WO2008089741A1 - Device for reducing the synchronization of neuronal brain activity and suitable coil for said device - Google Patents

Abstract

The invention relates to a method for reducing the desynchronization of neuronal brain activity and to a device and coil suitable for said method. The device comprises a sensor for measuring a physiological feature and at least one magnetic coil (2) which administers magnetic pulses to the brain, which either reduce or completely suppress a pathological brain activity.

Description

 Device for reducing the synchronization of neuronal brain activity and suitable coil

The invention relates to a device for desynchronization of neuronal brain activity and a suitable coil.

In patients with neurological or psychiatric disorders such as Parkinson's disease, essential tremor, dystonia or obsessive-compulsive disorder, neuronal bandages are present in circumscribed areas of the brain, e.g. of the thalamus and the basal ganglia, pathologically active, for example, exaggerated synchronously. In this case, a large number of neurons synchronously action potentials, that is, the participating neurons fire excessively synchronous. In a healthy patient, the neurons in these brain areas fire qualitatively differently, for example in an uncorrelated way.

In Parkinson's disease, the pathologically synchronous activity alters neuronal activity in areas of the cerebral cortex, such as in the primary motor cortex, for example, impelling it to rhythm, so that eventually the muscles controlled by these areas exhibit phatic activity, e.g. a rhythmic trembling, unfold.

In patients who can no longer be treated with medication, a deep electrode is implanted on one or both sides, depending on the clinical picture and whether the disease is unilateral or bilateral. Under the skin, a cable leads from the head to the so-called generator, which comprises a control unit with a battery, and is implanted, for example, in the region of the collarbone under the skin. About the depth electrodes is a continuous stimulation with a high-frequency periodic sequence (pulse train with a frequency of> 100 Hz) of individual pulses, z. B. rectangular pulses performed. The aim of this method is to suppress the firing of neurons in the target areas. The mechanism of action on which the standard depth stimulation is based is not yet sufficiently clarified. The results of several studies suggest that standard depth stimulation works like a reversible lesion, ie, a reversible tissue rejection: the standard depth stimulation suppresses the firing of neurons in the target areas and / or associated hirsut - real.

The disadvantage of this form of stimulation is that the energy consumption of the generator is very high, so that the generator including the battery often has to be replaced operationally after about one to three years. Even more disadvantageous is that the high-frequency duration stimulation as unphysiologischer (unnatural) input in the brain, z. As the thalamus or the basal ganglia, in the course of a few years to adapt the affected nerve cell associations can lead. In order to achieve the same stimulation success, then must be stimulated as a result of this adaptation with higher stimulus amplitude. The greater the stimulus amplitude, the greater is the likelihood that side effects such as dysarthria (dysfunctions), dysesthesia (sometimes very painful discomfort), cerebellar ataxia (inability to be safe without help from others), or dysarthria due to irritation of neighboring areas Schizophrenia-like symptoms etc. - comes. These side effects can not be tolerated by the patient. The treatment therefore loses its effectiveness in these cases after a few years.

Therefore, another method has been proposed, as described in DE 102 11 766.7 "Device for the treatment of patients by brain stimulation, an electronic component and the use of the device and the electronic component in The purpose of this method / device is not simply to suppress morbidly synchronous firing, as in standard deep stimulation but to bring it closer to the physiological, uncorrelated firing pattern, which on the one hand reduces current consumption and, on the other hand, prevents adaptation processes of the nervous tissue, which can lead to side effects by increasing the stimulation amplitude.

The German patent application 103 18 071 discloses a method and an apparatus in which a demand-controlled application of stimuli takes place in the respective target area. The aim of these methods and apparatus is not simply to suppress the morbid synchronous firing, but to bring it closer to the physiological, uncorrelated firing pattern. As a result, on the one hand, the power consumption is reduced and on the other hand, adaptation processes of the nervous tissue that require an increase in the stimulation amplitude are prevented.

The main disadvantage of the mentioned stimulation methods is that they require an operative placement of the stimulation electrodes in the brain and that in the case of the standard depth stimulation - due to the power consumption - the generator is inclusive

Battery often have to be replaced surgically after one to three years. The short time available for calibration during an operation prevents the calculation of frequency-dependent phase resetting maps that can effectively desynchronize pathological synchronization processes as the main frequency component changes. The use of these desynchronostomy methods takes place only in the late stages of the disease, that is to say that here the healing aspect of the disease Stimulation, for example, because of the advanced cell degeneration in the target area can not be fully exploited.

From WO 03/85546 A1 a method and a device are known with which the brain activity can be influenced. The method and the device serve to achieve a predetermined behavioral goal. The device described there uses electromagnetic fields using various algorithms, but it is not possible to derive from these algorithms an explicit therapeutic effect for the reduction of the synchronization of neuronal activity. Furthermore, this text does not include the therapeutically relevant component of modifying coupling topologies, nor does it provide an algorithm for reducing the synchronization of neuronal populations.

