологических показателей и визуальной аналоговой шкалы боли в комплексном лечении хронических мышечно-скелетных болей // Клиническая медицина и фармакология. 2017. Т. 3, № 3. С. 8-12. Б01: 10.12737/аШс1е_ 5а1а3М432аП.116613П.

12. Еськов В.М., Зинченко Ю.П., Филатова О.Е. Естествознание: от стохастики к хаосу и самоорганизации // Вестник кибернетики. 2017. № 1 (25). С. 121-127.

13. Еськов В.М., Вохмина Ю.В., Горбунов С.В., Шейдер А.Д. Кинематика гомеостатических систем // Вестник кибернетики. 2017. № 2 (26). С. 8793.

14. Еськов В.М., Зинченко Ю.П., Филатова О.Е. Признаки парадигмы и обоснование третьей парадигмы в психологии // Вестник московского университета. Серия 14: Психология. 2017. №. 1. С. 3-17.

15. Мирошниченко И.В., Григоренко В.В., Филатова Д.Ю., Мнацаканян Ю.В. Особенности поведения параметров сердечно-сосудистой системы школьников при широтных перемещениях // Сложность. Разум. Постнеклассика. 2018. № 1. С. 45-51.

16. Прохоров С.А., Белощенко Д.В., Шейдер А.Д., Горбунова М.Н. Методы теории хаоса-самоорганизации в оценке параметров систем третьего типа-сотр1ехку // Сложность. Разум. Постнеклассика. 2018. № 2. С. 32-41.

17. Пятин В.Ф., Еськов В.В., Алиев Н.Ш.,

organizma cheloveka nejrosetyami mozga [Features of regulation of the cardiovascular system of the human body by brain neural networks]. Vestnik novyh medicinskih tekhnologij. 2018;2:188-99. DOI: 10.24411/1609-2163-2018-16073. Russian.

10. Es'kov VV, Beloshchenko DV, Bazhenova AE, ZHivaeva NV. Vliyanie lokal'nogo holodovogo voz-dejstviya na parametry ehlektromiogramm u zhensh-chin [Effect of local cold exposure on the parameters of electromyograms in women]. EHkologiya cheloveka. 2018;9:42-7. Russian.

11. Es'kov VM, Tomchuk AG, Shirokov VA, Uraeva YaI. Stokhasticheskiy i khaoticheskiy analiz vertebro-nevrologicheskikh pokazateley i vizual'noy analogo-voy shkaly boli v kompleksnom lechenii khroni-cheskikh myshechno-skeletnykh boley [Stochastic and chaotic analysis of vertebroneurological indicators and visual analogue pain scale in the complex treatment of chronic musculoskeletal pain]. Klinicheskaya meditsina i farmakologiya. 2017;3(3):8-

12. DOI: 10.12737/article_ 5a1d3f9f432df1.11661311. Russian.

12. Es'kov VM, Zinchenko YUP, Filatova OE. Estest-voznanie: ot stokhastiki k khaosu i sa-moorganizacii [Natural science: from stochastics to chaos and self-organization]. Vestnik kibernetiki. 2017;1(25):121-7. Russian.

13. Es'kov VM, Vokhmina YUV, Gorbunov SV, SHejd-er AD. Kinematika gomeostaticheskikh sistem [Kinematics of homeostatic systems]. Vestnik kibernetiki. 2017;2(26):87-93. Russian.

14. Es'kov VM, Zinchenko YuP, Filatova OE. Priznaki paradigmy i obosnovanie tret'ey paradigmy v psikho-logii [Signs of the paradigm and the rationale for the third paradigm in psychology]. Vestnik moskovskogo universiteta. Seriya 14: Psikhologiya. 2017;1:3-17. Russian.

15. Miroshnichenko IV, Grigorenko VV, Filatova DYU, Mnacakanyan YUV. Osobennosti povedeniya parametrov serdechno-sosudistoj sistemy shkol'nikov pri shirotnyh peremeshcheniyah [Features of behavior of parameters of cardiovascular system of school students at latitudinal movements]. Slozhnost'. Razum. Postneklassika. 2018;1:45-51. Russian.

16. Prohorov SA, Beloshchenko DV, SHejder AD, Gorbunova MN. Metody teorii haosa-samoorganizacii v ocenke parametrov sistem tret'ego tipa-complexity [Methods of chaos theory-self-organization in the evaluation of the parameters of the third type of systems-complexity]. Slozhnost'. Razum. Postneklassika. 2018;2:32-41. Russian.

