APPARATUS AND METHOD FOR TREATING NEUROLOGICAL DISORDERS

Abstract

The present invention concerns an apparatus for treating neurological disorders comprising (i) at least one electrode implantable in the brain of a patient and (ii) a processing and stimulation device connected to the at least one electrode, wherein the processing and stimulation device comprises (a) at least one stimulation module adapted to generate a stimulation signal to be sent to the at least one electrode, the stimulation signal being characterised by a plurality of parameters, (b) at least one acquisition module of a signal characteristic of cerebral activity coming from the brain of the patient adapted to determine its power in at least one frequency band, and (c) at least one control module of at least one parameter of the stimulation signal as a function of the power of the signal characteristic of cerebral activity acquired, based on a transfer function having a saturating trend, wherein the transfer function is such as to set the at least one parameter (Va, Vd, Vf) of the stimulation signal (Vstim) differently dependent on a plurality of power ranges, by keeping the at least one parameter within a predetermined stimulation range ([Vi_HighThreshold; Vi_LowThreshold]) with i=a,d,f.

Description

The present invention concerns an apparatus and a method for treating neurological disorders. In greater detail, the present invention concerns an apparatus and a method for treating neurological disorders based on the feedback of deep brain stimulation, therefore capable of detecting biopotentials from a stimulation electrode or from neighbouring electrodes, correlating such signals with the clinical state of the patient and feeding the stimulation parameters back in order to optimise the therapy.

In the present description and in the subsequent claims, the expression “local biopotentials” means the electric potentials produced by groups of neurones close to the recording and stimulation electrodes.

In the present description and in the subsequent claims, the expression “local field potentials” means the sum of the pre and post synaptic activity around a stimulation electrode implanted in the brain.

In the present description and in the subsequent claims, the expression “base clinical state” of a patient means the state of the patient in the absence of therapies, both pharmacological and by stimulation, for a sufficiently long time period, so that the effects of the therapies have totally worn off, in general equal to at least 12 hours.

As known, deep brain stimulation, also known by the acronym DBS, is a therapeutic method based on the modulation of the activities of structures of the central nervous system through an electric stimulation delivered locally. For this purpose, electrodes are implanted by neurosurgery. Electric stimulation consists of delivering a train of electric pulses in the brain area of interest through the implanted electrode, which for this purpose is connected to a pulse generator.

Deep brain stimulation is currently used in clinical trials or for the treatment of epilepsy, migraines, some psychiatric disturbances, pain and movement disorders, such as dystonia, tremor and Parkinson's disease. Such a method allows the functional autonomy of patients to be improved, thus offering an improved quality of life. Since the use of deep brain stimulation is particularly consolidated in the treatment of Parkinson's disease, in the present description reference will mainly be made, although only as an example and not for limiting purposes, to such a disease.

In the case of Parkinson's disease, the implantation areas of the electrode most used in clinical practice are some structures of the grey nucleus, including in particular the internal globus pallidus and the subthalamic nucleus, as well as the inner peduncle pontine nucleus. The choice of the implantation area is at the discretion of the neurologist based on the symptoms to be tackled. As an example, it should be considered, indeed, that Parkinson's disease is characterised by a wide spectrum of motor and non-motor symptoms, where the motor symptoms comprise, amongst other things, rigidity, bradykinesia, rest tremor, akinesia, postural instability, inability to walk and so on. Clinical studies widely support the benefits brought by deep brain stimulation at the subthalamic nucleus, which is therefore the preferred implantation area in most treatments. Such a treatment method can cause various side effects including dyskinesia, language and psychiatric difficulties that can, nevertheless, be mitigated through a reprogramming of the stimulation parameters.

In the case of conventional brain stimulation, the parameter setting process is carried out empirically, requiring time, clinical experience and numerous visits to adjust to the optimal values. Moreover, such adjustment leads to setting that, even if initially optimal, is unable to adapt to variations that came afterwards. Indeed, once the parameters have been set, the stimulation is delivered independently from the clinical state of the patient, not being able to adapt to possible variations of such a clinical state. Therefore, clinical fluctuations characterising an advanced stage of Parkinson's disease remain uncontrolled.

In order to overcome the limitations described above, an apparatus for treating neurological disorders has been made that is capable of monitoring the clinical state of the patient by recording a physiological marker, also known as biomarker, as control variable, and of adapting the stimulation parameters to the clinical state detected. Such a system is known as adaptive brain stimulation. An example of such an apparatus is described in European patent EP 1 940 508. Through such an apparatus it is possible to decrease the time necessary to set the optimal parameters and decrease the occurrence of side effects.

From EP 1 940 508 it is known to use the local field potentials as a marker based on which to adapt the stimulation parameters. The local field potentials prove particularly suitable for the purpose, since they correlate with both motor and non-motor symptoms, they can be recorded by the stimulation electrode even during the stimulation itself, they can be influenced through stimulation and maintain such properties for a long time after the electrode has been implanted. Neurophysiological recordings carried out through an electrode for deep brain stimulation have, indeed, shown that the local field potentials are correlated to the clinical state of the patient, for example to the performance of movements, as well as to cognitive and behavioural stimuli.

In relation to the correlation between the local field potentials and the symptoms of the disorders to be treated, the Applicant has observed that a lot of the information content of the local field potentials relative to the disorders to be treated can be extrapolated through frequency analysis. In other words, the Applicant has found that there is a link between the power of the local field potentials in certain frequency bands and specific symptoms of the disturbance to be treated. Specifically, the Applicant has identified a relationship between sets of power values of the local field potentials and the variations in the clinical state in terms of symptoms to be treated. In particular, the significant oscillatory rhythms in terms of their correlation with the symptomatic state of the patient are: low frequencies (4-10 Hz), the beta band (alpha/low beta: 11-20 Hz, high beta: 20-35 Hz), the gamma band (60-80 Hz) and high frequencies (250-350 Hz).

In line with such an observation, patent EP 2 004 036 also describes a medical device that, for the treatment of a motor disorder, uses a physiological marker of the clinical state of a patient. In particular, the physiological marker is calculated based on two values representative of the amount of oscillatory activity of the physiological signal detected, each relative to a different range of frequencies. Patent EP 2 004 036 also refers to the possibility of adjusting the therapy in feedback based on the two values representative of the amount of oscillatory activity, in particular by adjusting the stimulation parameters so as to keep the biomarker calculated based on such two values, in a predetermined range. EP 2 004 036 does not, however, provide any indication concerning the adjustment logic for an optimal adaptation of the stimulation parameters.