From WO 03/098268 Al a magnetic field stimulation is known, which is used for diagnostic purposes. WO 03/082405 shows a method for increasing the performance. A method of suppressing brain activity is known from WO 00/74777 and WO 01/012236 A1. The document US 03/082507 Al discloses methods for the treatment of psychiatric and developmental diseases, such as stuttering or depression, using different stimulation protocols, eg single pulse, paired pulse, triple stimulation and low or high frequency repetitive stimulation, by means of one or more Stimulation coils are applied. There are two main forms of coils known, namely the simple O-shape and the shape of an eight 8. The so far known stimulation magnetic coils can stimulate the brain in most directions in two directions and thus stimulate at most two target populations or subpopulations. It is the object of the invention to provide a method and a device with which patients can be treated without the need for elaborate, high-risk and cost-intensive brain surgery. Preferably, it is intended to be able to treat patients who have a permanent or transient oscillatory component but are unable to perform surgery, or who have not had the suffering of the patients so far as to be aware of the risks and costs of brain surgery Take, that is, an operation should be superfluous. It should also be possible to treat patients in whom complex and prolonged calibration procedures must be used or in which the main frequency component of the pathological rhythmic activity may vary, such that the physiologically relevant coupling topologies are of physiological relevance through prolonged desynchronization internally or from other brain regions decoupled, that is, the diseased brain areas are brought to the healthy state. The aim is to create the possibility of frequency-dependent phase resetting maps, with the help of which the pathological synchronization processes with changing main frequency component can be effectively reduced. The term reduced or reduced also includes the complete elimination of the synchronization. It should be possible to treat patients in a previous stage of the disease. Not only should the decoupling of the affected neurons be decoupled, but a physiological change of the neurons is to be brought about, for example a recloning or a dephosphorylation of the ion channels. It should be created a coil with which a stimulation of more than in two directions is possible. Starting from the preamble of claim 1, the object is achieved with the features specified in the characterizing part of claim 1.

With the method and the device according to the invention, it is now possible to treat patients without the need for elaborate, high-risk and cost-intensive brain surgery. Preferably, patients may be treated who are based on a permanent or transient oscillatory component but are unable to perform an operation, or who have not been so severely distressed that they are at the mercy of the risks and costs of brain surgery take. Patients may also be treated for which complex and prolonged calibration procedures must be used, or where the major frequency component of pathological rhythmic activity may vary, such that the physiologically relevant coupling topologies of physiological relevance may be internally reduced by prolonged synchronization or decoupled from other brain regions, that is, the diseased brain areas are brought to the healthy state. It provides the ability to create frequency-dependent phase resetting maps that can effectively reduce pathological synchronization processes as the main frequency component changes. It is according to the invention a treatment of patients in an earlier stage of the disease possible. A physiological change of the neurons is brought about, for example a recloning or a dephosphorylation of the ion channels. Brain surgery can be dispensed with and the coil of the invention can be stimulated in more than two directions. For example, a stimulation in four directions is possible. Diseases can be successfully treated which have a pathologically synchronous oscillatory component is underlying. By way of example, but not limitation, epilepsy, Parkinson's disease, depression, obsessive-compulsive disorder may be mentioned. Tumor patients with synchronous high-amplitude neuronal populations near the tumor can be treated.

Advantageous developments of the invention are specified in the subclaims.

The method features specified in the description are embodied in the device according to the invention by a controller which is designed such that it controls the device for executing the method features. Each specified method step thus also discloses a device having a controller that carries out the method. In this way, disclosed process features are also to be construed as device features.

The figures show exemplary embodiments of the device and the coil according to the invention.

It shows:

Fig.l: A device according to the invention

Fig. 2a-e .: coils

Fig.3: Representation of excited subpopulations

Fig.4a: Stochastic multi-channel excitation by means of a pair of coils wound in a figure-of-eight form Fig.4b: Stochastic multi-channel excitation with a coil wound in a figure-eight.

FIG. 5a: Multi-channel excitation, in which the susceptibility of the pulses is reduced by means of a pair of eight-coil coils.

Fig.5b: Multi-channel excitation, in which the randomness of the pulses is reduced by means of an eight-coil.

Fig.6: Multi-channel excitation, in which the randomness of the pulses is reduced by means of a coil in figure-of-eight.

FIG. 7: Multi-channel excitation, in which the randomness of the pulses is reduced by means of a simple coil

Fog .8: Simulation results for the stimulation of a synchronous neuron population using a quadruple coil and stochastic multichannel excitation.

Fig. 9: Temporal evolution of the coupling among the neurons.