17. Pyatin VF, Es'kov VV, Aliev NSH, Vorob'eva LA.

JOURNAL OF NEW MEDICAL TECHNOLOGIES - 2018 - V. 25, № 4 - P. 304-314

Воробьева Л.А. Хаос параметров гомеостаза функциональных систем организма человека // Вестник новых медицинских технологий. 2018. Т. 25, № 1. С. 143-153. DOI: 10.24411/1609-2163-201815990.

Haos parametrov gomeostaza funkcional'nyh sistem organizma cheloveka [Chaos of homeostasis parameters of functional systems of human body]. Vestnik novyh medicinskih tekhnologij. 2018;25(1):143-53. DOI: 10.24411/1609-2163-2018-15990. Russian.

18. Филатова Д.Ю., Башкатова Ю.В., Филатов М.А., Иляшенко Л.К. Анализ параметров деятельности сердечно-сосудистой системы у школьников в условиях широтных перемещений // Экология человека. 2018. № 4. С. 30-35.

19. Эльман К.А., Срыбник М.А., Прасолова А.А., Волохова М.А. Сравнительный анализ функциональных систем организма коренного детско-юношеского населения в условиях Севера // Клиническая медицина и фармакология. 2017. Т. 3, № 3. С. 13-17.

20. Betelin V.B., Eskov V.M., Galkin V.A., Gavrilen-ko T.V. Stochastic Volatility in the Dynamics of Complex Homeostatic Systems // Doklady Mathematics. 2017. Vol. 95, No. 1. P. 92-94.

21. Eskov V.V., Filatova O.E., Gavrilenko T.V., Gorbu-nov D.V. Chaotic Dynamics of Neuromuscular System Parameters and the Problems of the Evolution of Complexity // Biophysics. 2017. Vol. 62, No. 6. P. 961-966.

18. Filatova DYU, Bashkatova YUV, Filatov MA, Ilya-shenko LK. Analiz parametrov deyatel'nosti serdech-no-sosudistoj sistemy u shkol'nikov v usloviyakh shi-rotnykh peremeshchenij [Analysis of the parameters of the cardiovascular system in schoolchildren in conditions of latitudinal displacement]. EHkologiya cheloveka. 2018;4:30-5. Russain.

19. EHl'man KA, Srybnik MA, Prasolova AA, Voloho-va MA. Sravnitel'nyj analiz funkcional'nyh sistem organizma korennogo detsko-yunosheskogo nasele-niya v usloviyah Severa [Comparative analysis of the functional systems of the indigenous youth population in the North]. Klinicheskaya medicina i farmako-logiya. 2017;3(3):13-7. Russian.

20. Betelin VB, Eskov VM, Galkin VA, Gavrilenko TV. Stochastic Volatility in the Dynamics of Complex Homeostatic Systems. Doklady Mathematics. 2017;95(1):92-4.

21. Eskov VV, Filatova OE, Gavrilenko TV, Gorbu-nov DV. Chaotic Dynamics of Neuromuscular System Parameters and the Problems of the Evolution of Complexity. Biophysics. 2017;62(6):961-6.

22. Eskov V.V., Gavrilenko T.V., Eskov V.M., Vochmi-na Yu.V. Static Instability Phenomenon in Type-Three Secretion Systems: Complexity // Technical Physics. 2017. Vol. 62, No. 11. P. 1611-1616.

23. Eskov V.M., Filatova O.E., Eskov V.V., Gavrilenko T.V. The Evolution of the Idea of Homeostasis: Determinism, Stochastics and Chaos-Self-Organization // Biophysics. 2017. Vol. 62, No. 5. P. 809-820.

22. Eskov VV, Gavrilenko TV, Eskov VM, Vochmi-na YuV. Static Instability Phenomenon in Type-Three Secretion Systems: Complexity. Technical Physics. 2017;62(11):1611-6.

23. Eskov VM, Filatova OE, Eskov VV, Gavrilenko TV. The Evolution of the Idea of Homeostasis: Determinism, Stochastics and Chaos-Self-Organization. Biophysics. 2017;62(5):809-20.

24. Eskov V.M., Eskov V.V., Gavrilenko T.V., Vochmi-na Yu.V. Formalization of the Effect of "Repetition without Repetition" Discovered by N.A. Bernshtein // Biophysics. 2017. Vol. 62, No. 1. P. 143-150.

25. Eskov V.M., Bazhenova A.E., Vochmina U.V., Filatov M.A., Ilyashenko L.K. N.A. Bernstein hypothesis in the Description of chaotic dynamics of involuntary movements of person // Russian Journal of Biomechanics. 2017. Vol. 21, No. 1. P. 14-23.

