SU525128A1 - Device for simulating spindle-shaped bioelectric activity of the brain - Google Patents

Description

(54) DEVICE FOR MODELING BELIEF-SHAPED BIOELECTRIC ACTIVITY OF THE BRAIN

but

The neural network model 1 (Fig. 1) is represented as a model 8 of the distribution of indiscrete and discrete signals in the network conductors connected by outputs 9 with formal neurons 9. The distribution model of indiscremental signals is a modified model of the distribution of electrotonic and cerielectrotonic currents in the nerve conductors. the matrix of active resistances, containing seven horizontal and fourteen vertical columns (Fig. 2). Between the vertical columns 10 and 11, in the first and seventh rows, there are resistances that have conditional twenty, and in the fourth line, resistance, which has a conditional value of ten. The extreme vertical columns on the right and on the left have resistances conventionally equal to zero, to which, via clamps 12 and 13, the voltage East does not change over time. conventionally called polarization potential. The time-varying voltage Ivh coming from the generator 4 sinusoidal pulses is supplied to terminals 14 and 15. The clamp 14 also serves to sum up the positive source of the discrete signals Id coming from the generator 5 discrete pulses. From the intersection points of the 16th row of resistances with columns 17-24, generators are made for connection to the inputs 25-32 of formal neurons (with indiscrete and discrete control).

For a formal neuron (with indiscrete and discrete control), the 9 input for discrete signals has a variable threshold depending on the polarity and magnitude of the control (indiscrete) signal, coming to the same input from one of the outputs of the indiscrete and discrete signals. in the network conductors.

The model 6 of the bioelectrical activity of the brain (Fig. 1) consists of models of a formal neuron, which in the form of a series-connected waiting multivibrator and a two-link integrating circuit, which is connected directly and via a capacitor to a diode limiter connected to a separation diode, and the inputs of the waiting multivibrators of models the formal neuron is connected to the input of the device, and the separation diodes of the models of the formal neuron are connected through appropriate switches to the load element, connected to the output of the device.

The proposed device works in the following way.

As a result of the addition of currents arising from the two applied to the distribution model, the indiscrete and discrete signals in the conductors of the time-varying voltage network. comes from a 4 sinusoidal pulse generator and a constant polarization potential. East The location of the point of convergence of currents in the fourth horizontal row of the impedance matrix becomes dependent on the value of the voltage Ivh-S depending on the location of this point, on one or the other of terminals 25-32, connected) 1e with formal neurons, current flows

a different direction. This current is used as an indiscrete signal. Since the minus of the source of discrete signals Id is applied to all formal neurons, then submitted after the direct source of Id

will arrive on the same fourth line. Id voltage is significantly less Ivh. therefore, it does not affect the distribution of currents in the resistance matrix. An indiscrete signal, arriving at the input of a formal neuron 9, depending on its polarity, lowers or raises the threshold for the excitation of a formal neuron from the incoming discrete signal. A spiky triggering impulse from the output of a formal neuron 9 enters the inputs of the formal neurons of model 6 bioelectric activator

brain and causes their triggering.

The impulse issued by a formal neuron has the form of an impulse produced by living neurons. With the simultaneous emergence of pulses of a number of formal neurons connected in parallel by both the inputs and the outputs, the pulses are summed both by the aggregate and the duration, and the total wave arises. | retains the pulse shape of an individual neuron.

In a device for simulating spindle-shaped bioelectrical activity of the brain, the smallest voltage value of a variable Ivh is obtained indiscretely by preparation for the excitation of a formal neuron connected to point 25.

A higher value of the Mgx voltage is obtained in the same way by preparing for the excitation of already two neurons that are connected to points 25 and 26. The higher

0 voltage the more neurons are prepared for excitation and are put into operation. Thus, depending on the phase of the positive sinusoidal half-wave of the generator 4, which is used as a variable voltage, the formal neurons of unit 1 are connected in series to work and give impulses from discrete signals to the input of unit 1. For the same reason, the formal neurons 9 of block 1 bpc; i think of work in

O reverse sequence.