The device according to FIG. 1 comprises an isolating amplifier 1, to which at least one magnetic coil 2 and sensors 3 for detecting physiological measuring signals are connected. The isolation amplifier is further connected to a signal processing and control unit 4 in communication with a stimulation unit 8 for signal generation stands. The stimulation unit 8 for the signal generation is connected to at least one magnetic coil 2 in connection. At the entrance of the sensors 3 in the isolation amplifier 1 is a relay 9 or transistor. The relay 9 is controlled by the stimulation unit. The unit 4 is connected via a line 10 to a telemetry transmitter 11 which communicates with a telemetry receiver 12, which in this example is outside the appliance and to which a means for visualization, transmission and storage of the data 13 is connected. For example, epicortical electrodes, depth electrodes, brain electrodes or peripheral electrodes can be used as sensors 3. The magnetic coils 2 can be fastened to the head of the patient.

2a shows a simple magnetic coil, which is known in the prior art.

2b, c: magnetic double coils which either correspond to two intersecting circles or which are composed of two substantially elliptical forms.

FIG. 2d shows a four-coil, which is composed of four individual coils whose centers mark a square.

FIG. 2e shows a four-coil, which is composed of four individual coils whose centers mark a parallelogram. The angle between the two axes, which run through the centers of the diagonally lying coil parts is preferably 80 ° 70 °.

FIG. 3

An exemplary representation of the four excited neural subpopulations (4-7) of a brain substrate (2) by means of a Four-coil (1). The remaining parts of the brain substrate are not excited (8). Overall, the quad coil can activate up to four separate subpopulations independently. The dashed lines show the induced magnetic field (3).

FIG. 4a

Stochastic multi-channel excitation by means of a four-coil

(here a pair of coils wound in a figure eight shape). The time course of the applied stimuli (3) on the first coil (1), and the second coil (2). The orientation (4) of the current flow in the coils is marked with the positive and negative values + and - of the pulses. The sequence of pulses (3) is stochastically distributed in time and the pulses (3) must not overlap.

FIG. 4b

Stochastic multi-channel excitation by means of a double coil

(here a spool wound in a figure eight). The timewide course of the applied stimuli (2) via the coil (1). The orientation (3) of the current flow in the coils is marked with the positive and negative values + and - of the pulses. The follow of pulses (2) is distributed cally in time stocha ^¬ and the pulses (3) must not be pen overlap-.

FIG. 5a

Reduced stochastic multi-channel excitation by means of a four-coil (here a pair of coils wound in a figure-eight shape). The time course of the applied stimuli (3) on the first coil (1), and the second coil (2). The orientation (4) of the current flow in the coils is marked with the positive and negative values + and - of the pulses. The sequence of pulses (3) is ordered in time, ie the sequence of pulses is fixed, eg the positive pulse of the pulse The first channel is followed by the positive pulse of the second channel, which in turn is followed by a negative pulse of the first channel, and so on. The time interval of the individual pulses to each other but has a stochastic characteristic. The pulses (3) must not overlap.

FIG. 5b

Reduced stochastic multi-channel excitation by means of a double coil (in this case a coil wound in a figure-eight shape). The temporal course of the applied stimuli (2) over the

Coil (1). The orientation (3) of the current flow in the coils is marked with the positive and negative values + and - of the pulses. The sequence of pulses (2) is ordered in time, i. the sequence of pulses is fixed, e.g. the positive pulse of the first channel is followed by the positive pulse of the second channel, which in turn is followed by a negative pulse of the first channel, etc. The time interval of the individual pulses to each other but has a stochastic characteristic. The pulses (3) must not overlap.

FIG. 6

A schematic representation of the currents (2), the magnetic fields (3) and the induced currents in the brain (5) during a simple excitation by means of a coil in figure-of-eight (1). The maximum of the induced currents has a focal character (5).

FIG. 7

A schematic representation of the currents (2), the magnetic fields (3) and the induced currents in the brain (5) during a simple excitation by means of a simple coil (1). The maximum of the induced currents has an annular character (5).

FIG. 8 The firing density (1), the extent of synchronization (2) and the mean synaptic coupling (3) of a neural network (10x10 neurons) before, during and after a stimulation with a reduced stochastic multichannel excitation. The beginning of the stimulation is indicated by the arrow (4) and the stochastic pulse sequences are marked with the black lines in the lower part of the graphs (5).

FIG. 9 The temporal development of the coupling matrix of the network from FIG. 8 before, during and after a stimulation with a reduced stochastic multichannel excitation. For each point in time, the coupling matrix is represented as a gray value column in the figure. The strength of the coupling is coded with a gray value. The beginning of the stimulation is indicated by the arrow (1) and the end of the stimulation is marked by the arrow (2).

In the morbid state before stimulation, the neurons fire in common mode and have a very strong coupling with each other. After switching on the stimulation, the couplings between the neurons are reduced. After prolonged stimulation (over 600 cycles in the simulation), stimulation is no longer needed because the network remains permanently in the desynchronous state and the abnormally strong linkages between the neurons have been diminished.