26. Filatova O.E., Eskov V.V., Filatov M.A., Ilyashenko L.K. Statistical instability phenomenon and evaluation of voluntary and involuntary movements // Russian Journal of Biomechanics. 2017. Vol. 21, No. 3. P. 224-232.

24. Eskov VM, Eskov VV, Gavrilenko TV, Vochmina YuV. Formalization of the Effect of "Repetition without Repetition" Discovered by N.A. Bernshtein. Biophysics. 2017;62(1):143-50.

25. Eskov VM, Bazhenova AE, Vochmina UV, Filatov MA, Ilyashenko LK. N.A. Bernstein hypothesis in the Description of chaotic dynamics of involuntary movements of person. Russian Journal of Biomechanics. 2017;21(1):14-23. Russian.

26. Filatova OE, Eskov VV, Filatov MA, Ilyashen-ko LK. Statistical instability phenomenon and evaluation of voluntary and involuntary movements. Russian Journal of Biomechanics. 2017;21(3):224-32.

JOURNAL OF NEW MEDICAL TECHNOLOGIES - 2018 - V. 25, № 4 - P. 304-314

27. Filatova O.E., Bazhenova A.E., Ilyashenko L.K., Grigorieva S.V. Estimation of the Parameters for Tremograms According to the Eskov-Zinchenko Effect Biophysics // Biophysics. 2018. Vol. 63, No. 2. P. 125-130.

27. Filatova OE, Bazhenova AE, Ilyashenko LK, Gri-gorieva SV. Estimation of the Parameters for Tremo-grams According to the Eskov-Zinchenko Effect Biophysics. Biophysics. 2018;63(2):125-30.

28. Ilyashenko L.K., Bazhenova A.E., Berestin D.K., Grigorieva S.V. Chaotic dynamics parameters of the tremorgrams at the stress exposure // Russian Journal of Biomechanics. 2018. Vol. 22, No. 1. P. 62-71.

28. Ilyashenko LK, Bazhenova AE, Berestin DK, Grigorieva SV. Chaotic dynamics parameters of the tre-morgrams at the stress exposure. Russian Journal of Biomechanics. 2018;22(1):62-71.

29. Leonov B.I., Grigorenko V.V., Eskov V.M., Kha-dartsev A.A., Ilyashenko L.K. Automation of the Diagnosis of Age-Related Changes in Parameters of the Cardiovascular System // Biomedical Engineering. 2018. Vol. 52, No. 3. P. 210-214.

29. Leonov BI, Grigorenko VV, Eskov VM, Khadart-sev AA, Ilyashenko LK. Automation of the Diagnosis of Age-Related Changes in Parameters of the Cardiovascular System. Biomedical Engineering. 2018;52(3):210-4.

30. Zilov V.G., Eskov V.M., Khadartsev A.A., Eskov V.V. Experimental confirmation of the effect of "Repetition without repetition" N.A. Bernstein // Bulletin of experimental biology and medicine. 2017. Vol. 1. P. 4-8.

30. Zilov VG, Eskov VM, Khadartsev AA, Eskov VV. Experimental confirmation of the effect of "Repetition without repetition" N.A. Bernstein. Bulletin of experimental biology and medicine. 2017;1:4-8.

31. Zilov V.G., Khadartsev A.A., Eskov V.V., Eskov V.M. Experimental Study of Statistical Stability of Cardiointerval Samples // Bulletin of experimental biology and medicine. 2017. Vol. 164, No 2. P. 115117.

31. Zilov VG, Khadartsev AA, Eskov VV, Eskov VM. Experimental Study of Statistical Stability of Car-diointerval Samples. Bulletin of experimental biology and medicine. 2017;164(2):115-7.

32. Zilov V.G., Khadartsev A.A., Ilyashenko L.K., Eskov V.V., Minenko I.A. Experimental analysis of the chaotic dynamics of muscle biopotentials under various static loads // Bulletin of experimental biology and medicine. 2018. Vol. 165, No. 4. P. 415-418.

32. Zilov VG, Khadartsev AA, Ilyashenko LK, Eskov VV, Minenko IA. Experimental analysis of the chaotic dynamics of muscle biopotentials under various static loads. Bulletin of experimental biology and medicine. 2018;165(4):415-8.

Библиографическая ссылка:

Инюшкин А.Н., Филатов М.А., Григорьева С.В., Булатов И.Б. Нейросети мозга и их моделирование с помощью нейроэмуляторов // Вестник новых медицинских технологий. 2018. №4. С. 304-314.

i Надоели баннеры? Вы всегда можете отключить рекламу.