Also known is the study of Santaniello et al. Described in the article titled “Closed-Loop Control of Deep Brain Stimulation: A Simulation Study” published in IEEE Transactions on neural systems and rehabilitation engineering, Vol. 19, No. 1, February 2011, pages 15-24, which has the purpose of developing a closed-loop control system for automatically adjusting the stimulation parameters based on the neuronal signals recorded through the same stimulation electrode, in particular based on the local field potentials. Such a control system comprises, downstream of the transfer function, a saturating trend function that limits the amplitude of the stimulation parameters if the transfer function determines an output value outside a predetermined range. Such a limitation is therefore carried out as a function of the output of the transfer function.

However, such a system is unable to correlate the power of the local field potentials calculated in specific frequency bands with the stimulation parameters, so as to take into account the link observed by the Applicant between the particular sets of band power values and the specific symptoms of the disturbance to be treated. In other words, the transfer function of the control system described in Santaniello applies the same law of variability independently from the set to which the power value that it receives in input belongs.

In light of the above, the Applicant has found the need to devise an apparatus and a method capable of administering a deep brain stimulation therapy in an optimised manner as a function of the instantaneous clinical state of the patient, expressed through a physiological marker calculated from the local field potentials measured on the patient.

The problem forming the basis of the present invention is therefore that of making an apparatus and a method for treating neurological disorders that is able to adapt the stimulation administered in an optimal manner as a function of the instantaneous clinical state of the patient, operating automatically and continuously.

For this purpose, the Applicant has perceived the need to identify a law of variability that correlates the stimulation parameters with the local field potentials, based on which it is possible to construct an automatic control of the therapy that adapts in an optimised and quick manner to possible changes in the clinical state of the patient.

In accordance with a first aspect thereof, the invention concerns an apparatus for treating neurological disorders comprising

• • at least one electrode implantable in the brain of a patient and
  • a processing and stimulation device connected to the at least one electrode, wherein the processing and stimulation device comprises
  • at least one stimulation module adapted to generate a stimulation signal to be sent to the at least one electrode, the stimulation signal being characterised by a plurality of parameters,
  • at least one acquisition module of a signal characteristic of cerebral activity coming from the brain of the patient, adapted to determine its power in at least one frequency band, and
  • at least one control module of at least one parameter of the stimulation signal as a function of the power of the acquired signal characteristic of cerebral activity based on a transfer function having a saturating trend, wherein the transfer function is such as to set said at least one parameter (V_a, V_d, V_f) of the stimulation signal (V_stim) differently dependent on a plurality of power ranges, keeping it within a predetermined stimulation range ([V_{i_HighThreshold}; V_{i_LowThreshold}]) with i=a,d,f. In the present description and in the subsequent claims, the expression “transfer function having a saturating trend” means a function that for input values greater than or less than respective first and second predetermined input values respectively places the output variable (different from the input variable) equal to a first and a second predetermined output value and for input values comprised between the first and the second predetermined input values, the output can take values based on a predetermined law of variability. Therefore, in the present description and in the subsequent claims, the definition of “saturating trend function” should not be taken in the narrow sense as a function that forces the input variable to not take values in output greater or less than predetermined thresholds.

The Applicant has observed that, for example, the oscillations in the beta band are suppressed following administration of a suitable drug (for example dopamine) and correlate with the preparation and performance of the movement. Such oscillations reflect the response of the grey nucleus to the drug and correlate with the motor state. In the same way, it has been demonstrated that a deep brain stimulation treatment induces a power reduction in the beta band.

In other words, when an improvement of the motor state of the patient is observed after the administration of the drug or after a deep brain stimulation therapy, a corresponding percentage reduction of the power of the local field potentials calculated in the beta band with respect to the base clinical state, i.e. the clinical state prior to the administration of therapies, is observed. The variation of the motor state of the patient is expressed in UPDRS (Unified Parkinson's Disease rating scale) as percentage values with respect to the base state.

Similarly, the Applicant has observed that the oscillations in the band of low frequencies (4-10 Hz) increases with a deep brain stimulation and/or pharmacological treatment. In other words, when a percentage decrease in the UPDRS scale is observed, i.e. an improvement in the clinical state of the patient after administration of the drug or the deep stimulation therapy, a corresponding increase in the power of the local field potentials calculated in the band of low frequencies with respect to the base clinical state is observed.

The Applicant therefore started from the assumption that an improvement in the clinical state of the patient is correlated to a percentage variation of the power of the local field potentials, and hypothesised a proportional relationship between such a power calculated in a certain frequency band, whether it is calculated in the frequency domain, or in the time domain, and the clinical state, for example expressed in UPDRS scale.

The Applicant has also hypothesised that the relationship between stimulation and local field potentials can be expressed through a proportional relationship between the stimulation signal and the power of the local field potentials. From such an assumption it follows that the clinical state can be expressed as a function of the stimulation signal.

In accordance with the assumptions outlined above it is possible to control the clinical state of the patient by modifying the stimulation parameters as a function of the power in a certain frequency band according to the mathematical function indicated above.

Such a function determines that, for certain ranges of input power values, the stimulation parameters can vary as a function of the input power, adapting to the instantaneous clinical state of the patient and for other ranges of input power values the stimulation parameters are set equal to respective saturation values at which the stimulation was still actually effective. In this way, the occurrence of side effects due to stimulation therapy is avoided, at the same time ensuring a definite benefit induced by the stimulation.

Similarly, in accordance with a second aspect thereof, the invention concerns a method for treating neurological disorders comprising the steps consisting of:

• • sending at least one stimulation signal characterised by a plurality of parameters to at least one electrode implantable in the brain of a patient;
  • acquiring at least one signal characteristic of cerebral activity coming from the brain of the patient and determining its power in at least one frequency band; and
  • adjusting at least one parameter of the stimulation signal as a function of the power of the acquired signal characteristic of cerebral activity based on a transfer function having a saturating trend, wherein the transfer function is such as to set said at least one parameter (V_a, V_d, V_f) of the stimulation signal (V_stim) differently dependent on a plurality of power ranges, keeping it within a predetermined stimulation range ([V_{i_HighThreshold}; V_{i_LowThreshold}]) with i=a,d,f.