The magnet coil 2 can basically be present in various embodiments. The term "magnetic coil" is intended to include in the content any body which is capable of inducing a magnetic field which is suitable for the establishment of an electric field which leads to the selective activation of the neuron population. According to the invention, the device comprises at least one magnetic coil 2. However, it may also comprise 2, 3, 4 or more magnetic coils 2.

Possible coil shapes are, for example: a) a simple magnetic coil (corresponding to FIG. 2a), b) an arrangement in front of two coils which are at a small distance from each other so that they either touch or nearly touch each other. The distance can be, for example, 1 to 5 mm. This arrangement is referred to below as a double coil. The coils can also touch, provided they are insulated. (Corresponds to Fig. 2b and 2c) bb) a coil whose wires are wound in a figure-of-eight. This coil shape is referred to as a double coil. (Corresponds to Fig. 2b and 2c) c) an arrangement of three coils whose centers preferably form the vertices of a triangle. d) in a particularly preferred embodiment, four coils (= four-coil), corresponding to a pair of coils wound in a figure eight, which are arranged substantially perpendicular, preferably perpendicular to each other. (Corresponds to Fig. 2d) dd) Four substantially annular coils (= four coil), the centers of which essentially form the vertices of a quadrilateral, preferably a square (corresponds to Fig. 2d and 2e).

The coils should have an electrical insulation, or there should be no short circuits. For the purposes of the invention, a four-coil of individual coils and / or eight-shaped coils can be composed.

The magnetic coils are examples of means for generating a magnetic field. The coils 2 are supplied via the controller 4 with electricity, which allow the construction of magnetic fields of, for example, 1.1 to 2.3 Tesla. The optimal magnetic field strengths can be determined by calibration. Basically, any magnetic field strength is suitable, which still has a physiological effect. Exemplary, lower magnetic field strengths are 0.1 Tesla or 0.01 Tesla. The information is not limiting.

Preferably, the coils have an inner diameter of 1 cm to 5 cm.

The outer diameter can be up to 8 cm, for example.

The signal processing and control unit 4 can comprise means for univariate and / or bivariate and / or multivariate data processing, as described, for example, in "Detection of n: m Phase Locking from Noisy Data: Application to Magnetoencephalography" by P. Tass, et.al. described in Physical Review Letters, 81.3291 (1998). the univaria ^¬ te and / or bivariate and / or multivariate data processing is preferably derived from statistical physics and preferably from the range of the stochastic phase reset tings.

According to the invention, the device is equipped with means which detect the signals of the sensors 3 as pathological and in the case of the presence of a pathological pattern on the magnetic coils 2 emit stimuli that cause the pathological neuronal activity either suppressed short-term or modified so that they natural, physiological activity comes closer. The pathological activity differs from the healthy activity by a characteristic change of its pattern and / or its amplitude. The means for detecting the pathological pattern are a computer which processes the measured signals of the sensor 3 and compares them with data stored in the computer. The computer has a data carrier, which stores data that was determined during a calibration procedure. By way of example, these data can be determined by systematically varying the stimulation parameters in a series of test stimuli and determining the success of the stimulation via the sensor 3 by means of the control unit 4. The determination can be performed by uni-, bi-, and multivariate data analysis to characterize the frequency characteristics and the interaction (eg coherence, phase synchronization, directionality and stimulus-response relationship), as for example in PA Tass: "Phase Resetting in Medicine and Biology , Stochastic Modeling and Data Analysis. "Springer Verlag, Berlin 1999.

The device according to the invention comprises means for establishing frequency-dependent phase resetting maps (fPRK). In the morbid state, the neurons fire very synchronously. If one measures the activity of this neural population, eg with the derivation of a local field potential, strong oscillatory components corresponding to the synchronous firing of the neurons can be detected in the measured signals. In practice, it is often observed that the frequency of the oscillatory activity is subject to frequent fluctuations. It can therefore not be assumed that during the oscillations in different frequencies the responses of the neurons to a stimulation remain constant. Therefore, one would not expect the vulnerable phase, in which stimulation of the network leads to its desynchronization, to be identical for all frequencies. The purpose of the frequency dependent phase resetting maps is to resolve the vulnerable phase as a function of the instantaneous NEN frequency of the oscillatory activity of the pathological

Area to determine. This makes it possible to perform stimulation in the vulnerable phase even for non-stationary oscillatory processes. Since the determination of the fPRK requires complex measurements, these can only be determined with the non-invasive stimulation method, magnetic field stimulation. Thus, the device according to the invention has a significant advantage over the devices (the PT double pulse and soft-phase resetting.)