The pulsing of neurons in block 1 triggers a series of blocks for modeling the bioelectrical activity of the brain. As a result, the total wave bundles (Fig. 3) coming from the outputs of blocks 8 on

5, terminal 33, have a dispersion of the beginning and end of the packs relative to those of neighboring blocks 6. At terminal 33, the wave packs are added along a Schlitd and duration. Since at different times has the addition of different numbers of waves

due to the diversification of wave packs, the result is

terminal 33 we get a total flash having different amplitudes and frequencies in different parts of it - a flash of a spindle shape. The duration and shape of the spindle-shaped flashes is depending on the frequency of the sinusoidal voltage coming from the generator 4. The smaller the frequency, the longer the spindle-shaped power and the increase and decrease in the amplitudes of the waves at the beginning and at the end of the flash (Fig. 5, a, b) ). In order to smoothly follow the flash waves one after another, triggering discrete impulses are fed to input 3 at the end of each wave. This is done as follows. From generator 5 to input 3 of block 1, triggering pulses are applied until they coincide (in time) with a minimum amplitude of the positive half-wave of a sinusoidal voltage coming from 1; generator 4; triggers one formal neuro and 9 block 1, connected to point 25, and starts the corresponding block 6; the resulting total wave from terminal 33 arrives at the input of the neuron feedback 7, at the output of which we get a sharp-pointed pulse arising with a delay at the end of the succession of the total wave; the sharp triggering impulse from the output of neuron 7 is fed to input 3 of block 1 and, coinciding with the next phase of the positive half-wave of the sinusoidal voltage of generator 4, already triggers two formal neurons 9 of block 1 (neurons connected to dots 25 to 26 and corresponding two blocks 6; At terminal 33, a second total flash wave with a corresponding amplitude and duration is obtained. The preceding and subsequent waves, despite the different duration, smoothly transform into one another, the process repeats itself and repeatedly About the moment of transition of the positive half-wave of the sinusoidal voltage of the generator 4 to the negative half-wave, At this moment the last wave of the spindle flash passes through terminal 33. Two start modes are possible. The first mode. 5. The spindle itself thanks to the feedback through the neurop 7 flows spontaneously, i.e. not

a triggering pulse is required for each spindle wave. The form of the beginning of the flash and its duration depend on the moment of coincidence in time of the external trigger pulse coming from generator 5 from one of the phases of the positive half-wave of the sinusoidal voltage of the generator 4. For example, if the trigger pulse coincides with the phase of the 90 ° sinusoidal half-wave, there is no gradual increase the flashes (Fig. 5c), since all blocks 6 (Fig. 1) are started at once without dispersion relative to each other.

Second mode. Spontaneous repetition of the spindle without the impulse of a superfluous trigger from generator 5 is accomplished by adjusting the sensitivity to the discrete input signal of one of the formal neurons 9 (Fig. 1) of one of the blocks 6. In this case, only one external impulse from the generator 5 circulates undamped pulses along the path 1-6 (a block with the increased sensitivity of one of the input neurons) 7-1, etc. At the moment of supplying the positive half-wave of the sinusoidal voltage from the generator 4, the circulating pulses serve as triggering and at the same time they themselves participate in the summation of the waves to form spindle-shaped flashes.

Claims (1)

Invention Formula A device for simulating spindle-shaped bioelectrical activity of the brain, comprising a discrete pulse generator and a sinusoidal pulse generator, characterized in that, in order to increase the simulation accuracy, it contains a model of a neural network, to which the outputs of the generators are connected. sinusoidal and discrete pulses; models of brain bioelectric activity, whose inputs are connected to the outputs of a neural network model; and a feedback neuron model, the output of which is connected to the output of a discrete pulse generator, and the input is connected to the outputs of brain bioelectric activity models. 10 11 -sch, n IB 20 21 22 Rig.2 23 2C L J 3.3 CL 1L t 15 / 45 and 17, c. 15 TsCh-18 - H