Advantageously, the method for treating neurological disorders according to the invention achieves the technical effects described above in relation to the apparatus.

The present invention, in at least one of the aforementioned aspects, can have at least one of the following preferred characteristics, which are in particular able to be combined with each other as desired in order to satisfy specific application requirements.

Preferably, the transfer function having a saturating trend as a function of a plurality of power ranges is a piecewise function, placing the stimulation parameter equal to a first value of the stimulation parameter for powers of the acquired signal greater than a first power limit value and placing the stimulation parameter equal to a second value of the stimulation parameter for powers of the acquired signal below a second power limit value according to the following law of variability:

$\mathrm{sat}\left(U\left(P_{BF}\right)\right)=\{\begin{array}{cc}{V}_{i2}& \mathrm{for}P_{\mathrm{BF}}\ge P_{\mathrm{BF}2}\\ U_i\left(P_{\mathrm{BF}}\right)& \mathrm{for}P_{\mathrm{BF}2}>P_{\mathrm{BF}}>P_{\mathrm{BF}1}\\ V_{i1}& \mathrm{for}P_{\mathrm{BF}}\le P_{\mathrm{BF}1}\end{array}$

with V_i1 and V_i2 alternatively respectively placed equal to the minimum value V_{i_LowThreshold} and to the maximum value V_{i_HighThreshold}, P_GF2 and P_BF1 equal, respectively, to a maximum limit value and a minimum limit value of the saturation ranges of the stimulation parameter V_i and U_i (P_BF) being a law of variability U_i(P_BF) of the stimulation parameter V_i outside of the saturation ranges. More preferably, the law of variability U_i(P_BF) of the stimulation parameter outside of the saturation ranges is of the following type:

$U_i\left(P_{BF}\right)=K_{1i}\frac{V_{i2}-V_{i1}}{P_{\mathrm{BF}2}-P_{BF1}}\left(P_{BF}-P_{BF1}\right)+V_{i1}.$

Even more preferably, the law of variability U_i(P_BF) of the stimulation parameter V_i outside of the saturation ranges comprises a further additional term (K_2i), thereby resulting in:

$U_i\left(P_{BF}\right)=K_{1i}\frac{V_{i2}-V_{i1}}{P_{\mathrm{BF}2}-P_{BF1}}\left(P_{BF}-P_{BF1}\right)+V_{i1}+K_{2i}.$

Alternatively, the law of variability U_i(P_BF) of the stimulation parameter outside of the saturation ranges is of the following type:

$U_i\left(P_{BF}\right)=K_{1i}\left(V_{i2}-V_{i1}\right)*\left(\frac{1}{1+e^{-p\left(P_{BF}-\frac{P_{BF2}-P_{\mathrm{BF}1}}{2}\right)}}\right)+V_{i1}.$

More preferably, the law of variability U_i(P_BF) of the stimulation parameter V_i outside of the saturation ranges comprises a further additional term (K_2i), thereby resulting in:

$U_i\left(P_{BF}\right)=K_{1i}\left(V_{i2}-V_{i1}\right)*\left(\frac{1}{1+e^{-p\left(P_{BF}-\frac{P_{BF2}-P_{\mathrm{BF}1}}{2}\right)}}\right)+V_{i1}+K_{2i}.$

Advantageously, the selected laws of variability in feedback allow a vast range of adjustment strategies to be carried out through a suitable calibration of the parameters thereof. This flexibility is an essential requirement that allows the use of the apparatus according to the present invention in association with multiple different neurological disorders, each characterised by mutually heterogeneous symptoms that in turn correlate with the local field potentials in different frequency bands and with different time dynamics.

Preferably, the at least one acquisition module comprises processing means for transforming the acquired signal characteristic of cerebral activity in the frequency domain.

In an even more preferred manner, the processing means for transforming the acquired signal characteristic of cerebral activity in the frequency domain implement a Fourier transform, preferably of the FFT (Fast Fourier Transform) type.

More preferably, the processing means for transforming the acquired signal characteristic of cerebral activity in the frequency domain are of the hardware or software type.

Preferably, the at least one acquisition module comprises an integral block for conditioning the acquired signal characteristic of cerebral activity.

Alternatively or in combination, the at least one acquisition module comprises a derivative block for conditioning the acquired signal characteristic of cerebral activity.

In this way, advantageously, it is possible to carry out prior conditioning of the signal fed to the control module that is independent and specific for each stimulation parameter.

Preferably, the at least one stimulation module is a pulse generator.

Preferably, the at least one control module is suitable for implementing the above law of variability U_i(P_BF).

Preferably, the frequency band BF and the parameters of the law of variability, such as a first V_i1 and a second V_i2 threshold value of a stimulation parameter V_i, a maximum power value P_BF2 and a minimum power value P_BF1 are obtained according to the following steps:

a) Identifying at least one maximum threshold value V_{i_HighThreshold} of a stimulation parameter V_i above which the patient shows signs of side effects caused by the stimulation and a minimum threshold value V_{i_LowThreshold} for which the patient shows the minimum or zero benefit induced by stimulation, and placing the extremes of the predetermined stimulation range ([V_i2; V_i1]) alternatively respectively equal to the maximum threshold value V_{i_HighThreshold} and minimum threshold value V_{i_LowThreshold} of the stimulation parameter or to a percentage thereof;b) Determining the frequency band BF, detecting a frequency peak of the power spectrum of a signal characteristic of cerebral activity of the patient recorded in the absence of stimulation, the frequency band BF being centred on such a frequency peak and having a bandwidth selected arbitrarily;c) Recording the trend over time of the power P_BF of a signal characteristic of cerebral activity calculated in the frequency band BF in the three conditions:

• • base state;
  • active stimulation at the maximum threshold value V_{i_HighThreshold} of the stimulation parameter V_i and pharmacological therapy absent; and
  • active stimulation at the maximum threshold value V_{i_HighThreshold} of the stimulation parameter V_i and pharmacological therapy administered and active;d) Identifying a maximum power value P_BF2 and a minimum power value P_BF1 of the trend recorded at step c).