For the determination of the fPRK a long sequence of single stimuli with constant parameters at different times is applied. In the data analysis, for each stimulus, its initial phase and the instantaneous frequency of the oscillation are determined, e.g. using Hilbert transformation and filtering, as well as the course of the phase during and after the stimulus application (the calculation period). Then, for each instantaneous frequency of the oscillation, a phase resetting curve is constructed for each time t in the entire calculation period. A phase resetting curve assigns the phase at time t to each value of the initial phase at time t after stimulation. In the next step, the number of turns of the phase resetting curves is determined for each time t. The number of turns indicates how often the phase after stimulation passes through the value range between 0 and 2PI at time t, while the initial phase passes through this range of values once. This is equivalent to the mean gradient of the phase resetting curve. The determination of the vulnerable phase is now carried out with the determination of the change in the number of turns from 1 to 0. The vulnerable

Phase is now in the range of the maximum absolute gradient of the last phase resetting curve with the number of turns 1. If for some instantaneous frequencies this transition between 1 and 0 is not found, repeat the stimulus application with a higher stimulation amplitude.

The PRPRs are then stored in the device. During single-stimulus stimulation, the device calculates the phase and the instantaneous frequency and selects the optimal combination of parameters from the fPRK.

The device according to the invention therefore comprises a computer which contains a data carrier which carries the data of the clinical picture, compares it with the measured data and emits a stimulus signal to the magnetic coil 2 in the event of the occurrence of pathological activity, so that a stimulation of the brain tissue takes place. The data stored in the data carrier of the clinical picture can be either person-specific, determined by calibration optimal stimulation parameters or a data pattern, which has been determined from a patient collective and typically occurring occurring optimal stimulation parameters. The computer recognizes the pathological pattern and / or the pathological amplitude.

The stimulus species used for the treatment of the pathological findings are known to the person skilled in the art. For example, longer periodic sequences of single stimuli or more complex stimulus sequences may be used. Examples of these complex stimuli are on the one hand a stochastic multi-channel excitation.

During stochastic multichannel stimulation, pulses or pulse trains across different coils and in different directions are not temporally superimposed and randomized at the time of onset. The amplitude and frequency of the individual pulses in pulse trains can vary randomly (noise processes / random processes), the intensity of the variations being adapted to the pathological feature. Stimulation via a channel is understood as influencing the neuronal activity of a subpopulation in the brain. at Multi-channel excitation influences the activity of several subpopulations. The excited subpopulations can overlap. A stimulation channel is the stimulation of a subpopulation by means of the magnetic field. The stochastic multichannel stimulus causes the stimulated subpopulations to display temporally independent fire behavior that translates into uncorrelated firing of the individual populations. Surprisingly, one obtains a significantly reduced synchronization, even if the extent of the stochastic variation of the time intervals between the pulses in different channels is significantly reduced. In principle, it can be achieved by changing the current direction in the coil that different subpopulations are excited. This means that two subpopulations can be excited with a simple coil. This also applies to two coils or an eight-shaped coil. When using two simple coils or an aft-shaped coil, the focusing of the resulting electric field can additionally be improved. The use of three coils allows the excitation of 6 different subpopulations. When 4 individual coils are used, arranged to substantially overlap at one point, even 12 different subpopulations can be excited by changing the combination of coils or the direction of current flow. A four-coil coil composed of two 8-row coils makes it possible to excite 4 subpopulations.

Since the excitation of the neurons in the target population depends on the angle of the axons to the electric field, up to 4 subpopulations can be selectively excited with the double coil. The individual subpopulations differ from each other by the angle they form with the induced electric field.

As a result of the applied stimuli, the pathological activity in the case of the use of longer periodic sequences of

Individual stimuli are typically briefly suppressed and in the case of the more complex stimulus sequences typically brought close to the natural, non-pathological activity or completely aligned.

By way of example, the following types of irritation can be cited:

Longer periodic sequence of single stimuli:

During such a sequence, a short single stimulus (one turn on and off of the stimulation) is repeatedly applied. Preferably, the individual stimuli are applied in the vulnerable phase of the oscillations

Complex stimulus sequence:

A complex sequence of stimuli refers to a set of on and off processes of the simulation unit which are distributed only over several stimulation channels and are variable in number and parameters (such as duration, amplitude, shape).

Pulse train: A pulse train is a set of on and off operations of the simulation unit called distributed only over a stimulation channel and are variable in number and parameters.

The device according to the invention is preferably designed such that, in the case in which the sensor 3 detects a discontinuation of the pathological activity after the stimulation, the stimulation is interrupted. For this purpose, the computer determines whether the pathologically increased amplitude or the pathologically increased pronounced pattern is present. This is done by means of the data analysis realized by the electronics. Once these pathological features are detected again, the next stimulation begins in the same way. The switching on and off of the stimulation takes place either by a control unit or by two control units communicating with one another, which are combined in FIG. 1 as a control unit 4.