Preferably, the stimulation signal comprises a train of pulses and the stimulation parameter V_i is chosen from the group consisting of:

• • The amplitude of the stimulation pulses;
  • The stimulation pulse repetition frequency;
  • The stimulation pulse duration.

Preferably, the maximum power value P_BF2 of the signal characteristic of cerebral activity calculated in the frequency band BF corresponds to the power value P_OFFOFF that can be determined when the patient is in the base state and the minimum power value P_BF1 corresponds to the power value P_ONON that can be determined when the patient is subjected both to pharmacological therapy and to stimulation therapy active at the maximum threshold value V_{i_HighThreshold} of the stimulation parameter.

More preferably, the frequency band BF is a sub-band of the beta band (10-35 Hz).

Alternatively, the minimum power value P_BF1 of the signal characteristic of cerebral activity calculated in the frequency band BF corresponds to the power value P_OFFOFF that can be determined when the patient is in the base state and the maximum power value P_BF2 corresponds to the power value P_ONON that can be determined when the patient is subjected both to pharmacological therapy and to stimulation therapy active at the maximum threshold stimulation parameter V_{i_HighThreshold}. Preferably, in this case, the frequency band BF is a sub-band of the low frequencies (4-10 Hz). Preferably, the signal characteristic of cerebral activity is a signal coming from the grey nucleus. Preferably, the power calculated in at least one frequency band of the acquired signal characteristic of cerebral activity is compared with a reference value and, based on the difference between the calculated power and the reference value, a control signal for a stimulation module is generated that sets the stimulation parameters.

Preferably, the power P_BF of the acquired signal characteristic of cerebral activity is integrated with a time constant τ.

Advantageously, the time constant τ is selected as a function of the adjustment that is wished to be obtained: the greater the time constant r, the greater the accuracy of the adjustment but also the delay in response of the system. Therefore, the time constant τ with which the power of band P_BF is integrated is chosen seeking the ideal compromise between accuracy and speed of response of the adjustment.

Alternatively or in addition, the power P of the acquired signal characteristic of cerebral activity is a derivative.

Further characteristics and advantages of the present invention will become clearer from the following detailed description of some preferred embodiments thereof, made with reference to the attached drawings.

The different characteristics in the single configurations can be combined with each other as desired according to the previous description, if it were necessary to have advantages resulting specifically from a particular combination.

In such drawings,

FIG. 1 is a schematic representation of an apparatus for treating neurological disorders according to a preferred embodiment of the present invention;

FIG. 2 is a block diagram of the main steps of the method for treating neurological disorders in accordance with the present invention;

FIG. 3 is a detailed block diagram of a step of the method for treating neurological disorders according to the present invention;

FIG. 4a is a diagram of the variation in power in beta band during the calibration step of the maximum and minimum spectral power thresholds in the case of Parkinson's disease;

FIG. 4b is a diagram of the variation in spectral power at the low frequencies during the calibration step of the maximum and minimum spectral power thresholds in the case of Parkinson's disease.

In the following description, to illustrate the figures identical reference numerals or symbols are used to indicate constructive elements with the same function. Moreover, for the sake of clarity of illustration, some references are not repeated in all of the figures.

With reference to FIG. 1, an apparatus for treating neurological disorders is shown, wholly indicated with 10.

The apparatus for treating neurological disorders 10 comprises at least one electro-catheter 11 suitable for being implanted in the brain of a patient to administer an electric stimulation. The electro-catheter 11 preferably comprises at least three metallic contacts accessible through external connections also called electrodes 12. However, it is obviously possible to hypothesise alternative solutions in which the electrodes are not necessarily carried by one same electro-catheter.

The electrodes 12 are connected to a processing and stimulation device 14 that, in the embodiment illustrated in FIG. 1, comprises three functional modules connected together in feedback and interoperating: a stimulation module 16, an acquisition module 18 and a control module 20.

The stimulation module 16 is adapted to generate a stimulation signal V_stim, characterised by a set of parameters V_a, V_d, V_f, and to send to the electrodes 12 the stimulation signal generated. In particular, the stimulation module 16 is a generator of pulses defined by the amplitude, frequency and duration of the pulses.

The acquisition module 18 is assigned to the acquisition of a signal characteristic of cerebral activity coming from the brain of the patient. In detail, the acquisition module 18 comprises processing means for transforming the acquired signal characteristic of cerebral activity in the frequency domain. Specifically, the processing means carry out an FFT (Fast Fourier Transform) and can be made through hardware means and/or software means. The acquisition module 18 also preferably comprises an integral block and a derivative block (not illustrated) of the signal characteristic of cerebral activity transformed in the frequency domain.

The control module 20 implements an adjuster, preferably a feedback controller. As illustrated more clearly in FIG. 3, the control module 20 is functionally connected, upstream, to the acquisition module 18 and, downstream, to the stimulation module 16 that determines the stimulation signal V_stim. As a function of the spectral power of the signal characteristic of cerebral activity acquired by the acquisition module 18, the control module 20 determines at least one signal based on which at least one parameter V_a, V_d, V_f of the stimulation signal V_stim set by the stimulation module 16 is defined.

Advantageously, the control module 20 receives in input the signal acquired in the time domain or transformed in the frequency domain by the acquisition module 18 to determine its power. Based on such a power, preferably integrated based on a time constant τ and/or derived according to specificities of the stimulation parameter to be adjusted, the stimulation parameters are calculated based on a transfer function having a saturating trend such as to also determine that the stimulation parameters are variable between two saturation values (V_{i_HighThreshold}; V_{i_LowThreshold}) between which the stimulation is actually effective.

In particular, the transfer function having a saturating trend is implemented as a piecewise function based on ranges of values of the input power, i.e. such as to place the stimulation parameter V_i equal to a maximum value V_{i_HighThreshold} or to a minimum value V_{i_LowThreshold} in the saturation ranges, allowing the stimulation parameter V_i to vary, outside the saturation ranges, as a function of the power P_BF of the signal acquired according to a law of variability U_i(P_BF).

This translates into the following transfer function, where the saturation ranges correspond to powers P_BF of the acquired signal greater than a first power limit value P_GF2 or powers P_BF of the acquired signal below a second power limit value P_BF1:

$\mathrm{sat}\left(U\left(P_{BF}\right)\right)=\{\begin{array}{cc}{V}_{i2}& \mathrm{for}P_{\mathrm{BF}}\ge P_{\mathrm{BF}2}\\ U_i\left(P_{\mathrm{BF}}\right)& \mathrm{for}P_{\mathrm{BF}2}>P_{\mathrm{BF}}>P_{\mathrm{BF}1}\\ V_{i1}& \mathrm{for}P_{\mathrm{BF}}\le P_{\mathrm{BF}1}\end{array}.$

In particular, the law of variability U_i(P_BF) of the stimulation parameter outside the saturation ranges is of the following type:

$U_i\left(P_{BF}\right)=K_{1i}\frac{V_{i2}-V_{i1}}{P_{\mathrm{BF}2}-P_{BF1}}\left(P_{BF}-P_{BF1}\right)+V_{i1}+K_{2i}.$

Alternatively, the law of variability U_i(P_BF) of the stimulation parameter outside of the saturation ranges is of the sigmoid type:

$U_i\left(P_{BF}\right)=K_{1i}\left(V_{i2}-V_{i1}\right)*\left(\frac{1}{1+e^{-p\left(P_{BF}\frac{P_{BF2}-P_{BF1}}{2}\right)}}\right)+V_{i1}+K_{2i}.$

In both cases, the parameter K_1i, represents a proportional adjustment factor preferably having unitary value so that the maximum excursion of P_BF corresponds to the maximum excursion of the output V_i(P_BF). In the case in which the parameter K_1i takes on values greater than 1, an additional saturation block is foreseen in order to ensure that the stimulation parameter V_i is kept within the predetermined stimulation range. In the sole case of sigmoid function, the parameter p is adjusted based on the desired value of the function U_i(P_BF) around the point P_GF2 and P_BF1.

Specifically, since deep brain stimulation uses a stimulus defined by three stimulation parameters V_a, V_d, V_f relative, respectively, to amplitude, duration and frequency of the stimulation signal V_stim, the control module 20 foresees to implement a respective law of variability V_a(P_BF), V_d(P_BF), V_f(P_BF) for each stimulation parameter V_a, V_d, V_f. The stimulation signal V_stim in output from the stimulation module 16 is characterised by the parameters V_a, V_d, V_f, calculated based on the respective output of the control module 20.

Advantageously, the time constant T based on which the integration of the spectral power P_BF takes place, is selected as a function of the control requirements: the greater the time constant T the smaller the variance on the evaluation of the spectral power P_BF of the acquired signal and therefore on the instantaneous clinical state of the patient. However, increasing the time constant T increases the delay in identifying the clinical state of the patient.

Preferably, the control allows variable setting of the time constant T, so that the most suitable time constant can be set each time based on the specific application, taking into account the compromise between speed and accuracy of detection of the power.

The adjuster implemented by the control module 20 is characterised by a wide degree of flexibility thanks to the possibility of calibrating the adjustment parameters K_1i, K_2i P_BF2, P_BF1, τ, V_i1 V_i2. It is thus possible to carry out a vast range of adjustment strategies.

The operating method 100 of the apparatus for treating neurological disorders 10 is schematically illustrated in FIG. 2.

Prior to the stimulation treatment there is an initialisation session (step 110). The initialisation session is carried out after the patient has spent a sufficient time (generally 12 hours) without pharmacological medication. After such a time period without pharmacological therapy, the patient is considered to be in the so-called “OFF-OFF” or base clinical condition, i.e. in the absence of stimulation and with effect of pharmacological therapy having worn off.

Then there is the identification (step 111) of at least one of the threshold values of the specific stimulation parameters of the patient based on which to set the treatment. In particular, in simplifying terms, the amplitude of the stimulation voltage is identified.

Such identification takes place through an expert, like for example a neurologist specialised in the treatment of neurological disorders through deep brain stimulation. The step of identifying the parameters (step 111) therefore takes place through a series of stimulation tests with different parameter values, based on which the expert decides the maximum threshold value V_{i_HighThreshold}, to obtain the maximum clinical effect before the appearance of side effects, and minimum threshold value V_{i_LowThreshold}, to obtain the minimum or zero clinical effect. The saturation values of the stimulation parameters V_i2 and V_i1 can be placed equal, respectively, to the maximum threshold value V_{i_HighThreshold} and minimum threshold value V_{i_LowThreshold} or alternatively, respectively equal to the minimum threshold value V_{i_LowThreshold} and maximum threshold value V_{i_HighThreshold}.

In the case in which the parameter analysed is the amplitude V_a of the stimulation signal V_stim, the step of identifying the stimulation parameters leads to determining the maximum and minimum amplitude of the voltage that can be set V_{a_HighThreshold} and V_{a_LowThreshold}. Similarly, the maximum/minimum frequency and/or the maximum/minimum duration of the stimulation signal V_stim can be identified.

Then there is a recording (step 112) of the neurophysiological signal: the harmonic content of the neurophysiological signal of the specific patient in the absence of stimulation is detected and analysed to identify the characteristics of the specific frequency spectrum of the patient. In particular, at least one frequency peak is identified, with respect to which the at least one frequency band BF is centred based on which to calculate the spectral power of the signal. The frequency bands are defined through a minimum frequency and a maximum frequency: fw_min<BF<fw_max and correspond to the frequency bands with which the symptoms of the neurological disorders that it is wished to counteract most probably correlate. The frequencies fw_min and fw_max can be selected arbitrarily.

Once the stimulation parameters and the frequency band have been defined, there is a calibration step (step 113) in which the signal characteristic of cerebral activity is recorded in a plurality of different conditions to extract the values of the power in the frequency band defined previously. Specifically, for the calibration step (step 113), the neurophysiological signal of the patient at the base state is detected, i.e. in the absence of therapies of any kind (pharmacological or stimulation) also called OFF-OFF state, and the power in the band of interest is stored. The recording of the neurophysiological signal at the OFF-OFF state takes place for an initial period, in general equal to 20 minutes. Once the initial period has ended, the stimulation is brought to the maximum threshold stimulation values V_{i_HighThreshold} determined previously. The processing and stimulation device proceeds to store the power of the signal characteristic of cerebral activity of the patient in the clinical OFF-ON state, i.e. in the absence of pharmacological therapy (LEVOdopamine), but in the presence of stimulation. After a further time period, in general equal to another 20 minutes, the pharmacological therapy is started again proceeding to store the power. After a third time period, the drug taken is considered to be completely assimilated and the patient is in the clinical ON-ON state, i.e. in the presence of pharmacological therapy and of stimulation. The storage of the power of the signal characteristic of cerebral activity of the patient is therefore ended.