The control unit 4 may include, for example, a chip or other electronic device with comparable computing power.

The control unit 4 controls the magnetic coil 2 preferably in the following manner. Targeted stimuli are then passed on via the magnetic coils 2 to the target region in the brain via the stimulator unit 8.

Furthermore, the device according to the invention preferably communicates with means for visualizing and processing the signals and for data backup 13 via the telemetry receiver 12. The unit 13 may have the data analysis methods mentioned below.

Furthermore, the device according to the invention can be connected via the telemetry receiver 12 with an additional reference database in order, for example, to accelerate a calibration process.

FIGS. 2a-e show, by way of example, magnetic coils which are suitable for magnetic stimulation. These are:

2a shows a simple magnetic coil, which is known in the prior art.

2b, c: magnetic double coils, which either correspond to two intersecting circles or which are composed of two substantially elliptical forms. FIG. 2d shows a four-coil, which is composed of four individual coils whose centers mark a square.

FIG. 2e shows a four-coil, which is composed of four individual coils whose centers mark a parallelogram. The angle between the two axes, which run through the centers of the diagonally lying coil parts is preferably 80 ° 70 °.

The coils 2b, 2c, 2d and 2e are particularly effective. With them, a penetration depth of the magnetic field in the cortex up to 2 cm can be effected.

The coils 2d and 2e are particularly effective in reducing synchronicity and in dividing a synchronous neuron population into two or more uncorrelated subpopulations.

In the following, the invention will be described in its general form.

According to the invention, the magnetic coils 2 are used to induce magnetic fields which generate electric fields in the brain and thus cause the intracortical currents to apply stimulus patterns in the target areas of the brain.

The respective magnetic field is perpendicular to the current direction.

If a simple magnetic coil 2 according to the example of FIG. 2a) is used, it induces a simple magnetic field, which causes a circular electrical gradient in the brain tissue, which can be used for stimulus application. In the embodiment according to the examples Fig. 2b) and 2c) forms a very good focus of the magnetic fields below the point of the coils where they come closest, so that a very good stimulation is effected. The direction of the current in the area of the maximum spatial approximation of the coils must have the same direction, so that the currents and thus the magnetic field do not cancel each other out or reduce. Depending on the direction of the current flow, the magnetic field has an opposite orientation to one of two directions.

In the embodiments of the example FIGS. 2d) and 2e), the formation of current flow in the brain in four directions is possible. In each case two opposite directions are generated by two double coils, which preferably form the diagonal corner points of a quadrilateral. The formed by this coil shape temporally not overlapping trained electric fields preferably include an angle of 90 °. This can cause a special decoupling of the neural activity. In a less preferred embodiment, the arrangement of the coils differs from that of an ideal symmetry and / or there is overlap of the coils, so that an angle can be formed between the two electric fields which is different from 90 °. The angle can reach 75 °, for example. However, the best healing and treatment successes are achieved when an angle of 90 ° is formed. Good results are also achieved with a deviation of 90 ° + 5 °.

In principle, other coil shapes are possible.

With the electric fields generated by the magnetic coils or their temporal sequence and orientation, it is possible to generate stimuli in defined brain regions, which cause disunctivation and even a healing of the patients Effect regions. With the training electric

Fields can also apply electrical electromagnetic stimulus patterns, which have a definite temporal sequence and spatial extent and / or spatial orientation.

Various stimulus patterns can be applied.

The stimuli are individual pulses with a duration in the sub-millisecond range of, for example, 200-500μs.

With these single pulses, a low-frequency stimulation of a frequency of less than one Hz can be made. It is also possible to apply high-frequency stimulus sequences of a frequency of more than one Hz to, for example, 200 Hz.

With the method and the device according to the invention, a desynchronisation of neuronal activity is achieved technically differently than with the devices which make use of stimulating electrodes. Instead of placing a stimulation electrode in the pathologically synchronous neurocellular network, a magnetic coil is positioned over the area so that the maximum of the electrical gradient induced by the magnetic coil in the brain is in the target population. (= Induction of electric fields in the brain) can be done with different stimulation methods. These may be selected depending on the disease: "Single stimulus, compound stimuli, stochastic and deterministic multichannel stimulation." The device is not designed for a particular stimulation technique, but may vary the algorithms for reducing the synchronization as needed The change in coupling topologies within or between neuronal populations is essentially the result of an uncorrelated activity in the brain tissue and minimizes the stimulation Achieving the treatment goal, that is, no

Stimulation after healing is more necessary.

Example: Reduction of the synchronization by means of the stochastic multichannel excitation: If, for example, the stimulation is used by means of coordinated reset by means of a double coil 2d, the following stimuli can be delivered via two stimulation coils: A set of pulse trains is applied via each coil, the stimulation phases being should not overlap and two of the pulse trains are applied by means of the same coil, only with reverse polarity.

FIG. 8 shows by way of example the stochastic multi-channel excitation. In the morbid state before the stimulation, the neurons fire in common mode and have a very strong coupling with each other. This can be seen in the high fire density and the high degree of synchronization. After switching on the stimulation by means of the stochastic multichannel excitation (4), the synchronous activity in the network is significantly reduced. In addition, as a result of continuous desynchronization, the couplings between the

Neurons reduced. This leads to a weakening of the synchronization tendency in the network and a reduction in the frequency of the stimulation application. After prolonged stimulation (over 600 cycles in the simulation), stimulation is no longer needed because the network remains permanently in the desynchronous state, i. the network is healed.

In this case, a four-coil with α = 90 ° (Fig.2d) is used. The excitation of the four subpopulations happens randomly or randomized with a randomization factor RF = K + S * epsilon with (K = T / 4, S = O, 2) ie the individual subpopulations are separated with a time interval of A = T / 4 + S * Epsilon excited, where Epsilon is a normally distributed random process with the expected value 0 and the variance equal to 1, and T is the period of the rhythm to be desynchronized. After a pulse train with duration D is switched on at time ti, the turn-on point t _{2 of} the next pulse train t ₂ = ti + T / 4 + S * epsilon is calculated. If the turn-on time t _{2 is} less than (ti + D), ie the pulse trains would intersect, additional factors (T / 8 + S * epsilon) are added until t _{2 is} greater than (ti + D). The individual subpopulations are successively stimulated one after the other. During the stochastic multichannel excitation, the pathological synchronization is clearly suppressed (FIG. 8 a, b), that is, the firing of the neuronal population shows a clearly reduced amplitude (FIG. 8 a) and also a reduction of the synchronization indices (FIG. 8 b). If the stochastic multi-channel excitation is applied over a long period of time, it leads to a recovery / coupling reduction of the target area (FIG. 8, 9), the decoupling having a significantly slower dynamic than the synchronization behavior. The healthy target area no longer needs stimulation.

In an alternative embodiment of the stochastically reduced multichannel excitation, pulse trains are applied offset in time by T / 4, where T is the mean period of the rhythm to be desynchronized.

This stimulation method is a mild, non-invasive procedure that exploits the neurons in the affected areas to undergo plastic processes that are dependent on the intensity of firing synchronicity. High synchronicity leads to the strengthening of the coupling between neurons and / or neuron polulations and low synchronicity reduces the coupling between neurons and / or neuron populations.

Example: According to the invention, the pathological neuronal activity A) is measured via a sensor such as a) brain electrode, eg a depth electrode, b) epicortocal electrode or via c) a muscle electrode and serves as feedback signal, ie control signal, for demand-controlled stimulation B). The feedback signal from the sensor 3 is transmitted via a line to the buffer amplifier 1. Alternatively, the feedback signal can also be transmitted telemetrically without the use of an isolation amplifier. In the case of telemetric transmission, sensor 3 is connected to an amplifier via a cable. The amplifier 9 is connected to a telemetry transmitter 11 via a cable. In this case, sensor 3 and amplifier 9 and telemetry transmitter are implanted, for example, in the area of an affected limb, while the telemetry receiver is connected to the control unit 4 via a cable. This means that, unlike the standard continuous stimulus, the activity is measured and the measurement signal is used as the trigger for on-demand stimulation. For the measurement A) of the neuronal activity, there are the following different possibilities:

I. Measurement via the sensor 3, for example measurement of neuronal activity originating from the cerebral cortex, either via an implanted electrode b) or preferably an atraumatic epicortical electrode b) (sensor 3), i. an electrode which rests on the brain and is fixed but does not penetrate into the tissue, and thus a local electroencephalogram from an affected area of the cerebral cortex, e.g. the primary motor cortex.

II. In patients who primarily suffer from tremor, the measurement of muscular activity by electrodes c) (sensor 3, preferably tele- metric connected to control unit 4) in the area of the affected muscles.

The pathological neuronal activity can in principle also occur in different neuron populations. Because of this, several signals measured via sensors 3 can also be used to control the stimulation. Whenever a pathological feature of the activity is detected in at least one of the neuron polypulations, an irritation is triggered.

The measuring signal or the measuring signals serve or serve as feedback signals. This means that stimulation takes place as a function of the activity detected via the measuring signal. Whenever a pathological feature of neuronal activity (ie, pathologically increased amplitude or pathologically enhanced activity pattern) begins and increases, it is stimulated.

In this case, the stimulation B) can take place in various ways.

Demand-driven stimulation to reduce synchronization:

These procedures are used when pathologically synchronized nerve branch activity is present in the target area (eg in epilepsy disease in areas of the cortex) or in another area or muscle relevant to the disease (derived via sensors 3). This is determined, for example, by bandpass filtering the signals measured via sensors 3 in the frequency range that is characteristic of the pathological activity. As soon as a band-pass filtered measurement signal exceeds a threshold determined in the calibration procedure, it becomes via control unit 4, the next control pulse to the stimulator unit 8, which generates or passes on the magnetic coil 2 stimuli. The aim here is not simply to suppress the firing of the neurons, as in the standard continuous stimulation. Instead, only the morbidly increased synchronization of the Nerver cells should be remedied as needed. This means that the nerve cell groups in the target area are uncorrelated, whereby they continue to be active, ie form action potentials. This is to bring the affected nerve cells closer to their physiological - ie uncorrelated firing - state, rather than simply that their activity is completely suppressed. For this, several different synchronization reduction methods based on the principle of "stochastic phase resetting" can be used, taking advantage of the fact that a synchronized neuron population can be desynchronized by applying an electrical stimulus of the correct intensity and duration, provided , the stimulus is administered in a vulnerable phase of the abnormal rhythmic activity, these optimal stimulation parameters

(Intensity, duration and vulnerable phase) are determined in the calibration procedure, for example by systematic variation of these parameters and comparison with the stimulation success (eg attenuation of the amplitude of the bandpass filtered feedback signal). In case of using the telemetry device 11-13, the calibration can be improved by using the above-mentioned frequency-dependent phase resetting curves. Single-pulse stimulation is only efficient if the stimulus is applied at or close enough to the vulnerable phase of the activity to be stimulated. Alternatively, complex forms of stimulation can be used. These consisted of a resetting (controlling the dynamics of the neuron subpopulation to be stimulated, for example restarting). The advantage of these more complex procedures is that the complex forms of stimulation are independent of dynamic state of the neuron population to be stimulated cause a reduction in synchronization.

In the case of use of individual stimuli must control unit 4 when exceeding the threshold determined by the calibration lenwerts means by the electronics (control unit 4) rea ^¬ ized standard prediction algorithms predict the temporal occurrence of the vulnerable period to these precise enough to meet. In the case of the use of complex stimuli, control unit 4, when exceeding the threshold determined by the calibration, must only cause a new complex stimulus of the same kind.

Simple stimuli are, for example, a) single-pulse stimulations.

Complex stimuli are, for example, b) double pulse stimulation, c) stochastic multichannel stimulation

In a preferred embodiment, the apparatus is provided with means for wireless transmission of data, such as the measurement signals and pacing control signals, to allow data transfer from the patient to an external receiver, for example, for the purpose of therapy monitoring and optimization. In this way it can be recognized early on whether the stimulation parameters used are no longer optimal. In addition, by wireless transmission of data to a reference database and resorted early to typical Verände- ^■ stakes of irritability to respond in the target tissue.

Claims

claims An apparatus for reducing neuronal brain activity desynchronization comprising means for stimulating brain regions and a sensor for measuring a pathological feature, characterized in that the means for stimulating neuronal brain activity is a means for generating a magnetic field. 2. Device according to claim 1, characterized in that the means for generating a magnetic field comprises at least one magnetic coil (2). 3. A device according to claim 2, characterized in that the means for generating a magnetic field (2) is an arrangement of two or three coils. 4. Apparatus according to claim 2, characterized in that the means for generating a magnetic field (2) comprises four coils. 5. Apparatus according to claim 4, characterized in that the centers of the coils form substantially the corner points of a square. 6. Device according to one of claims 3 or 4, characterized in that the two or four coils are given by one or two guided in the form of an 8 coils. 7. Device according to one of claims 1 to 6, characterized in that the means for generating a magnetic field form a magnetic field strength of 1.1 to 2.3 Tesla. 8. Device according to one of claims 1 to 7, characterized in that it has a control unit (4) which applies magnetic fields. 9. Device according to claim 8, characterized in that the control unit (4) comprises a univariate and / or bivariate and / or multivariate data analysis. 10. The device according to claim 9, characterized in that at least one of the univariate, bivariate and multivariate data processing works with methods of statistical physics. 11. The device according to claim 10, characterized in that the method of statistical physics comes from the field of stochastic phase resetting. 12. The device according to one of claims 1 to 11, characterized in that the controller detects the occurrence of a pathological feature, which is measured by the sensor (3), and emits at least one pulse to the solenoid upon occurrence of the physiological feature. 13. Device according to one of claims 1 to 12, characterized in that the controller (4) detects the elimination of a pathological feature and turns off the magnetic pulse. 14. means for generating a magnetic field, characterized in that it comprises at least three coils, through which a Electricity is passed. 15. A magnetic field generating device according to claim 14, characterized in that it comprises four coils through which a current is passed. 16. means for displaying a magnetic field according to any one of claims 14 or 15, characterized in that it generates magnetic fields between 1.1 and 2.3 Tesla.