Once the storage has ended, in the three initialisation steps 110, of the power calculated in the frequency band identified initially, the maximum power value P_BF2 and minimum power value P_BF1 are extrapolated.

In the specific case illustrated in FIG. 4a, relative to a frequency band of interest coinciding with a sub-band of the beta band (10-35 Hz), the maximum power value P_BF2 coincides with the power value in the OFF-OFF state (N_βOFFOFF), whereas the minimum power value P_BF1 coincides with the power value in the ON-ON state (N_βONON).

In example terms, the case is also shown in which the frequency band of interest coincides with the low frequencies (4-10 Hz). In this case, as shown in FIG. 4b, the minimum power value P_BF1 coincides with the power value in the OFF-OFF state (P_lowOFFOFF), whereas the maximum power value P_BF2 coincides with the power value at the ON-ON state (P_lowONON).

Once the initialisation step 110 of the therapy has ended, the method 100 for treating neurological disorders comprises the repetition of the following steps.

The delivery of the deep brain stimulation is started (step 120).

Thereafter (step 130), the acquisition module 18 records a signal characteristic of cerebral activity of the patient (sub-step 131), preferably the local field potentials LFP recorded at the grey nucleus, and thereafter transforms it (sub-step 132) preferably in the frequency domain, for example through FFT (Fast Fourier Transform), determining its spectral power P_BF (sub-step 133). Preferably, the power is also integrated based on the time constant τ (sub-step 134).

Finally, based on the spectral power P_BF recorded, there is a step of updating the stimulation parameters (step 140).

For this purpose, the control module 20 preferably compares the power P_BF with a range of reference values [P_BF2;P_BF1] that correlate more with the band of interest. Based on the difference between the power P_BF calculated and the lower limit P_BF1 of such a reference range [P_BF2;P_BF1] a control signal for a stimulation module (16) is generated which sets the stimulation parameters (V_a, V_d, V_f) according to the law of variability V_i(P_BF) given above.

In the case in which the frequency band of interest coincides with a sub-band of the beta band (10-35 Hz) that, as will be seen hereinafter, proves particularly suitable for the treatment of some symptoms of Parkinson's disease, the upper extreme P_BF2 of the range of values is equal to the power P_βOFFOFF at the OFF-OFF state and the lower extreme P_BF1 of the range of values is equal to the power P_βONON at the ON-ON state determined in the initialisation step, with P_BF2>P_BF1.

In example terms, the case is now discussed in which the treatment method according to the invention is specifically used for treating clinical-motor functions of Parkinson's disease that most correlate with the beta band β, which has therefore been identified as the reference frequency band BF based on which to calculate the spectral power of the acquired local field potentials. For this treatment it has also proven sufficient to carry out an adaptation of just the stimulation amplitude V_a.

In this case, the law of variability takes the form:

$U_i\left(P_{BF}\right)=K_{1i}\frac{V_{i2}-V_{i1}}{P_{\mathrm{BF}2}-P_{\mathrm{BF}1}}\left(P_{BF}-P_{\mathrm{BF}1}\right)+V_{i1}+K_{2i}$

With the adjustment parameters K_1i, P_BF2, P_BF1, V_i1, V_i2, K_2i equal to:

P_BF=P_β;P_BF2=P_βOFFOFF P_BF1=P_βONON V_i2=V_{a_HighThreshold} V_i1=V_{a_LowThreshold} K_1i=K_a=1

K_2i=0

In the case of the treatment of the symptoms of Parkinson's disease that correlate with the beta band, there is therefore a simplified adjustment model based on the following law of variability:

$\begin{array}{c}{V}_a\left(P_g\right)=\mathrm{sat}\left(P_g\right)\\ =\{\begin{array}{cc}{V}_{\mathrm{a\_HighThreshold}}& \mathrm{for}P_{\beta }\ge P_{\beta \mathrm{OFFOFF}}\\ \begin{array}{c}\frac{V_{\mathrm{a\_HighThreshold}}-V_{\mathrm{a\_LowThreshold}}}{P_{\beta \mathrm{OFFOFF}}-P_{\beta \mathrm{ONON}}}\\ \left(P_g-P_{\beta \mathrm{ONON}}\right)+V_{\mathrm{a\_LowThreshold}}\end{array}& \begin{array}{c}\mathrm{for}P_{\beta \mathrm{OFFOFF}}>P_{\beta }>P_{\beta \mathrm{ONON}}\\ \end{array}\\ V_{\mathrm{a\_LowThreshold}}& \mathrm{for}P_{\beta }\le P_{\beta \mathrm{ONON}}\end{array}\end{array}$

Alternatively, the case is now discussed in which the treatment method according to the invention is specifically used for the treatment of the clinical-motor fluctuations of Parkinson's disease that most correlate with the band of low frequencies (4-10 Hz), which has therefore been identified as reference frequency band BF based on which to calculate the spectral power of the acquired local field potentials. For this treatment it also proved sufficient to carry out an adaptation of just the stimulation amplitude V_a.

In this case, the law of variability takes the form

$U_i\left(P_{BF}\right)=K_{1i}\frac{V_{i2}-V_{i1}}{P_{\mathrm{BF}2}-P_{BF1}}\left(P_{\mathrm{BF}2}-P_{BF1}\right)+V_{i1}+K_{2i}.$

With the adjustment parameters K_1i, P_BF2, P_BF1, V_i1, V_i2, K_2i equal to:

P_BF=P_low P_BF2=P_lowONON P_BF1=P_lowOFFOFF V_i2=V_{a_LowThreshold} V_i1=V_{a_HighThreshold} K_1i=K_a=1

K_2i=0

In the case of treatment of the symptoms of Parkinson's disease that correlate most with the low frequencies, there is therefore a simplified adjustment model defined by the following law of variability:

$\begin{array}{c}{V}_a\left(P_{\mathrm{low}}\right)=\mathrm{sat}\left(P_{\mathrm{low}}\right)\\ =\{\begin{array}{cc}{V}_{a_{\mathrm{LowThreshold}}}& \mathrm{for}P_{\mathrm{low}}\ge P_{\mathrm{low}\mathrm{ONON}}\\ \begin{array}{c}\frac{V_{a_{\mathrm{LowThreshold}}}-V_{a_{\mathrm{HighThreshold}}}}{P_{\mathrm{low}\mathrm{ONON}}-P_{\mathrm{low}\mathrm{OFFOFF}}}\\ \left(P_{\mathrm{low}}-P_{\mathrm{low}\mathrm{OFFOFF}}\right)+V_{a_{\mathrm{HighThreshold}}}\end{array}& \begin{array}{c}\mathrm{for}P_{\mathrm{low}\mathrm{OFFOFF}}>P_{\mathrm{low}}>P_{\mathrm{low}\mathrm{ONON}}\\ \end{array}\\ V_{a_{\mathrm{HighThreshold}}}& \mathrm{for}P_{\mathrm{low}}\le P_{\mathrm{low}\mathrm{OFFOFF}}\end{array}\end{array}$

From the description that has been made the characteristics of the apparatus and of the method for treating neurological disorders object of the present invention are clear, just as the relative advantages are also clear.

From the embodiments described above further variants are possible, without departing from the teaching of the invention.

Finally, it is clear that an apparatus and a method for treating neurological disorders thus conceived can undergo numerous modifications and variants, all of which are covered by the invention; moreover, all of the details can be replaced by technically equivalent elements. In practice, the materials used, as well as the sizes, can be whatever according to the technical requirements.

Claims

1-23. (canceled)

24: An apparatus for treating neurological disorders comprising: an electrode implantable in the brain of a patient;a processing and stimulation device connected to the electrode, wherein the processing and stimulation device comprises: a stimulation module adapted to generate a stimulation signal (Vstim) to be sent to the electrode, the stimulation signal (Vstim) having a plurality of stimulation parameters including a stimulation amplitude, stimulation duration, and stimulation frequency (Va, Vd, Vf),an acquisition module configured to acquire cerebral activity signals from the brain of the patient and to calculate a power (PBF) of a beta frequency band of the acquired cerebral activity signals, anda control module configured to adjust a stimulation parameter (Va, Vd, Vf) of the stimulation signal (Vstim) when the power (PBF) of the beta frequency band meets or exceeds an upper power value threshold (PBF2),wherein the upper power value threshold (PBF2) is a power of the beta frequency band of the patient's cerebral activity signals measured in the absence of pharmacological therapy (POFF).

25: The apparatus of claim 24, wherein when the power (PBF) of the beta frequency band meets or exceeds the upper power value threshold (PBF2), a stimulation parameter of the stimulation signal (Vstim) is set to a maximum value (Vi_HighThreshold) of the stimulation parameter that provides a maximum clinical effect before side effects appear.

26: The apparatus of claim 24, wherein the control module is further configured to adjust a stimulation parameter of the stimulation signal (Vstim) when the power (PBF) of the beta frequency band is at or below a lower power value threshold (PBF1), wherein the lower power value threshold (PBF1) is a power of the beta frequency band of the patient's cerebral activity signals measured in the presence of pharmacological therapy (PON).

27: The apparatus of claim 26, wherein when the power (PBF) of the beta frequency band is at or below the lower power value threshold (PBF1), a stimulation parameter of the stimulation signal (Vstim) is set to a minimum value (Vi_LowThreshold) of the stimulation parameter that provides a minimum clinical effect.

28: The apparatus of claim 26, wherein the upper power value threshold (PBF2) is a power of the beta frequency band of the patient's cerebral activity signals measured in the absence of pharmacological therapy and in the absence of electrical stimulation (POFFOFF).

29: The apparatus of claim 28, wherein the lower power value threshold (PBF1) is a power of the beta frequency band of the patient's cerebral activity signals measured in the presence of pharmacological therapy and in the presence of electrical stimulation (PONON).

30: The apparatus of claim 29, wherein when the power (PBF) of the beta frequency band meets or exceeds POFFOFF, a stimulation parameter of the stimulation signal is set to a maximum value (Vi_HighThreshold) of the stimulation parameter that provides a maximum clinical effect before side effects appear.

31: The apparatus of claim 30, wherein when the power (PBF) of the beta frequency band is at or below PONON, a stimulation parameter of the stimulation signal is set to a minimum value (Vi_LowThreshold) of the stimulation parameter that provides a minimum clinical effect.

32: The apparatus of claim 26, wherein the control module is configured to adjust a stimulation parameter (Va, Vd, Vf) of the stimulation signal (Vstim) according to a piecewise transfer function based on the power (PBF) of the beta frequency band, wherein the piecewise transfer function is:

33: The apparatus of claim 32, wherein the law of variability Ui(PBF) is:

34: A method for treating a neurological disorder of a patient, the method comprising: acquiring cerebral activity signals from the patient;calculating a power (PBF) of a beta frequency band of the acquired cerebral activity signals; andadjusting a stimulation parameter of a stimulation signal (Vstim) when the power (PBF) of the beta frequency band meets or exceeds an upper power value threshold (PBF2), wherein the upper power value threshold (PBF2)1 is a power of the beta frequency band of the patient's cerebral activity signals measured in the absence of pharmacological therapy (POFF).

35: The method of claim 34, wherein adjusting a stimulation parameter of the stimulation signal (Vstim) comprises setting a stimulation parameter of the stimulation signal (Vstim) to a maximum value (Vi_HighThreshold) of the stimulation parameter when the power (PBF) of the beta frequency band meets or exceeds the upper power value threshold (PBF2), wherein the maximum value (Vi_HighThreshold) is a stimulation parameter value that provides a maximum clinical effect before side effects appear.

36: The method of claim 34, wherein adjusting a stimulation parameter of the stimulation signal (Vstim) further comprises adjusting a stimulation parameter when the power (PBF) of the beta frequency band is at or below a lower power value threshold (PBF1), wherein the lower power value threshold (PBF1) is a power of the beta frequency band of the patient's cerebral activity signals measured in the presence of pharmacological therapy (PON).

37: The method of claim 36, wherein adjusting a stimulation parameter of the stimulation signal (Vstim) further comprises setting a stimulation parameter of the stimulation signal (Vstim) to a minimum value (Vi_LowThreshold) of the stimulation parameter when the power (PBF) of the beta frequency band is at or below the lower power value threshold (PBF1), wherein the minimum value (Vi_LowThreshold) is a stimulation parameter value that provides a minimum clinical effect.

38: The method of claim 36, wherein the upper power value threshold (PBF2) is a power of the beta frequency band of the patient's cerebral activity signals measured in the absence of pharmacological therapy and in the absence of electrical stimulation (POFFOFF).

39: The method of claim 38, wherein the lower power value threshold (PBF1) is a power of the beta frequency band of the patient's cerebral activity signals measured in the presence of pharmacological therapy and in the presence of electrical stimulation (PONON).

40: The method of claim 39, wherein adjusting a stimulation parameter of the stimulation signal (Vstim) further comprises setting a stimulation parameter of the stimulation signal to a maximum value (Vi_HighThreshold) when the power (PBF) of the beta frequency band meets or exceeds POFFOFF, wherein the maximum value (Vi_HighThreshold) is a stimulation parameter value that provides a maximum clinical effect before side effects appear.

41: The method of claim 40, wherein adjusting a stimulation parameter of the stimulation signal (Vstim) further comprises setting a stimulation parameter of the stimulation signal to a minimum value (Vi_LowThreshold) of the stimulation parameter when the power (PBF) of the beta frequency band is at or below PONON, wherein the minimum value (Vi_LowThreshold) is a stimulation parameter value that provides a minimum clinical effect.

42: The method of claim 36, wherein adjusting a stimulation parameter of the stimulation signal (Vstim) comprises adjusting a stimulation parameter (Va, Vd, Vf) of the stimulation signal (Vstim) according to a piecewise transfer function based on the power (PBF) of the beta frequency band, wherein the piecewise transfer function is:

43: The method of claim 42, wherein the law of variability Ui(PBF) is:

44: The method of claim 34, further comprising identifying a maximum value (Vi_HighThreshold) of a stimulation parameter of the stimulation signal by measuring a value of the stimulation parameter at which side effects appear.

45: The method of claim 34, further comprising identifying a minimum value (Vi_LowThreshold) of a stimulation parameter of the stimulation signal by measuring a value of the stimulation parameter below which no clinical effect is observed.

46: An apparatus for treating neurological disorders comprising: an electrode implantable in the brain of a patient;a processing and stimulation device connected to the electrode, wherein the processing and stimulation device comprises:a stimulation module adapted to generate a stimulation signal (Vstim) to be sent to the electrode, the stimulation signal (Vstim) having a plurality of stimulation parameters including a stimulation amplitude, stimulation duration, and stimulation frequency (Va, Vd, Vf),an acquisition module configured to acquire cerebral activity signals from the brain of the patient and to calculate a power (PBF) of a low frequency band (4-10 Hz) of the acquired cerebral activity signals, anda control module configured to adjust a stimulation parameter (Va, Vd, Vf) of the stimulation signal (Vstim) when the power (PBF) of the low frequency band meets or exceeds an upper power value threshold (PBF2),wherein the upper power value threshold (PBF2) 1 is a power of the low frequency band of the patient's cerebral activity signals measured in the presence of pharmacological therapy (PON).

47: The apparatus of claim 46, wherein the control module is further configured to adjust a stimulation parameter of the stimulation signal (Vstim) when the power (PBF) of the low frequency band is at or below a lower power value threshold (PBF1), wherein the lower power value threshold (PBF1) is a power of the low frequency band of the patient's cerebral activity signals measured in the absence of pharmacological therapy (POFF).

48: The apparatus of claim 47, wherein the upper power value threshold (PBF2) is a power of the low frequency band of the patient's cerebral activity signals measured in the presence of pharmacological therapy and in the presence of electrical stimulation (PONON), and wherein the lower power value threshold (PBF1) is a power of the low frequency band of the patient's cerebral activity signals measured in the absence of pharmacological therapy and in the absence of electrical stimulation (POFFOFF).

49: The apparatus of claim 47, wherein when the power (PBF) of the low frequency band meets or exceeds the upper power value threshold (PBF2), a stimulation parameter of the stimulation signal (Vstim) is set to a minimum value (Vi_LowThreshold) of the stimulation parameter that provides a minimum clinical effect, and wherein when the power (PBF) of the low frequency band is at or below the lower power value threshold (PBF1), a stimulation parameter of the stimulation signal (Vstim) is set to a maximum value (Vi_HighThreshold) of the stimulation parameter that provides a maximum clinical effect before side effects appear.

50: The apparatus of claim 47, wherein the control module is configured to adjust a stimulation parameter (Va, Vd, Vf) of the stimulation signal (Vstim) according to a piecewise transfer function based on the power (PBF) of the low frequency band, wherein the piecewise transfer function is:

51: The apparatus of claim 50, wherein the law of variability Ui(PBF) is:

52: A method for treating a neurological disorder of a patient, the method comprising: acquiring cerebral activity signals from the patient;calculating a power (PBF) of a low frequency band of the acquired cerebral activity signals; andadjusting a stimulation parameter of a stimulation signal (Vstim) when the power (PBF) of the low frequency band (4-10 Hz) meets or exceeds an upper power value threshold (PBF2), wherein the upper power value threshold (PBF2) is a power of the low frequency band of the patient's cerebral activity signals measured in the presence of pharmacological therapy (PON).

53: The method of claim 52, wherein adjusting a stimulation parameter further comprises adjusting a stimulation parameter of the stimulation signal (Vstim) when the power (PBF) of the low frequency band is at or below a lower power value threshold (PBF1), wherein the lower power value threshold (PBF1) is a power of the low frequency band of the patient's cerebral activity signals measured in the absence of pharmacological therapy (POFF).

54: The method of claim 53, wherein adjusting a stimulation parameter of the stimulation signal (Vstim) comprises adjusting a stimulation parameter (Va, Vd, Vf) of the stimulation signal (Vstim) according to a piecewise transfer function based on the power (PBF) of the low frequency band, wherein the piecewise transfer function is:

55: The method of claim 54, wherein the law of variability Ui(PBF) is: