WO2024204823A1 - Deep brain stimulation device - Google Patents

Abstract

A deep brain stimulation device is anticipated that makes it possible to obtain a therapeutic effect while reducing electric power consumption. This deep brain stimulation device comprises: a brain wave detector 1; a drive signal generation device 10 to which an output signal from the brain wave detector 1 is inputted; and a deep brain stimulation element 8 to which a drive signal outputted from the drive signal generation device 10 is applied. The drive signal generation device 10 extracts a signal component in a γ2 frequency band, determines an occurrence rate of waveforms having an intensity greater than or equal to a reference value and a duration greater than or equal to a reference period of time among the extracted signal components, and generates a drive signal having a frequency and an amplitude corresponding to the occurrence rate. The higher the occurrence rate is, the higher the frequency of the drive signal is and the higher the amplitude of the drive signal is.

Description

Deep Brain Stimulator

This disclosure relates to a deep brain stimulation device.

Deep brain stimulation (DBS) is a method to reduce symptoms exhibited by patients with movement disorders (Parkinson's disease, dystonia, dyskinesia, essential tremor, etc.) by electrically stimulating deep parts of the brain, such as the thalamus and the basal ganglia, including the subthalamic nucleus and globus pallidus. The basal ganglia, together with the motor cortex of the cerebral cortex, which outputs motor commands to muscles, form a loop circuit (neural network) via the thalamus, and any impairment of function within the loop will result in movement disorders. The basal ganglia are a group of nerve cells located deep within the cerebrum, and consist of the striatum, globus pallidus, subthalamic nucleus, substantia nigra, etc.

　Known deep brain stimulation therapies include adaptive deep brain stimulation (aDBS) and the conventional constant deep brain stimulation (cDBS), which is widely used in clinical practice. At the clinical stage, activity in the low beta band (13 Hz to 30 Hz) in the basal ganglia, particularly the subthalamic nucleus, may serve as a biomarker for adaptive deep brain stimulation, but further improvements are required. Patent Documents 1 to 4 are known as technologies related to deep brain stimulation.

Non-Patent Document 1 discloses adaptive deep brain stimulation technology.

US Patent Application Publication No. 2016/0263380 US Patent Application Publication No. 2014/0350634 U.S. Pat. No. 1,071,6943 U.S. Pat. No. 1,047,8085

Swann NC, et al., “Adaptive deep brain stimulation for Parkinson’s disease using motor cortex sensing”, Journal of Neural Engineering, May 9, 2018, Vol. 15, Number 4, 46006

　Even with conventional deep brain stimulation devices, it is possible to achieve therapeutic effects by stimulating targeted areas in the brain. For example, when the potential of an EEG signal exceeds a threshold, stimulation of the targeted area can begin at a preset fixed stimulation intensity. Meanwhile, in the medical equipment field, hopes are high for deep brain stimulation devices that can achieve therapeutic effects while reducing power consumption.

The inventors of the present application have conducted extensive research into the mechanism of decreased movement in Parkinson's disease. As a result, it has become clear that in Parkinson's disease, motor commands are prevented from passing through the basal ganglia, making it impossible to initiate movement. When the intention to move arises in the brain, the electrical potential in the brain changes before the actual movement of a body part. The main area that the inventors of the present application focused on as the area where the electrical potential change occurs is the primary motor cortex. The electrical potential of the higher motor cortex (supplementary motor area, premotor cortex) also correlates with movement. Therefore, if this change in the electrical potential in the brain is detected, a predetermined signal processing is performed on the detection result, and the subthalamic nucleus, etc. is stimulated in accordance with the passage of the motor command, making it easier for the motor command to pass, it is possible to treat movement disorders such as Parkinson's disease.

The deep brain stimulation device disclosed herein comprises an EEG detector, a drive signal generating device that receives an output signal from the EEG detector, extracts signal components in a frequency band that includes a center frequency selected from 80 Hz to 200 Hz and has a lower limit frequency at least higher than 30 Hz, determines the frequency of waveforms among the extracted signal components that have an intensity equal to or greater than a reference value and a duration equal to or greater than a reference time, and generates a drive signal having a frequency and amplitude according to the frequency, and a deep brain stimulation element that is provided with the drive signal output from the drive signal generating device.

The gamma 2 frequency band of the cortical EEG in the primary motor cortex is a frequency band that includes a center frequency selected from 80 Hz to 200 Hz, and has a lower limit frequency that is at least higher than 30 Hz. The signal components in the gamma 2 frequency band in the primary motor cortex are extracted and detected from the output signal of the EEG detector as the potential before the actual movement of the body part (motor preparation potential). When the extracted signal components are subjected to a specified signal processing to generate a drive signal for stimulation, it is possible to obtain an effective treatment effect for movement disorders (Parkinson's disease) with little power consumption.

The above-mentioned predetermined signal processing is a process of determining the frequency of waveforms among the extracted signal components whose intensity is equal to or exceeds a reference value and whose duration is equal to or exceeds a reference time, and generating a drive signal having a frequency and amplitude according to the frequency. In other words, the number of times that a waveform that satisfies the condition occurs within a certain period of time is determined, and the frequency and amplitude of the drive signal are determined according to this number of occurrences. The drive signal generating device can preferably generate a drive signal such that the higher the frequency, the higher the frequency of the drive signal, and the higher the frequency, the larger the amplitude of the drive signal. In other words, both the frequency and amplitude of the drive signal monotonically increase with frequency.

Of course, an upper limit can be set for the frequency and amplitude of the drive signal, and in this case, the frequency and amplitude will saturate if the upper limit is exceeded. Therefore, it is not necessary to always monotonically increase in the signal processing. As the frequency increases, the frequency and amplitude may increase not only linearly but also nonlinearly. Even with nonlinear control, stimulating the subthalamic nucleus etc. makes it easier for motor commands to pass, making it possible to treat movement disorders. Furthermore, the frequency and amplitude can be changed not only continuously but also discretely with respect to the frequency. Even with discrete control, stimulating the subthalamic nucleus etc. makes it easier for motor commands to pass, making it possible to treat movement disorders.

An analog filter or a digital filter can be used to extract the above frequency band in the drive signal generating device. When an analog filter is used, a window discriminator can be placed in the subsequent stage. The window discriminator generates one pulse voltage and outputs a pulse signal every time a waveform whose intensity is equal to or greater than a reference value and whose duration is equal to or greater than a reference time is detected. The control device can control the frequency and amplitude of the drive signal according to the frequency of the pulse signal output from the window discriminator. When a digital filter is used, the output signal from the EEG detector can be digitized and input to the control device to generate a drive control signal having a frequency and amplitude according to the frequency, and this can be input to the drive circuit.

The drive circuit can use an analog isolator that has an input section to which a drive control signal from a control device is input, and an output section that generates a drive signal isolated from the input based on the drive control signal. Since isolated signal transmission is performed between the input and output sections, the maximum value of the drive signal can be limited on the output section side, improving safety. The isolator has the function of electrically isolating between its input terminal and output terminal, regardless of the type of input, such as direct current or pulse current.

The drive signal generating device is housed in a sealed container, and a signal line extending outside the sealed container is connected to the drive signal output terminal of the drive signal generating device, and the signal line is preferably connected to the deep brain stimulation element. Since the drive signal generating device that performs electrical processing is housed inside the sealed container, it is possible to protect the drive signal generating device. It is also possible to implant the sealed container inside the body.

The deep brain stimulation device of the present disclosure comprises an EEG detector, a drive signal generating device that extracts signal components of a predetermined frequency band from the output signal of the EEG detector, determines the frequency of waveforms among the extracted signal components whose intensity is equal to or exceeds a reference value and whose duration is equal to or exceeds a reference time, and generates a drive signal having a frequency and amplitude corresponding to the frequency, and a deep brain stimulation element to which the drive signal output from the drive signal generating device is applied. The signal components of the predetermined frequency band are frequency bands that correlate with movement, and are preferably the frequency bands described above. The drive signal generating device preferably generates a drive signal such that the higher the frequency, the higher the frequency of the drive signal, and the higher the frequency, the larger the amplitude of the drive signal.

The deep brain stimulation device disclosed herein comprises an EEG detector, a drive signal generating device that extracts signal components of a predetermined frequency band from the output signal of the EEG detector, determines the frequency of waveforms among the extracted signal components whose intensity is equal to or exceeds a reference value and whose duration is equal to or exceeds a reference time, and generates a drive signal having a frequency or amplitude corresponding to the frequency, and a deep brain stimulation element to which the drive signal output from the drive signal generating device is applied. The signal components of the predetermined frequency band are frequency bands that correlate with movement, and preferably the above-mentioned frequency bands. Increasing the frequency and amplitude of the drive signal increases the power supplied to the deep brain stimulation element, and increases the amount of stimulation. Even when only one of these parameters is increased, a therapeutic effect can be expected, but it is more effective to increase both parameters.

This deep brain stimulation device makes it possible to obtain therapeutic effects while reducing power consumption.

FIG. 1 is a diagram showing a deep brain stimulation apparatus according to a first embodiment. FIG. 2 is a timing chart of the input signal _SF and the pulse output signal _SP of the window discriminator. FIG. 3 is a plan view of the deep brain stimulation element (FIG. 3(A)) and an enlarged view of the tip portion (FIG. 3(B)). FIG. 4 is a graph showing temporal changes in EEG signals in the beta, gamma 1, and gamma 2 bands (FIG. 4(A) shows data for monkeys A and B combined, FIG. 4(B) shows data for monkey A, and FIG. 4(C) shows data for monkey B). FIG. 5 is a graph showing the success rate of operations that satisfy the criteria conditions in the DBS-OFF, aDBS, and cDBS states. FIG. 6 is a graph showing the power of the output signal in the gamma 2 band during the premovement, movement, and return periods (FIG. 6(A) shows data for monkeys A and B combined, FIG. 6(B) shows data for monkey A, and FIG. 6(C) shows data for monkey B). FIG. 7 is a graph showing the time required to complete an operation in the DBS-OFF, aDBS, and cDBS states, normalized with respect to the DBS-OFF state. FIG. 8 is a graph showing the ratio of the charge supplied by the aDBS to the charge supplied by the cDBS. FIG. 9 is a diagram showing a deep brain stimulation apparatus according to the second embodiment. FIG. 10 is a diagram showing a deep brain stimulation device as a medical device. FIG. 11 is a diagram showing a deep brain stimulation device equipped with a plurality of detection elements. FIG. 12 is a diagram showing a circuit configuration for receiving and processing a plurality of electroencephalogram signals. FIG. 13 is a diagram showing a circuit configuration of the detection element. FIG. 14 is a block diagram of a control device to which a plurality of electrocorticogram signals are input. FIG. 15 is a block diagram of a control device to which a plurality of electrocortical signal are input. FIG. 16 is a graph (A) showing the relationship between the input frequency f _in (Hz) and the frequency f (Hz) of the drive signal, and a graph (B) showing the relationship between the input frequency f _in (Hz) and the amplitude A of the drive signal. FIG. 17 is a graph (A) showing the relationship between the input frequency f _in (Hz) and the frequency f (Hz) of the drive signal, and a graph (B) showing the relationship between the input frequency f _in (Hz) and the amplitude A of the drive signal. FIG. 18 is a graph (A) showing the relationship between the input frequency f _in (Hz) and the frequency f (Hz) of the drive signal, and a graph (B) showing the relationship between the input frequency f _in (Hz) and the amplitude A of the drive signal. FIG. 19 is a graph (A) showing the relationship between the input frequency f _in (Hz) and the frequency f (Hz) of the drive signal, and a graph (B) showing the relationship between the input frequency f _in (Hz) and the amplitude A of the drive signal. FIG. 20 is a graph (A) showing the relationship between the input frequency f _in (Hz) and the frequency f (Hz) of the drive signal, and a graph (B) showing the relationship between the input frequency f _in (Hz) and the amplitude A of the drive signal.

Various exemplary embodiments will be described in detail below with reference to the drawings. Note that the same or corresponding parts in each drawing are denoted by the same reference numerals, and duplicated explanations will be omitted.
First Embodiment
FIG. 1 is a diagram showing a deep brain stimulation device 100 according to the first embodiment.

The deep brain stimulation device 100 comprises an EEG detector 1, a drive signal generating device 10, and a deep brain stimulation element 8 to which the drive signal output from the drive signal generating device 10 is applied. In this example, the deep brain stimulation element 8 is a deep brain stimulation probe (electrode) that applies an electrical signal to a target area. The deep brain stimulation element 8 can also be an element that converts an electrical signal into an optical (electromagnetic wave) signal and applies it to the target area.

The drive signal generating device 10 receives the output signal of the EEG detector 1, generates a drive signal, and supplies the drive signal to the deep brain stimulation element 8. The drive signal generating device 10 includes a filter 2 (analog filter), a window discriminator 3, a first interface 4, a control device 5, a second interface 6, and a drive circuit 7 (analog isolator).

The electroencephalogram detector 1 receives an electrocortical electroencephalogram signal from the motor cortex of the cerebral cortex. The output signal of the electroencephalogram detector 1 is input to the filter 2, which extracts signal components in a predetermined frequency band (including the γ2 band). The output signal of the filter 2 is input to the window discriminator 3. The window discriminator 3 generates a pulse signal S _P from the extracted signal components according to the occurrence frequency of events correlated with brain activity. Specifically, the window discriminator 3 generates one pulse voltage and outputs the pulse signal S _P every time one waveform having an intensity equal to or greater than a reference value and a duration equal to or greater than a reference time is detected.

The control device 5 receives the pulse signal output from the window discriminator 3. The control device 5 processes the pulse signal S P to generate a drive control signal S _C according to the frequency of event occurrence. The drive control signal S _C has a frequency and amplitude according to the frequency of the pulse signal S _P. The drive circuit 7 generates a drive signal _{S D} having a frequency and amplitude according to the drive control signal S _C output from the control device 5 and electrically insulated from S _C. The drive signal S _D (stimulation pulse) is synchronized with the drive control signal S _C and is input to the deep brain stimulation element 8. When the drive signal S _D is input to the deep brain stimulation element 8, the subthalamic nucleus STN that the deep brain stimulation element 8 contacts is stimulated by an electrical signal. This stimulation reduces neurological movement disorders and can treat movement disorders. This will be described in detail below.

The following describes the EEG detector 1.

The EEG detector 1 includes detection elements (first detection element 1A, second detection element 1B) arranged in the primary motor area M1 of the motor cortex, a preamplifier 1C to which the output signals of the detection elements are input, and a main amplifier 1D (biological signal amplifier) connected to the rear stage of the preamplifier 1C. In this example, the first detection element 1A and the second detection element 1B form a bipolar electrode. Each of these pair of electrodes is a stainless steel conductor, and is covered with an insulating material, fluororesin (polytetrafluoroethylene), except for the tip. The distance between the pair of electrodes is 2 mm. The pair of electrodes is connected to the positive and negative input terminals of the preamplifier 1C in the amplifier, respectively. The EEG detector 1 in this example uses a bipolar induction detection method, but other detection methods such as unipolar induction can also be used. The EEG detector 1 detects cortical EEG (local field potential (LFP)) in the motor cortex (primary motor area M1) and outputs an EEG signal.

Explain filter 2.

Filter 2 is a bandpass filter that receives the output signal from EEG detector 1, extracts components in the gamma 2 wave frequency band (80 Hz to 200 Hz), and outputs the extracted signal.

The frequency band to be extracted (80 Hz to 200 Hz) is not an absolute range, but rather a frequency band that is highly correlated with brain activity immediately before the movement of a body part may be selected. From this perspective, the upper limit frequency that filter 2 passes is not limited to 200 Hz, but since unnecessary frequency components become noise, it is desirable to block them if not necessary. In addition to gamma 2 waves, a portion of gamma 1 waves (30 Hz to 80 Hz) can also be included in the frequency band extracted by filter 2, but the lower limit frequency that filter 2 passes is set to at least a frequency higher than the upper limit frequency (30 Hz) of beta waves (13 Hz to 30 Hz). In order to clearly distinguish between gamma 2 waves and beta waves, the lower limit frequency that filter 2 passes can be set to 40 Hz or higher, 50 Hz or higher, 60 Hz or higher, or 70 Hz or higher. In order to clearly distinguish between gamma 2 waves and gamma 1 waves, the lower limit frequency that filter 2 passes can be set to 80 Hz or higher, 90 Hz or higher, or 100 Hz or higher.

From this viewpoint, it is preferable to select a center frequency f _C between 80 Hz and 200 Hz as the γ2 frequency band of the cortical electroencephalogram in the primary motor cortex, and to extract a frequency band including the selected center frequency f _C. When the filter 2 extracts a band of the center frequency fc±the allowable frequency deviation Δf, the frequency band is given by fc±Δf.

If the lower limit frequency for passing through filter 2 is 40 Hz and the upper limit frequency is 200 Hz, the frequency band is the center frequency fc (120 Hz) ± the allowable frequency deviation Δf (80 Hz). The width of the frequency band fw is 200 Hz - 40 Hz = 160 Hz.

If the lower limit frequency for passing through filter 2 is 60 Hz and the upper limit frequency is 200 Hz, the frequency band is the center frequency fc (130 Hz) ± the allowable frequency deviation Δf (70 Hz). The width of the frequency band fw is 200 Hz - 60 Hz = 140 Hz.

If the lower limit frequency for passing through filter 2 is 80 Hz and the upper limit frequency is 200 Hz, the frequency band is the center frequency fc (140 Hz) ± the allowable frequency deviation Δf (60 Hz). The width of the frequency band fw is 200 Hz - 80 Hz = 120 Hz.

If the lower limit frequency for passing through filter 2 is 100 Hz and the upper limit frequency is 200 Hz, the frequency band is the center frequency fc (150 Hz) ± the allowable frequency deviation Δf (50 Hz). The width of the frequency band fw is 200 Hz - 100 Hz = 100 Hz.

If the lower limit frequency for passing through filter 2 is 120 Hz and the upper limit frequency is 200 Hz, the frequency band is the center frequency fc (160 Hz) ± the allowable frequency deviation Δf (40 Hz). The width of the frequency band fw is 200 Hz - 120 Hz = 80 Hz.

In these examples, the allowable frequency shift Δf is set by selecting from the range of 40 Hz to 80 Hz. If the center frequency fc is set between 80 Hz and 200 Hz, the allowable frequency shift Δf may be set, for example, to Δf = 30 Hz, Δf = 20 Hz, or Δf = 10 Hz. That is, the allowable frequency shift Δf can be set by selecting from the range of 10 Hz to 80 Hz, but if brain activity can be monitored, a value outside this range can also be selected. The frequency band width fw is exemplified above as 80 Hz to 160 Hz, but brain activity can be detected even if the frequency band width fw is set smaller or larger.

The frequency band γ2 correlated with the movement extracted by filter 2 was applied to monkeys (primates) in the experiment, but it can also be applied to humans (humans), who are also primates.

Explain window discriminator 3.

The window discriminator 3 receives the output signal S _F from the filter 2. During brain activity, nerve cells in the brain generate pulse-like action potentials, and the superposition of these potentials is considered to constitute γ2 of the cortical electroencephalogram. The window discriminator 3 monitors the components of the received signal and determines whether the signal components satisfy the following criterion condition (α):

Reference condition (α): A signal component having an intensity equal to or greater than the reference value (Vth) continues for a period equal to or greater than the reference period (Tth).

When the result of the determination of the reference condition (α) is true, it is assumed that one event related to brain activity has occurred, and one pulse signal _SP corresponding to this is output.

FIG. 2 is a timing chart of the output signal _SF and the pulse signal _SP of the filter 2. As shown in FIG.

There are various types of structure for the window discriminator 3, so an example will be described below. The maximum value of the received signal S _F from the filter 2 is set to 100%. If the window discriminator 3 is equipped with a comparator, a reference value (Vth) (threshold value) of the comparator is set. In this example, Vth is set to 75% of the maximum value of the received signal S _F , but it can also be set to another value. The comparator outputs "1" when the intensity of the signal component of the received signal S _F is equal to or greater than the reference value (Vth).

The output signal of this comparator is sampled periodically in synchronization with the sampling clock signal. If all the outputs of the comparator sampled within the reference period (Tth) are "1", the judgment result of the above-mentioned reference condition (α) is true. That is, it can be judged that the signal component having the intensity equal to or greater than the reference value (Vth) has continued for the reference period (Tth) or more. In this case, the window discriminator 3 judges that the component (waveform BS) contained in the received signal S _F (filtered electroencephalogram signal) satisfies the reference condition (α), and outputs one pulse signal S _P (pulse voltage P) corresponding to this. An exemplary pulse width of one pulse voltage P is 60 microseconds (μs). In this example, the reference period (Tth) is set to 0.5 milliseconds (ms). The judgment circuit that performs the above judgment in the window discriminator 3 can be configured, for example, using a transistor-transistor logic circuit (TTL circuit).

Each time an event (waveform BS) related to brain activity occurs, one pulse voltage P is generated. In the brain, multiple events occur in a time series. Therefore, a pulse signal S _P including multiple pulse voltages P is output from the window discriminator 3. The pulse signal S _P has information on the frequency of signal components (waveform BS of an electroencephalogram signal that satisfies the reference condition (α)) that satisfy the above-mentioned reference condition (α). In short, the number of occurrences (frequency) of the waveform BS that satisfies the reference condition within a certain period of time is found.

The pulse signal S _P output from the window discriminator 3 is input to the control device 5 via the first interface 4. The control device 5 processes the received pulse signal S _P to generate a drive control signal S _C , and inputs it to the drive circuit 7 via the second interface 6. Specifically, the control device 5 generates a drive control signal S C having a frequency and amplitude according to the number of pulse voltages _P included in a unit period (the frequency of signal components that satisfy the reference condition (α) included in the output signal of the filter 2).

Explain the control device 5.

A method for generating the drive control signal S _C in the control device 5 will be further described. The drive signal S _D is synchronized with the drive control signal S _C , and the frequency f of the drive signal S _D is the same as the frequency f of the drive control signal S _C. The amplitude A of the drive signal S _D is proportional to the amplitude A of the drive control signal S _C.

The control device 5 measures the number of pulses of the pulse signal S _P within the judgment period T _P (the number (= frequency) of the pulse voltage P within a unit period). In this example, the judgment period T _P is 100 times the reference period (Tth=0.5 ms) and is set to 50 milliseconds (ms). In this example, the moving average μ TTL of the number of pulses included in the judgment period T _P is used.

When the number of pulses (average value μTTL) in the judgment period T _P increases, in other words, when the frequency of occurrence of brain activity events increases, the control device 5 increases the frequency f and amplitude A of the drive control signal S _C so as to monotonically increase (e.g., proportionally) with respect to the number of pulses. The frequency f saturates when it exceeds an upper limit frequency f _limit (e.g., 150 Hz) and does not increase even if the number of pulses increases.

With respect to frequency f, this relationship can be expressed mathematically as follows:
(Condition 1) If f ₀ <f _limit : f = f ₀
(Condition 2) If f _limit ≦f ₀ : f = f _limit
(monotonically increasing function) _f0 = Threshold _f + (μTTL/Threshold _f ) × Gain _f
The frequency f (and frequency f ₀ ) is a function of the average value μTTL of the number of pulses per unit time. Threshold _f is the frequency of the drive control signal S _C when the average value μTTL of the number of pulses is zero, and for example, Threshold _f = 50 Hz. In this case, if the horizontal axis is μTTL and the vertical axis is frequency f, the frequency f is frequency f ₀ until it reaches μTTL = 5000/Gain _f , and increases at a slope (Gain _f /50). Gain _f is a coefficient that determines the slope of this monotonically increasing function (linear function).

When the number of pulses (average value μTTL) in the judgment period _TP increases, in other words, when the frequency of occurrence of brain activity events increases, the control device 5 increases the amplitude A of the drive control signal _SC so as to monotonically increase (e.g., proportionally) with respect to the number of pulses. The amplitude A saturates when it exceeds an upper limit amplitude A _limit (e.g., 1.5 mA) and does not increase even if the number of pulses increases.

Regarding the amplitude A, this relationship can be expressed mathematically as follows:
(Condition 1) If _A0 < _Alimit , then A= _A0
(Condition 2) If A _limit ≦A ₀ : A = A _limit
(Monotonically increasing function) _A0 = Threshold _A + (μTTL/Threshold _f ) × Gain _A
The amplitude A is a function of the average value μTTL of the number of pulses per unit time. Threshold _A is the amplitude of the drive control signal S _C when the average value μTTL of the number of pulses is zero, and for example, Threshold _A = 0.5 mA. In this case, when Threshold _f = 50 Hz, if the horizontal axis is μTTL and the vertical axis is amplitude A, the amplitude A increases with a slope (Gain _A / 50) until it reaches μTTL = 50 / Gain _A. Gain _A is a coefficient that determines the slope of this monotonically increasing function (linear function). Note that the frequency threshold Threshold _f and the amplitude threshold Threshold _A are each greater than 0, and it is preferable that the drive signal S _D is provided to the deep brain stimulation element 8 even when the subject is in a resting period.

Gain _f and Gain _A are set so that the frequency and amplitude are maintained lower than the predetermined upper limit, and even if the input is maximized in normal use, the output does not reach the upper limit. The drive control signal S _C has a frequency f and an amplitude A, and the duration of the pulse can be set to, for example, 60 microseconds. The above relational expression is a function in which the frequency f and the amplitude A increase linearly according to the pulse generation frequency (average value μTTL), but other function expressions can also be used. That is, it is sufficient that the frequency f and the amplitude A increase monotonically according to the pulse generation frequency (average value μTTL). For example, it is possible that the frequency and the amplitude increase nonlinearly with the increase in the frequency. Even in nonlinear control, stimulating the subthalamic nucleus or the like makes it easier for motor commands to pass, so that movement disorders can be treated. In addition, the frequency and the amplitude can be changed not only continuously but also discretely with respect to the frequency. Even in discrete control, stimulating the subthalamic nucleus etc. makes it easier for motor commands to pass through, making it possible to treat movement disorders.

When the frequency f and amplitude A of the drive control signal S _C (drive signal S _D ) are increased, the power supplied to the deep brain stimulation element 8 is increased, and the amount of stimulation is increased. Although a therapeutic effect can be expected even when only one of these parameters is increased, it is more effective to increase both parameters. When only one parameter is changed, the drive signal generating device 10 extracts the signal components of the frequency band corresponding to the above-mentioned gamma 2 wave, obtains the frequency of the waveform of the signal components that satisfy the reference condition (α), and generates a drive signal S _D having a frequency corresponding to the frequency. Alternatively, the drive signal generating device 10 extracts the signal components of the frequency band corresponding to the above-mentioned gamma 2 wave, obtains the frequency of the waveform of the signal components that satisfy the reference condition (α), and generates a drive signal S _D having an amplitude corresponding to the frequency.

The drive circuit 7 will now be explained.

The drive circuit 7 generates a drive signal S _D synchronized with the input drive control signal S _C and inputs the drive signal S _D to the deep brain stimulation element 8. The higher the frequency of occurrence of events corresponding to brain activity, the higher the frequency and amplitude indicated by the drive control signal S _C , i.e., the supply power. The higher the frequency and amplitude of the drive control signal S _C , the higher the frequency and amplitude of the drive signal S _D , and the higher the power supplied to the deep brain stimulation element 8. The drive signal S _D in this example supplies a drive current to the deep brain stimulation element 8, and is a current signal having the same frequency as the drive control signal S _C, which is a voltage pulse signal, and the amplitude of this current signal is proportional to the amplitude of the voltage pulse signal.

The drive circuit 7 (analog isolator) includes an input section to which the drive control signal S _C is input, and an output section to which the drive control signal S _C is transmitted in an electrically insulated state from the input section and which generates the drive signal S _D. That is, these input section and output section are isolated from each other so that no current flows directly between them. The signal transmission structure between the input section and the output section includes a structure using a transformer and a structure using a photocoupler. The use of an isolator has the advantage that the influence on the drive signal S _D can be suppressed in the event of a malfunction on the input section side. That is, the maximum value of the drive signal S _D can be limited on the output section side, improving safety. It is also possible to adopt a circuit other than an analog isolator as the drive circuit 7.

Explain about the deep brain stimulation element 8.

Figure 3 shows a plan view of the deep brain stimulation element 8 (Figure 3(A)) and an enlarged view of the tip portion (Figure 3(B)).

The deep brain stimulation element 8 has a tip 8A that contacts the deep brain, an input 8B to which a drive signal is input, and a support 8C that connects the tip 8A and the input 8B.

The surface of the support 8C is covered with a coating film 8a made of resin (e.g., polyimide), and a conductor of a high melting point metal (e.g., tungsten) runs inside the coating film 8a.

The tip portion 8A is equipped with a first stimulation portion 8b, a second stimulation portion 8c, and a third stimulation portion 8d, and these stimulation portions are provided on the surface of a cylindrical body 8e made of resin (e.g., epoxy resin). A tip potential detection portion 8f is exposed from the tip of the cylindrical body 8e. In this example, the tip potential detection portion 8f is made of a high impedance electrode (up to 500 kΩ). In this example, the first stimulation portion 8b, the second stimulation portion 8c, and the third stimulation portion 8d are each made of a low impedance electrode (up to 10 kΩ). In this example, the center-to-center distance between the first stimulation portion 8b, the second stimulation portion 8c, and the third stimulation portion 8d is 0.75 mm.

Inside the cylindrical body 8e, there are multiple conductors connected to the first stimulating section 8b, the second stimulating section 8c, the third stimulating section 8d, and the tip potential detecting section 8f. These conductors are electrically insulated from each other and are each connected to multiple conductors that pass through the inside of the coating film 8a.

The first stimulating section 8b, the second stimulating section 8c, and the third stimulating section 8d are made of an electrode material that is less reactive to chemical substances in the body. As the electrode material, a high melting point metal, platinum, or an alloy containing at least one metal selected from these metal materials may be used. As an example, a platinum-iridium alloy may be used as the electrode material. Each stimulating section has a shape formed by wrapping a wire made of an electrode material around the cylindrical body 8e multiple times (e.g., eight times), but may have other shapes as long as it can provide electrical stimulation. The material of the tip potential detecting section 8f is also made of an electrode material that is less reactive to chemical substances in the body, like the material of the first to third stimulating sections. The tip potential detecting section 8f can be used to confirm whether the deep brain stimulating element is located in the target area (subthalamic nucleus STN), and can be used to measure neural activity.

When a drive signal (drive current) is supplied to the first stimulator 8b, the second stimulator 8c, and the third stimulator 8d, electrical stimulation is given to the parts where they are in contact. It is not necessary to supply a drive current to all of the stimulators. That is, after inserting the deep brain stimulation element 8 into the brain, a magnetic resonance imaging (MRI) device is used to measure the positions of these stimulators (electrodes), and two of the three stimulators that are in appropriate positions are selected, and a drive signal is given to the selected stimulators.
(Experimental Example)
An experiment was carried out to verify the therapeutic effect of using the above-described deep brain stimulation device 100.

First, we will explain the subjects and training. Two female Japanese macaques (Macaca fuscata) were prepared as subjects. Monkey A weighed 5.0 kg, and Monkey B weighed 5.4 kg. The monkeys were trained five days a week for four months. After this training, the monkeys were treated with MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine), and an experiment with the same content as the training was conducted. In this training, the monkeys were asked to complete a simple vertical upper limb reaching task.

During training, the monkeys sat in monkey chairs facing a fixed plastic tower. A green light-emitting diode (LED) was embedded in the tower as a visual target. A home key (a horizontal platform) was fixed at the base of the tower, 15 cm below the LED, allowing the monkeys to rest their hands comfortably on the home key with their elbows bent at 90°. To monitor hand movements, infrared photoelectric sensors (manufactured by Keyence Corporation) were attached to the home key and the target LED.

The task began after the monkey placed its hand on the home key for at least 1 second. When the monkey was stimulated (for 0.3 seconds) by a visual (LED light) and an auditory (buzzer) signal, the monkey was required to reach and touch the target within 6 seconds. If the task was successful, the monkey was rewarded with a drop of water 0.1 seconds after touching the target. The monkey was required to return its hand to the home key in 12 seconds. If the monkey was unable to return its hand to the home key, the experimenter placed the monkey's hand on the home key to start the next trial.

In this assignment, four time periods were defined.
The rest period (Rest) is a period during which no task is being performed, specifically the period from one second before the onset of stimulation by visual and auditory signals until the onset of said stimulation.
The premovement period begins at the onset of the visual and auditory cues and ends when the hand is released from the home key.
A movement period begins when the monkey removes his hand from the home key and ends when he touches the target.
The return period began when the monkey released his hand from the target and ended when he returned his hand to or near the home key.

A "series" was defined as 25 trials in which the monkeys successfully reached the target, regardless of whether they returned to the home key. Monkeys were deprived of water for 24 h prior to the start of the experiment, and water intake was regulated by fruit and subcutaneous fluid supply as needed.

Next, we will explain the surgical procedure for head fixation and electrode implantation. Under general anesthesia, a surgical procedure was performed to fix the pipes to the skull. This procedure was performed to fix the head in a stereotaxic frame during the experiment. The anesthetics used were ketamine hydrochloride (5-8 mg/kg (body weight), intramuscular injection), xylazine hydrochloride (0.5-1 mg/kg (body weight), intramuscular injection), and propofol (6-7 μg/ml, blood concentration, intravenous injection).

Ten days after this surgery, a detection element (electrode) for detecting electrocortical electroencephalogram signals was implanted in the head. Under general anesthesia, the head was fixed in a stereotaxic frame, and then the skull above the primary motor cortex M1 on the opposite side to the hand used to perform the task was removed. Ketamine hydrochloride (5-8 mg/kg (body weight), intramuscular injection) and xylazine hydrochloride (0.5-1 mg/kg (body weight), intramuscular injection) were used as anesthetics. The upper limb area of the primary motor cortex M1 was identified by electrophysiological mapping.

The electrodes, which served as the detection elements for the EEG signals, were made of 200 μm diameter polytetrafluoroethylene-coated stainless steel wires (California Fine Wire). The wires were paired and impedance matched (≤1 kΩ) to ensure optimal common-mode rejection. For subsequent recording of local field potentials (LFPs), the wires were implanted in layers III-V of the upper limb region of the primary motor cortex M1 with a center-to-center distance of 2 mm and mechanically fixed with acrylic resin. A rectangular plastic chamber covering the primary motor cortex M1 was fixed to the skull with acrylic resin.

Induction of Parkinson's disease. Monkeys were administered MPTP to induce moderate to severe Parkinson's disease. After induction of general anesthesia, the external carotid artery, internal carotid artery, and common carotid artery on the opposite side to the hand used for performing the task were incised in the neck region. The anesthetic agents used were ketamine hydrochloride (5-8 mg/kg (body weight), intramuscular injection), xylazine hydrochloride (0.5-1 mg/kg (body weight), intramuscular injection), and propofol (6-7 μg/ml, blood concentration, intravenous injection). The external carotid artery was temporarily clamped, and MPTP (2 mg/mL) dissolved in saline was injected into the common carotid artery (Monkey A: 0.6 mg/kg, Monkey B: 1.3 mg/kg).

　Subsequently, under general anesthesia with ketamine hydrochloride (5-8 mg/kg body weight, intramuscular injection) and xylazine hydrochloride (0.5-1 mg/kg body weight, intramuscular injection), further MPTP was injected intramuscularly (Monkey A, 0.6 mg/kg (single injection), 5 injections) and further MPTP was injected intravenously via the saphenous vein (Monkey B, 0.8 mg/kg and 0.7 mg/kg, 2 injections). The total dose of MPTP administered to Monkey A was 3.6 mg/kg and to Monkey B was 2.8 mg/kg.

Electrophysiological recordings of parkinsonian states were started 5-8 weeks after the last MPTP injection, when parkinsonian symptoms had stabilized for 2 weeks. The severity of parkinsonian symptoms was assessed using the Primate Parkinson's Disease Rating Scale (maximum score of 20 points indicates the worst symptoms) on the side opposite to the MPTP injection into the carotid artery. Monkeys responded to treatment with L-dopa and carbidopa (200-500 mg/day, orally).

　We describe cortical electrophysiological recordings and their conversion into stimulation pulses. Behavioral tasks and electrophysiological recordings were performed in a drug-off state (at least 12 hours after withdrawal). The monkey's head was restrained in a soundproof room shielded with copper netting to block external electromagnetic noise, and the monkey was seated in a monkey chair. A head amplifier (Blackrock Microsystems, Cerebus (registered trademark)) was placed 10 cm away from the recording chamber, and a direct current (12 V) power supply was used to amplify local field potentials (LFPs) from the cortical recording electrode in the primary motor cortex M1. The amplifier output signal was transmitted to two separate circuits.

The first circuit uses a brain-machine interface and is the online processing system shown in Figure 1. This circuit modulates the control parameters of the stimulation pulses (STN-DBS parameters) as a function of γ2 activity from the primary motor cortex M1.

The second circuit is the offline processing system. This circuit is a data recorder that digitizes and stores data from the headstage, brain-machine interface, deep brain stimulator, and behavioral device for offline analysis of the experimental results.

In the brain-machine interface, the local field potential (LFP) measured in the primary motor cortex M1 was amplified (10,000 times, frequency band 50-300 Hz) using a biosignal amplifier (Nihon Kohden Corporation, product name MEG-6116/AB610J), and the amplifier's output signal was then input to a filter (NF Corporation, product name FV-664) to extract the signal components indicating γ2 activity (80-200 Hz).

Before starting the first trial series, local field potentials (LFPs) were recorded for 15 min in the primary motor cortex M1 while the monkey was moving its arms, legs and trunk freely. This allowed the determination of the maximum oscillation amplitude of the gamma 2 band signal. A window discriminator (Nihon Kohden Corporation, product name EN-611J) detected gamma 2 band oscillations (components with an amplitude greater than 75% of the maximum oscillation amplitude of the gamma 2 band and a duration greater than 0.5 milliseconds (ms)). For further processing, the pulse signal S _P (transistor-transistor logic (TTL) pulse) generated by the window discriminator as a function of the gamma 2 band activity was sent to the control device (computer) via a first interface (digital input/analog output card: PCI6713 (National Instruments)).

In the control device, as described above, the instantaneous frequency of the γ2 drive pulse was moved and averaged over a 50 ms window to modulate the frequency f and amplitude A of the drive control signal S _C (drive signal S _D ) for deep brain stimulation. The drive control signal S _C generated in the control device is a pulse signal having a frequency f, an amplitude A, and a pulse width of 60 μs. This pulse signal was input to a drive circuit (analog isolator: Dagan, BSI-950 (product name)) via a second interface (digital input/analog output card: PCI6713 (manufactured by National Instruments)), and output from the drive circuit as a drive signal S _D , _which was supplied to the deep brain stimulation element.

The pulse width of the drive control signal S _C (drive signal S _D ) was set to 60 microseconds (μs) in both aDBBS and cDBS. This pulse width does not cause artifacts in the local field potential (LFP) of the primary motor cortex M1 and has been applied clinically. Because there is debate about the clinical effects associated with low-frequency stimulation, the frequency range of the drive control signal S _C (drive signal S _D ) was set to 50-150 Hz. The amplitude A of the drive signal S _D (stimulation pulse, drive pulse current) in the experiment was 1 mA in cDBS and 0.5-1.5 mA in aDBS.

In addition, preliminary experiments on amplitude optimization showed that optimal performance was obtained when a drive signal _SD of constant frequency and constant amplitude (amplitude A = 1 mA) was applied to a deep brain stimulator implanted in a monkey. Performance decreased when amplitude A exceeded 2 mA, and muscle contractions and dyskinesias were observed when amplitude A was 3 mA or higher. These side effects are interpreted as being due to current spreading to adjacent regions of the subthalamic nucleus (STN), such as the internal capsule.

In the experiment, the therapeutic effect was measured under the following conditions (a) to (c). In the constant deep brain stimulation (cDBS), the frequency of the drive signal supplied to the deep brain stimulation element was 100 Hz, the amplitude was 1 mA, and the pulse width was 60 microseconds (μs).
(a) Condition without deep brain stimulation (DBS-OFF).
(b) Conditions for performing adaptive deep brain stimulation (aDBS).
(c) Condition for constant deep brain stimulation (cDBS).

A deep brain stimulator (deep brain stimulation probe (electrode)) as shown in Figure 3 was prepared and inserted vertically or obliquely (inclined 36 degrees forward from the vertical in the monkey's sagittal plane) into the subthalamic nucleus (STN) along the same trajectory in each experiment. Multi-unit recording was performed from the high-impedance tip potential detection unit located at the tip of the deep brain stimulator, and the sensory response to passive joint movement of the arm was recorded to identify the dorsolateral motor subregion. Two adjacent stimulating units (contact electrodes) included in the deep brain stimulator were placed in this motor region, and bipolar stimulation was performed through these stimulating units.

We explain the post-processing of local field potentials (LFPs) recorded from the primary motor cortex.

For offline analysis, local field potentials (LFPs) from the primary motor cortex were recorded as bipolar signals and signal components in the frequency band from 3 to 200 Hz were extracted using a double reverse Butterworth filter, which has a characteristic of -3 dB at 200 Hz with zero phase shift. The extracted signals were stored in the memory of the control device (computer) and downsampled to 500 Hz. The signals were checked by the experimenter, and those containing artifacts longer than 3% in duration were excluded from the analysis.

First, local field potentials (LFPs) from the primary motor cortex were analyzed under conditions without deep brain stimulation (DBS-OFF) to define the behavior in the beta, gamma1, and gamma2 bands. We hypothesized that task-related changes in oscillatory activity would be frequency band specific, small in amplitude, and highly dynamic. A dynamic autoregressive model based on the Kalman smoother was used to track the instantaneous frequency bands of local field potentials (LFPs) from the primary motor cortex (beta band (13-30 Hz), gamma1 band (30-80 Hz), gamma2 band (80-200 Hz)).

　We will explain the measurement results under DBS-OFF conditions.

Figure 4 is a graph showing the change over time in activity in the beta, gamma 1, and gamma 2 bands. Figure 4(A) shows data for monkey A and monkey B combined, Figure 4(B) shows data for monkey A, and Figure 4(C) shows data for monkey B.

The activity of the primary motor cortex in the beta, gamma1, and gamma2 bands represents the cortical EEG signal (voltage) in each frequency band, and changes during the task periods (rest period (Rest), premovement period (Premovement), movement period (Movement), and return period (Return)) under DBS-OFF conditions. These four task periods were normalized in time, with each period set to 100 time units and linked on the graph (400 time units in total). The activity of the beta, gamma1, and gamma2 bands was calculated using a Kalman filter and normalized using the median value during the rest period (Rest). The data in the middle row of the graph indicates the median, and the area between the data in the upper and lower rows indicates the interquartile range (25th to 75th percentile). Additionally, the partially interrupted horizontal line at the bottom during the premovement to return periods indicates periods of significant change (P value < 0.05) compared to the data during the rest period.

FIG. 5 is a chart showing the success rate of each task under the conditions of no deep brain stimulation (DBS-OFF), adaptive deep brain stimulation (aDBS), and constant deep brain stimulation (cDBS). Data for monkeys A and B combined (upper row), monkey A's data (middle row), and monkey B's data (lower row) are shown. The chart shows the percentage (success rate, %) of attempts in which the target was reached (Reach) and attempts in which the target was returned to the home key (Return). As shown in this chart, when the adaptive deep brain stimulation (aDBS) condition of this embodiment was used, the success rate of attempts in which the target was reached (Reach) was almost equal to the success rate under the constant deep brain stimulation (cDBS) condition. In other words, when the signal component of the gamma 2 frequency band in the primary motor cortex was extracted and a driving signal for stimulation was generated as described above (aDBS), an effective therapeutic effect against Parkinson's disease was obtained. Furthermore, aDBS consumes less power than cDBS.

Pulse train: For each period, the pulse train of the drive control signal (drive signal S _D ) is characterized by the time between pulses and its amplitude. The central tendency and variance of these quantities were used to evaluate the variation of the stimulation train between periods and DBS paradigms. The central tendency was estimated by the median and the variance by the median absolute deviation. The delivered charge Q, expressed in coulombs (C)/sec, is delivered by the isolator and is calculated as Q=intensity×pulse width×frequency.

Statistical analysis of the data is described. All data are reported as median and interquartile range (25th to 75th percentiles). Comparisons between data from different conditions were performed using nonparametric Kruskal-Wallis and Mann-Whitney tests for pairwise comparisons adjusted by the Benjamini-Hochberg correction (P value < 0.05). The Page test was applied to the time series of instantaneous frequencies to test the null hypothesis that the ordered series is monotonic (i.e., no trend) and the alternative hypothesis that the series increased with a P value = 0.95. The Benjamini-Hochberg correction is exemplified in "Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: A practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B (Methodol.) 57, 289-300 (1995)". The Kruskal-Wallis test and Mann-Whitney test are exemplified in "Campbell, R. C. "Statistics for Biologists" (Cambridge University Press, 1989)". The Page test is exemplified in "Page, E. B. Ordered hypotheses for multiple treatments: A significance test for linear ranks. J. Am. Stat. Assoc. 58, 216-230 (1963)".

Next, we will further explain the experimental results.

First, we describe the severity of parkinsonian symptoms and task performance. After MPTP treatment, the two monkeys showed cardinal signs of parkinsonian syndrome, including akinesia, bradykinesia, rigidity, and flexed posture, which were more severe on the contralateral side of the carotid MPTP injection. Primate Parkinson's Rating Scale scores remained stable throughout the experimental period. Both monkeys responded to oral administration of L-dopa and carbidopa (L-dopa: 200 mg/day) with a decrease in parkinsonian scores (Monkey A: 4, Monkey B: 6). On the other hand, hyperkinetic symptoms were observed at higher doses (L-dopa > 500 mg/day). The two subjects were able to perform 1 to 6 series of 25 trials in one session. The dataset includes 2583 trials (Monkey A: 1112, Monkey B: 1471) performed under DBS-OFF conditions.

As shown in Figure 5, the average rate of successful target arrival was 87.5% (Monkey A: 92.2%, Monkey B: 86.6%), and the average rate of successful return to the home key was 61.0% (Monkey A: 90.0%, Monkey B: 55.7%). The median premovement period was 546 ms with an interquartile range of 420-833 ms, the median movement period was 408 ms with an interquartile range of 292-632 ms, and the median return period was 998 ms with an interquartile range of 627-1001 ms.

Next, we will explain the local field potentials (LFPs) of the primary motor cortex M1 during the task under DBS-OFF conditions. First, we investigated which of the frequency bands, beta, gamma1, and gamma2, in the primary motor cortex M1 can be used as biomarkers of movement. Instantaneous frequency analysis of the local field potentials of the primary motor cortex (M1-LFPs) under DBS-OFF conditions showed that the power of the beta, gamma1, and gamma2 bands changed along the task period (Figure 4).

Beta band activity tended to decrease during the premovement and movement periods (Page test, P value > 0.99), but not during the return period (Page test, P value < 0.01) (Figure 4).

The activity of the gamma 1 band did not show a stable monotonic trend during the premovement period (Page test, P value < 0.01), but showed a decreasing trend during the movement period (Page test, P value > 0.99), and no trend during the return period (Page test, P value < 0.55) (Figure 4).

The activity of the gamma 2 band showed a tendency to increase during the premovement and movement periods (Page test, P value > 0.99), but showed a tendency to decrease during the return period (Page test, P value > 0.99) (Figure 4).

When examining the power spectral density during each task period, the median power in the beta band decreased during the premovement, movement, and return periods (P value < 0.001).

The median gamma 1 band power for both monkeys combined did not change during the premovement period but increased during the movement and return periods, whereas the median gamma 1 power for monkey B did not change during the movement period (P value > 0.5).

The median gamma-2 band power increased during the premovement, movement, and return periods in each monkey (P value < 0.0002). These dynamic changes and the median gamma-2 band power suggest that gamma-2 band activity in the primary motor cortex M1 (M1-gamma-2 activity) is a suitable biomarker for movement in the parkinsonian state. With the exception of gamma-1 band activity, the Kalman filter and fast Fourier transform (FFT) analyses were consistent.

Next, we will explain the pulse interval and amplitude of aDBS. In aDBS, the pulse interval and amplitude were modulated within the task period (both features, P value < 0.001).

During the rest period (Rest), the pulse interval was 0.012 seconds (interquartile range: 0.011 seconds to 0.016 seconds) and the amplitude was 0.653 mA (interquartile range: 0.488 mA to 0.801 mA).

During the premovement and movement periods, the pulse interval decreased (-12.05%, -13.61%, P < 0.001), while the amplitude increased during these periods (+23.55%, +25.09%, P < 0.001). Compared to the movement period, the pulse interval during the return period increased (+4.35%, P < 0.001), but was lower than that observed during the rest period (-9.68%, P < 0.001). During the return period, the pulse amplitude decreased (-4.80%, P < 0.001), but was higher than that during the rest period (+19.35%, P < 0.001).

The variance of the pulse interval and the variance of the amplitude were also modulated during the task period. The variance of the pulse interval was decreased during the premovement period (-29.41%, P value < 0.001), during the movement period (-64.71%, P value < 0.001), and increased during the return period (50.01% increase compared to the movement period, P value < 0.001), which remained higher than the rest period (29.41%, P value < 0.001).

The variance of the stimulation pulse amplitude decreased during the premovement period (-13%, P value < 0.001) and during the movement period (-38%, P value < 0.001). The variance of the pulse amplitude increased during the return period (33%, P value < 0.001) and remained 10% lower than the value during the rest period (P value < 0.001). The changes in the central tendency and the variance of the interstimulus interval and amplitude were similar in the two monkeys.

Figure 6 shows the power of the gamma 2 band output signal during the premovement, movement, and return periods (Figure 6(A) shows data for monkeys A and B combined, Figure 6(B) shows data for monkey A, and Figure 6(C) shows data for monkey B). The power spectral density of the gamma 2 band during the motor task is shown. The top panel shows aDBS, and the bottom panel shows cDBS. Data are expressed as a percentage of the median during the rest period. The median is shown in the center of the box plot, and the upper and lower positions indicate the interquartile range (25th to 75th percentile). The + symbol in the figure means that the value is significantly different from the premovement period (P value < 0.05), and the diamond symbol means that the value is significantly different from the movement period (P value < 0.05).

Activity in the gamma 2 band in the primary motor cortex M1 increased during the movement period and decreased during the return period with aDBS and cDBS (P value < 0.001) (Figures 6 (A), (B), (C)).

We now explain the behavioral changes caused by aDBS and cDBS. The success rate of reaching the target was similar for aDBS and cDBS (91.5%, Figure 5), and the success rate of returning to the home key was higher for cDBS (75.9% (aDBS), 80.7% (cDBS), Figure 5). Monkey B influenced these results.

Figure 7 is a graph showing the normalized time required to complete each task (premovement, movement, return) in DBS-OFF, aDBS, and cDBS states. The time required to complete a movement in DBS-OFF is set to 100%. The range in parentheses indicates the interquartile range (25th to 75th percentile). Regarding the P values in the figure, P1 is the P value when post hoc comparisons are applied to the DBS-OFF case, and P2 is the P value when post hoc comparisons are applied to the cDBS case.

In aDBS, the premovement and movement periods were shortened (-20.1% and -15.8%, respectively, P value < 0.001), but the return period did not change (P value > 0.05).

In cDBS, the premovement and movement periods were shortened (-12.0% and -10.0%, P value < 0.001), and the return period was increased (+6.5%, P value < 0.001).

aDBS and cDBS are clinically beneficial in the premovement and movement periods. aDBS is more effective in the premovement period (8.1% difference, P value < 0.001) and in the movement period (5.8% difference, P value < 0.001) compared to cDBS.

The difference in clinical efficacy between aDBS and cDBS was evident in the return period, which increased in monkey B (33.2%, P value < 0.001) but did not increase significantly in monkey A (P value > 0.05).

Figure 8 is a graph showing the ratio of charge supplied by aDBS to charge supplied by cDBS.

During the cDBS period, the charge supplied by the isolator was 6 μC/sec. The aDBS supplied only 53.7% (26.5%-71.8%, P value < 0.001) of the supply during the rest period (Rest) of the cDBS, and 75.7% (69.7%-106.6%, P value < 0.05) of the supply during the movement period (Movement) of the cDBS. That is, the aDBS can reduce power consumption significantly more than the cDBS. Note that the aDBS also consumes a small amount of power during the rest period (Rest).

Both aDBS and cDBS improved task performance more than DBS-OFF. aDBS delivered less charge than cDBS, but achieved the same therapeutic effect. The frequency (interval between stimulation pulses) and amplitude of the drive signal during task execution were adjusted. During the task execution period, the interval between stimulation pulses was minimized, the amplitude of stimulation pulses was maximized, and the variance of the interval and amplitude of stimulation pulses was minimized.

It is preferable to minimize the possibility of side effects. In this monkey study, the location of the stimulation site (electrode position of the deep brain stimulator) in the dorsal motor area of the subthalamic nucleus was confirmed every time using brain coordinate guidance, electrophysiological mapping, and responses to sensory and motor stimulation. Furthermore, according to previous findings, when a drive signal of 150 Hz and 2 mA or more is supplied by cDBS, side effects may occur, so the frequency of cDBS was limited to 100 Hz and the amplitude to 1 mA. Similarly, the frequency of aDBS was limited to 50 Hz to 150 Hz and the amplitude to 0.5 mA to 1.5 mA. In addition, the pulse width of the drive signal was set to 60 μs in both aDBS and cDBS. This pulse width is commonly used in clinical applications of DBS, but this short pulse width may cause slight artifacts in DBS of local field potentials in the primary motor cortex. For wider pulse widths (>200 μs), additional processing to avoid artifacts may be warranted in a closed loop system.

The above treatment improved motor dysfunction due to Parkinson's disease, improving reaction and movement times. The clinical effects of DBS are known to exhibit hysteresis. Simply switching DBS from OFF to ON delays changes in tremor, rigidity, and bradykinesia by several seconds or minutes, despite immediate changes in basal ganglia circuit dynamics. Thus, these delays may be reduced when the driving signal provided by DBS is modulated rather than switched between ON and OFF states. In aDBS, stimulation during periods prior to voluntary body part movement, i.e., rest and premovement periods, is useful.

In contrast to cDBS, in aDBS, the pulse interval (frequency) and amplitude of the drive signal have a variance. This variance is the basis for irregular pattern generation, which allows clinical utility while reducing the charge supply compared to cDBS. In aDBS, power consumption can be reduced when the patient is not engaged in motor activity, thereby extending battery life.

Varying the frequency of the drive signal can have a variety of effects. Parkinson's symptoms depend on the frequency of stimulation to the subthalamic nucleus. For example, bradykinesia responds nonlinearly to stimulation across multiple different frequencies. The frequency of the drive signal can be multiplexed to include multiple frequencies. Other possible alternatives include surrogate aDBS techniques. In these alternative aDBS techniques, the surrogate drive signal has similar statistical properties to the original drive signal (pulse train), but is not correlated with the temporal biomarkers. Such surrogate drive signals (pulse trains) can be generated by shuffling the pulse intervals of the pulse trains in the original drive signal.

As described above, the above-mentioned deep brain stimulation device supplied a driving signal with a higher pulse amplitude and a higher pulse frequency to the deep brain stimulation element during the premovement and movement periods (Movement) compared to the rest period (Rest). Also, aDBS supplied only 2/3 of the charge compared to cDBS, but was able to obtain sufficient therapeutic results.
Second Embodiment
FIG. 9 is a diagram showing a deep brain stimulation apparatus according to the second embodiment.

Compared to the deep brain stimulation device of the first embodiment, the deep brain stimulation device of the second embodiment has a filter and window discriminator realized by a digital processing configuration, and the other configurations are the same as those of the first embodiment.

That is, the drive signal generating device 10 in the second embodiment includes a control device 5 that digitizes and inputs the output signal from the brain wave detector 1, and a drive circuit 7 that generates a drive signal _SD having a frequency f and an amplitude A according to the drive control signal SC output from the control device 5. The brain wave detector 1 outputs an analog brain wave signal _SDET , and the first interface 4 converts the brain wave _{signal SDET} into a digital signal and inputs it to the control device 5.

The control device 5 is preferably a computer equipped with a central processing unit and a storage device, and the central processing unit processes the digital signals according to the procedures instructed by the software (program) stored in the storage device.

The control device 5 is equipped with a digital filter 5A that extracts signal components in the above-mentioned frequency band (including the γ2 band) from the digitized signal. The frequency band extracted by the digital filter 5A is the same as the frequency band extracted by the filter 2 (analog filter) shown in the first embodiment. In other words, the software of the digital filter 5A stored in the memory device of the control device 5 digitizes the brainwave signal, which is the input signal, performs filtering, and extracts components in the desired frequency band.

The control device 5 includes a measurement unit 5B that measures the frequency of signal components (waveform BS of electroencephalogram signal that satisfies the reference condition (α)) that satisfy the above-mentioned reference condition (α) from the signal components extracted by the digital filter 5A. In short, the measurement unit 5B determines the number of occurrences (frequency) of the waveform BS that satisfies the reference condition (α) within a certain period of time. The control device 5 includes a signal generation unit 5C that generates a drive control signal S _C having a frequency f and an amplitude A according to the above-mentioned frequency measured by the measurement unit 5B. The function of the measurement unit 5B is the same as that of the window discriminator 3 in the first embodiment, except that it processes digital signals. In other words, the software of the measurement unit 5B stored in the storage device of the control device 5 performs the same processing as the window discriminator 3 on the input electroencephalogram signal of a specific frequency band, and outputs the occurrence frequency of events correlated with brain activity in a time series.

The deep brain stimulation device 100 can reduce the number of parts by adopting a configuration that performs digital signal processing. More specifically, in this configuration, the EEG signal output as an analog signal from the EEG detector 1 is AD converted by an analog-to-digital converter (AD conversion) that serves as an interface. The waveform of the signal components in the digital signal is a discrete value, and has amplitude information that is continuous in a time series. If one data block has time information and amplitude information, the discrete value waveform is made up of a group of data blocks that are continuous in a time series. A program stored in the memory device of a control device such as a computer is executed by a central processing unit, and signal processing is performed, for example, by the following steps.

In the first step, the function of the digital filter 5A is used to extract signal components in the above-mentioned frequency band.

In the second step, the measurement unit 5B executes the function of a window discriminator. That is, it counts the number of consecutive data blocks in a chronological order that have an amplitude equal to or greater than a reference value (Vth), and when the count value reaches or exceeds a value corresponding to a reference period (Tth), it determines that an intracerebral event has occurred, and outputs a judgment output (True) of "1" corresponding to one pulse voltage. The number of judgment outputs (True) measured during a judgment period (e.g., 50 ms) indicates the degree of activity that correlates with intracerebral movement.

In the third step, the signal generating unit 5C substitutes the input number of judgment outputs (activity correlated with intracerebral movement, in other words, the occurrence frequency of a waveform satisfying the reference condition (α)) into a predetermined calculation formula and outputs the calculation result. That is, the occurrence frequency of a waveform satisfying the reference condition (α) (moving average μTTL of the number of pulses per unit time) is input into the formula represented by the above-mentioned (Condition 1), (Condition 2), and (monotonically increasing function), and a frequency f and an amplitude A corresponding to this frequency are output. The signal generating unit 5C generates a drive control signal S _C having the calculated frequency f and amplitude A, and outputs it to the drive circuit 7.

The function of the signal generating unit 5C is the same as the function of generating the drive control signal S _C in the control device 5 in the first embodiment. The drive control signal S _C output from the control device 5 via the second interface 6 is input to a drive circuit 7 (analog isolator) and supplied to the deep brain stimulation element 8 as a drive signal S _D.

The deep brain stimulation device of the second embodiment performs digital signal processing, which allows the device size to be reduced. This is therefore useful when the deep brain stimulation device is applied to medical equipment. Next, a specific example of the deep brain stimulation device according to the first and second embodiments applied to medical equipment will be described.

Figure 10 shows a deep brain stimulation device as a medical device.

This deep brain stimulation device is a device in which the above-mentioned drive signal generating device 10 is housed in a sealed container BOX. As described above, the EEG detector 1 includes detection elements (first detection element 1A, second detection element 1B) arranged in the primary motor area M1 of the cerebral cortical motor area, a preamplifier to which the output signals of the detection elements are input, and a main amplifier. The first detection element 1A and the second detection element 1B are each a plate-shaped chip electrode, and are shaped to have little effect on the brain. The first detection element 1A and the second detection element 1B are connected to the input terminal of the preamplifier via the first signal line W1. The EEG signal S _B (voltage signal) output from the first detection element 1A and the second detection element 1B is transmitted through the first signal line W1 and input to the preamplifier.

The sealed container BOX contains a drive signal generating device 10 and a preamplifier and a main amplifier in the electroencephalogram detector 1. A second signal line W2 extending to the outside of the sealed container BOX is connected to an output terminal of a drive signal _SD of the drive signal generating device 10. The second signal line W2 is connected to a deep brain stimulation element 8.

The deep brain stimulation element 8 in this example is made of a soft material. The deep brain stimulation element 8 shown in FIG. 3 has a conductor wire made of a high melting point metal coated with a resin such as polyimide. A thin conductor wire made of a high melting point metal may be passed through the inside of a soft silicone resin tube, and a plurality of stimulation parts may be formed on the outer circumferential surface of the tip of the tube. The material of the stimulation part may be a conductive material with low reactivity, for example, a platinum-iridium alloy. The stimulation part may be a thin film. When the stimulation part is made of a material that generates a light (electromagnetic wave) signal, it is also possible to embed a semiconductor light emitting element inside the tube. The conductor wire inside the tube is connected to the stimulation part, and a drive signal S _D is supplied.

The first signal line W1 is electrically connected to the first detection element 1A and the second detection element 1B in a physically continuous manner, but if the EEG signal is AC, it is also possible to connect them in an electrically isolated manner via a capacitor. By differentially amplifying the output signal from the first detection element 1A and the signal from the second detection element 1B, it is possible to reduce noise and detect weak signals. In the case of a wireless connection, it is also possible to store a very small wireless transmission amplifier inside the skull and connect the first detection element 1A and the second detection element 1B to the wireless transmission amplifier. It is also possible to supply power to the wireless transmission amplifier using induced electromotive force or ultrasound. The second signal line W2 can be connected to the deep brain stimulation element 8 in the same way as the first signal line W1.

The material of the sealed container BOX is not particularly limited, but it can be made of fluororesin or general resin materials. The sealed container BOX contains a drive signal generating device that performs electrical signal processing, and can protect these devices. It is also possible to implant the sealed container BOX inside the body. The drive signal generating device 10 and the amplifier can be made of semiconductor integrated circuits. The analog isolator used as the drive circuit can also be made of a digital isolator. Furthermore, since semiconductor integrated circuits can be made extremely small, it is also possible to place some or all of these devices inside the skull. The control device can be a computer, but it can also be made of a logic circuit, or an application specific integrated circuit (ASIC) can also be used.

As described above, the deep brain stimulation device 100 detects the patient's voluntary movement from the gamma 2 band electrocortical electroencephalogram signal, generates a drive signal, and uses adaptive deep brain stimulation (aDBS) to stimulate the deep brain. Normally, when a specific movement command exceeds the functional threshold of the internal pallidum (GPi), the intended movement is performed and other movement plans are blocked. In Parkinson's disease, the decision threshold of the internal pallidum increases, and all movement plans are blocked. Although activity in the gamma 1 band (30 Hz < frequency f) can also be used as an electrophysiological biomarker, the above embodiment uses activity in the gamma 2 band (80 Hz to 200 Hz) in the primary motor cortex M1. It is believed that the deep brain stimulation device 100 detects the advance generation of a movement plan in the primary motor cortex M1, lowers the functional threshold of the internal pallidum, and increases the selectability of the movement plan in this time window.

Figure 11 shows a deep brain stimulation device equipped with multiple detection elements.

The following detection elements are arranged for the left side of the brain. That is, the first left detection element S1L is arranged in the primary motor area M1. The second left detection element S2L is arranged in the supplementary motor area SMA. The third left detection element S3L is arranged in the premotor area PM. Similarly, the detection elements are arranged for the right side of the brain in the same corresponding relationship as the left side. The output signals of these detection elements are input to a processing circuit 101 including an amplifier and a drive signal generating device 10. The processing circuit 101 has an internal circuit 101A arranged inside the skull 102 and an external circuit 101B arranged outside the skull 102. Each circuit element in the processing circuit 101 is allocated and arranged in the internal circuit 101A and the external circuit 101B. A circuit with a large volume, such as a battery, can be arranged in the external circuit 101B. The drive signal output from the processing circuit 101 is input to the deep brain stimulation element 8.

Figure 12 shows the circuit configuration for receiving and processing multiple EEG signals.

The deep brain stimulation device of this example has multiple detection elements. The detection element group of this example consists of a first left detection element S1L, a second left detection element S2L, a third left detection element S3L, a first right detection element S1R, a second right detection element S2R, and a third right detection element S3R. These detection elements have the same structure. In the case of a unipolar induction structure, each detection element has a single detection electrode 1S and an indifferent electrode 1RE that provides a reference potential φRef. Each detection element may have a bipolar induction structure.

The output signal of each detection element is input to an EEG detector 1 equipped with an amplifier. In the figure, multiple EEG detectors 1 equipped with an amplifier for each input signal are shown, but a single EEG detector 1 equipped with multiple input terminals and multiple output terminals and having an amplification function for each input signal can also be used. The output signal from each detection element is amplified by the EEG detector 1 and input to a drive signal generating device 10 as a cortical EEG signal. The drive signal generating device 10 is part of a processing circuit 101. A control device 5 in the processing circuit 101 generates a drive control signal based on the multiple input cortical EEG signals. The generated drive control signal is input to a drive circuit 7. The drive circuit 7 supplies a drive current to a deep brain stimulation element 8.

Figure 13 shows the circuit configuration of the detection element.

Detection element S1 represents one of the structures of the six detection elements described above. In a preferred embodiment, detection element S1 has a single detection electrode 1S for detecting brain waves and multiple reference electrodes 1R, and has a unipolar induction configuration. In this embodiment, the multiple reference electrodes 1R are arranged to surround the single detection electrode 1S. The number of reference electrodes 1R may be one. Each reference electrode 1R is connected to a resistor Z and wired. The connection position becomes the node that constitutes the indifferent electrode 1RE.

As described above, the EEG detector 1 comprises a detection electrode 1S for detecting EEG, a plurality of reference electrodes 1R arranged around the detection electrode 1S, and a plurality of resistors Z each having one end connected to the plurality of reference electrodes 1R. The EEG detector 1 comprises an amplifier having a reference potential input terminal to which a reference potential φRef of the node to which the other ends of the plurality of resistors Z are connected is input, and a detection potential input terminal to which a detection potential φ(S1) (local potential that becomes a cortical EEG signal) from the detection electrode 1S is input. The indifferent electrode 1RE with this structure has a stable potential, so noise contained in the cortical EEG signal is reduced, allowing for even more stable control. The EEG detector 1 outputs a cortical EEG signal S(S1).

FIG. 14 is a block diagram of a control device 5 to which multiple EEG cortical signals are input.

The basic functions of the digital filter 5A, the measuring unit 5B, and the signal generating unit 5C are as described in FIG. 9. In response to the output signals from each detection element, the EEG detector 1 (see FIG. 12) outputs EEG signals (cortical EEG signals). That is, in response to the output signals from the first left detection element S1L, the second left detection element S2L, the third left detection element S3L, the first right detection element S1R, the second right detection element S2R, and the third right detection element S3R, the EEG detector 1 outputs the first left EEG signal S(S1L), the second left EEG signal S(S2L), the third left EEG signal S(S3L), the first right EEG signal S(S1R), the second right EEG signal S(S2R), and the third right EEG signal S(S3R).

Each EEG signal is converted to a digital value by an AD converter and input to the digital filter 5A. When the input EEG signals are analog signals, the digital filter 5A extracts the signal components in the gamma 2 band and inputs them to the measurement unit 5B. The measurement unit 5B executes the function of the window discriminator described above. If a component corresponding to the occurrence of a brain event is detected in the input EEG signals, the measurement unit 5B outputs a judgment output of "1". If a component exceeding a threshold continues for a reference period or longer in the EEG signal, it is judged that a brain event has occurred.

The threshold determination unit 5C1 sequentially receives the determination outputs of multiple (e.g., six) channels output from the measurement unit 5B. The threshold determination unit 5C1 counts the number of multiple determination outputs "1" corresponding to the occurrence of a brain event during the determination period. In this example, six count values corresponding to six EEG signals are obtained. The count values corresponding to the left and right EEG signals are CL1, CL2, CL3, CR1, CR2, and CR3. When one of these count values (e.g., CL1) exceeds a threshold THC set in correlation with brain activity (e.g., THC<CL1), the threshold determination unit 5C1 notifies the selection unit 5C2 of a trigger signal indicating the channel that exceeded the threshold. When the count values of multiple channels exceed the threshold, the threshold determination unit 5C1 can select the channel that gives the largest count value and notify the selection unit 5C2 of a trigger signal indicating the channel that exceeded the threshold.

The selection unit 5C2 sequentially receives a plurality of judgment outputs output from the measurement unit 5B. The selection unit 5C2 can temporarily store the plurality of judgment outputs that have been input. When the selection unit 5C2 receives a trigger signal from the threshold judgment unit 5C1, it selects and outputs the judgment output of the channel corresponding to the trigger signal. For example, the channel with the count value CL1 is selected, and the time-series judgment output temporarily stored in this channel is input to the main signal generation unit 5C3.

The main signal generating unit 5C3 counts the number of input judgment outputs "1" within a certain period of time, and generates a drive control signal S _C having a frequency f and amplitude A corresponding to this count value (e.g. CL1), i.e., the number of judgment outputs. This count value also corresponds to the frequency of judgment outputs. Since this frequency is the number of brain events occurring per unit time, if this is considered as the number of output waves per unit time, it is the frequency of occurrence of brain events and corresponds to the input frequency (f _in ) to the downstream circuit. This frequency also corresponds to the moving average μTTL of the number of pulses per unit time mentioned above.

It is also possible to omit the selection section and directly input the threshold-exceeding count value (e.g., CL1) measured in the threshold determination section 5C1 to the main signal generation section 5C3 as the occurrence frequency (f _in ) of brain events.

The main signal generating unit 5C3 generates a drive control signal S _C having an output frequency f and amplitude A according to the occurrence frequency of a brain event (hereinafter, input frequency f _in (frequency)). When the input frequency f _in and the output frequency f of the drive control signal S _C have a linear function relationship, the frequency f of the drive control signal S _C is given by (f = a ₁ × f _in + b ₁ ) (a ₁ is a coefficient, b ₁ is a coefficient). If an upper limit value for the output frequency f is set, the output frequency f will be constant when it exceeds the upper limit value.

When the input frequency f _in and the amplitude A have a linear function relationship, the amplitude A of the drive control signal S _C is given by (A=a ₂ ×f _in +b ₂ ) (a ₂ is a coefficient, and b ₂ is a coefficient). If an upper limit value for the amplitude A is set, the amplitude A becomes constant when it exceeds the upper limit value. The generated drive control signal S _C is input to the drive circuit 7, and a drive signal S _D (drive current) having a frequency and amplitude proportional to the output frequency f and amplitude A is supplied to the deep brain stimulation element.

In the above configuration, the digital filter 5A can be replaced with an analog filter, and the measurement unit 5B can be replaced with a window discriminator. In this case, the output of the window discriminator is AD converted and input to the signal generation unit at the subsequent stage.

As described above, in the deep brain stimulation device equipped with the control device shown in Fig. 14, the number of output signals (cortical electroencephalogram signals) of the electroencephalogram detector 1 is multiple, and the control device 5 of the drive signal generating device calculates the above-mentioned frequency for each of the multiple cortical electroencephalogram signals, and when any one of the frequencies (e.g., the frequency of occurrence of a brain event corresponding to the count value CL1 within a unit time (input frequency f _in )) exceeds a threshold value, it generates a drive control signal S _C (drive signal S _D ) having a frequency f and amplitude A corresponding to that frequency (input frequency f _in ). Since stimulation is performed when any of the multiple signals correlated with intracerebral movement indicates the occurrence of a brain event, it is possible to perform stimulation control with more stable and faster response, and it is considered that an effective therapeutic effect can be obtained.

FIG. 15 is a block diagram of the control device 5 to which multiple brainwave signals are input.

In the control device 5 shown in the figure, multiple cortical EEG signals are input to a digital filter 5A, which extracts gamma 2 band components and inputs them to a measurement unit 5B, which then sequentially outputs multiple judgment outputs. The configuration is the same as that shown in FIG. 14.

The signal generating unit 5C includes a mixer unit 5CM and a main signal generating unit 5C3. The judgment outputs of multiple channels (e.g., six) output from the measuring unit 5B are input sequentially to the mixer unit 5CM. The mixer unit 5CM counts the number of judgment outputs "1" corresponding to the occurrence of a brain event within the judgment period. In this example, six count values corresponding to six cortical EEG signals are obtained. The count values corresponding to the left and right cortical EEG signals are designated CL1, CL2, CL3, CR1, CR2, and CR3, respectively.

The mixer unit 5CM weights and adds the obtained count values to obtain a weighted average count value CW. For example, the weighting coefficients are _wL1 , _wL2 , _wL3 , _{wR1, wR2 ,} and _wR3 ( _wL1 + _wL2 + _wL3 + wR1 + wR2 + wR3 = 1). In this case, the count value CW = (CL1 x _wL1 + CL2 x _wL2 + CL3 x _wL3 + _CR1 x _wR1 + CR2 _{x wR2} + _CR3 x _wR3 ). As an example, _wL1 = _0.5 /2, _wL2 = 0.25/2, _wL3 = 0.25/2, wL1 = _wR1 , _wL2 = _wR2 , _wL3 = _wR3 . The weighting coefficients ( _wL1 , _wR1 ) in the primary motor cortex M1 are set to be larger than the other coefficients. This weighted average count value corresponds to the occurrence frequency (input frequency f _in , frequency) of the above-mentioned brain events.

As described above, in the deep brain stimulation device equipped with the control device shown in Fig. 15, the number of output signals (cortical electroencephalogram signals) of the electroencephalogram detector 1 is multiple, and the drive signal generating device calculates the frequency (count value counted within a unit time) for each of the multiple output signals, and generates a drive signal having a frequency f and an amplitude A according to an input frequency f _in corresponding to a value (weighted average count value in the above case) calculated by weighting and adding these frequencies. Since signals from multiple locations correlated with movements in the brain are used, it is possible to perform more precise stimulation control, and it is believed that an effective therapeutic effect can be obtained.

The main signal generating unit 5C3 generates a drive control signal S _C (∝ drive signal S _D ) having an output frequency f and amplitude A according to the input frequency f _in input from the mixer unit 5CM. An example of a method for calculating the frequency f and amplitude A of the drive control signal S _C is as described above, and the deep brain stimulation element can be controlled according to the calculated parameters. Next, a calculation control method using such calculation will be described.

The frequency f and amplitude A of the drive signal S _D are functions _of the input frequency f in corresponding to the frequency. In the case of an analog system, the frequency is the number of judgment outputs (pulses) output from the window discriminator 3 per judgment period, and in the case of a digital system, the frequency is the number of judgment outputs (digital value = 1) output from the measurement unit 5B per judgment period. This frequency is the number of brain events occurring per unit time, so it corresponds to the input frequency f _in when considered as the number of output waves per unit time. In the following, the relationship between the input frequency f _in and the output (frequency f and amplitude A of the drive signal S D) will be described with respect to the generation of the drive control signal S _C (∝ drive signal S _{D )} in the control device.

In the following description, _aN , _bN , _aM , and _bM are coefficients, and N and M are natural numbers (1, 2, 3, 4, etc.). In addition, the range of the input frequency f _in described in the following description ( _fN ≦f _in ≦f _N+1 , _fM ≦f _in ≦f _M+1 ) can be made to match (f _N =f _M , f _N+1 =f _M+1 ). In the following description, the input frequency threshold f _TH is an input frequency f _{in that} satisfies the condition of providing an upper limit frequency f _limit (e.g., 150 Hz) and an upper limit amplitude A _limit (e.g., 1.5 mA) on the output side. Under the condition of f _TH ≦f _in , the frequency f and the amplitude A are saturated, and the frequency f=b _MAX1 (constant) and the amplitude A=b _MAX2 (constant). The slope of the function of the frequency f and amplitude A of the drive signal is determined, thereby determining the input frequency threshold _fTH .
In many cases, the input frequency threshold _fTH was 50 Hz ± 20 Hz, particularly 50 Hz ± 10 Hz.

Fig. 16 is a graph explaining the first calculation control method. Fig. 16(A) is a graph showing the relationship between the input frequency f _in (Hz) and the frequency f (Hz) of the drive signal S _D. Fig. 16(B) is a graph showing the relationship between the input frequency f _in (Hz) and the amplitude A of the drive signal S _D. With _a1 , _b1 , _a2 , and _b2 being coefficients greater than 0, within a predetermined input frequency range (0≦f _in ≦f _TH ), the frequency (f) and amplitude (A) satisfy the following relationship.
・f=a ₁ ×f _in +b ₁
・A=a ₂ ×f _in +b ₂

As described above, the first calculation control method calculates the frequency f and amplitude A that are proportional to the input frequency f _in , generates a drive signal (drive control signal) having these parameters, and supplies this to the deep brain stimulation element. This method is simple in calculation and enables high-speed processing, and therefore can provide effective treatment.

17A and 17B are graphs for explaining the second calculation control method. Fig. 17A is a graph showing the relationship between the input frequency f _in (Hz) and the frequency f (Hz) of the drive signal S _D. Fig. 17B is a graph showing the relationship between the input frequency f _in (Hz) and the amplitude A of the drive signal S _D.

When multiple input frequency intervals are defined, each input frequency interval is fN _{≦ fin} ≦fN ₊₁ . The frequency f of the drive signal (drive control signal) is given by the following formula in the frequency interval _fN ≦ _fin ≦fN ₊₁ ( _fN <fN ₊₁ ). The frequency interval is a value that is not 0 Hz (0< _fN+1 - _fN ), and ( _{fN+1- fN} ) is set to, for example, 5 Hz or more.
・f=a _N ×f _in +b _N
aN > _aN+1
bN < _bN+1
The graphs of adjacent input frequency intervals are continuous and satisfy _{aNfN +1} + _bN =aN ₊ 1fN ₊₁ +bN ₊₁ .

When multiple input frequency intervals are defined, each input frequency interval is _fM ≦ _fin ≦ _{fM +1} . The amplitude A of the drive signal (drive control signal) is given by the following formula in the frequency interval fM≦ _fin ≦ _fM+1 ( _fM < _fM+1 ). The frequency interval is a value that is not 0 Hz (0< _fM+1 - _fM ), and ( _fM+1 - _fM ) is set to, for example, 5 Hz or more.
・A=a _M ×f _in +b _M
aM > _aM+1
bM < _{bM + 1}
The graphs of adjacent input frequency intervals are continuous and satisfy _{aMfM +1} + _bM =aM _{+1fM +1} + _bM+1 .

_aN , _bN , _aM , and _bM are coefficients greater than 0. These graphs of frequency f and amplitude A are line graphs formed by combining graphs of linear functions. In the initial state (range where input frequency f _in is small) where symptoms of Parkinson's disease are observed, the slope is large, and the deterioration of symptoms is quickly suppressed. In the period where input frequency f _in is large, the slope is made small to prevent overstimulation, and the increase in stimulation is suppressed. On the other hand, since this graph is obtained by combining linear functions, the calculation is simple, high-speed processing is possible, and the effectiveness of treatment is therefore increased.

18A and 18B are graphs for explaining the third calculation control method. Fig. 18A is a graph showing the relationship between the input frequency f _in (Hz) and the frequency f (Hz) of the drive signal S _D. Fig. 18B is a graph showing the relationship between the input frequency f _in (Hz) and the amplitude A of the drive signal S _D.

When multiple input frequency intervals are defined, each input frequency interval is fN _≦ f _in ≦f _N+1 . The frequency f of the drive control signal (drive signal) is given by the following formula where the frequency interval is _fN ≦f _in ≦f _N+1 (f _N <f _N+1 ). The frequency interval is a value that is not 0 Hz (0<f _N+1 -f _N ), and (f _N+1 -f _N ) is set to, for example, 5 Hz or more.
f = _bN
bN < _bN+1

When multiple input frequency intervals are defined, each input frequency interval is _fM ≦f _in ≦fM ₊₁ . The amplitude A of the drive control signal (drive signal) is given by the following formula in the frequency interval _fM ≦f _in ≦ _fM+1 ( _fM < _fM+1 ). The frequency interval is a value that is not 0 Hz (0< _{fM+1 -fM} ), and ( _{fM+1 -fM} ) is set to, for example, 5 Hz or more.
・A= _bM
bM < _{bM + 1}

_bN and _bM are coefficients greater than 0. A graph of these frequencies f and amplitudes A is a stepped graph. In this case, since control is performed using constants that change stepwise, calculation is simple and high-speed processing is possible, thereby increasing the effectiveness of treatment.

19A and 19B are graphs for explaining the fourth calculation control method. Fig. 19A is a graph showing the relationship between the input frequency f _in (Hz) and the frequency f (Hz) of the drive signal S _D. Fig. 19B is a graph showing the relationship between the input frequency f _in (Hz) and the amplitude A of the drive signal S _D.

The frequency f and amplitude A of the drive control signal (drive signal) in this graph satisfy the condition that the input frequency intervals ( _fN ≦f _in ≦f _N+1 ) and ( _fM ≦f _in ≦f _M+1 ) are minimized to the limit in the example shown in the broken line type graph (FIG. 17) described above. That is, there are intervals where (f _N+1 -f _N ) approaches 0 and (f _M+1 -f _M ) approaches 0. In this case, the graph of frequency f and amplitude A is a graph that changes smoothly. In other words, the output frequency f and amplitude A are functions of the input frequency f _in , the second derivative of this function is negative, and the slope of the tangent to this function (first derivative) is 0 or more.

That is, the frequency f is a function of the input frequency f in which the slope decreases as the input frequency f _in increases under the condition of 0≦f _in ≦f _TH , and f _TH is the input frequency f _in that gives the upper limit of the frequency _f . The amplitude A is a function of the input frequency f in which the slope decreases as the input frequency f _in increases under the condition of 0≦f _{in ≦} f _TH .

These graphs of frequency f and amplitude A have a steep slope in the initial state where Parkinson's disease symptoms are observed (in the range where the input frequency f _in is small), and the deterioration of symptoms is quickly suppressed. In the range where the input frequency f _in is large, the slope is made small to prevent overstimulation, and the increase in stimulation is suppressed. Therefore, the effectiveness of treatment is increased. Note that a logarithmic function can also be adopted as the nonlinear function.

Fig. 20 is a graph for explaining the fifth calculation control method. Fig. 20(A) is a graph showing the relationship between the input frequency f _in (Hz) and the frequency f (Hz) of the drive signal S _D. Fig. 20(B) is a graph showing the relationship between the input frequency f _in (Hz) and the amplitude A of the drive signal S _D. In the case of f _in ≦f _TH , these graphs satisfy the condition that the input frequency ranges (f _N ≦f _in ≦f _N+1 ) and (f _M ≦f _in ≦f _M+1 ) are minimized in the relationship given by the following formula. Note that a _N and a _M are coefficients greater than 0.

The frequency f of the drive control signal (drive signal) is given by the following formula in the frequency range fN _{≦ fin} ≦fN ₊₁ ( _fN <fN ₊₁ ).
・f=a _N ×f _in +b _N
aN < _aN+1
bN > _bN+1

The amplitude A of the drive control signal (drive signal) is given by the following formula in the frequency range fM _{? fin} ? _fM+1 ( _fM < _fM+1 ).
・A=a _M ×f _in +b _M
aM < aM _+1
bM > _bM+1

When f _TH ≦f _in , the frequency f and the amplitude A are saturated as described above.

The frequency f is a function of the input frequency _f in which the slope increases with increasing input frequency f in the condition of 0≦f _in ≦f _TH , and f _TH is the input frequency f _in that gives the upper limit value of the frequency f. The amplitude A is a function of the input frequency _f in _which the slope increases with increasing input frequency f in the condition of 0≦f _in ≦f _TH . These graphs of the frequency f and the amplitude A have a small slope and do not stimulate much in the initial state (in the range where the input frequency f _in is small) where symptoms of Parkinson's disease are observed. In the range where the input frequency f _in is large, the slope becomes large and the stimulation becomes strong when the condition becomes severe. On the other hand, when _{the input frequency f in exceeds} the threshold value f _TH , the output frequency f and the amplitude A become constant values and are suppressed so as not to stimulate too much. Therefore, the effectiveness of the treatment is increased. In addition, since the output frequency f and the amplitude A are set small in the initial stage, the power consumption can be suppressed. In addition, an exponential function can be adopted as the nonlinear function.

The above-mentioned drive signal generating device 10 includes a signal generating unit that generates a drive signal S _D having a frequency f and an amplitude A according to an input frequency f _in corresponding to the above-mentioned frequency. This signal generating unit is a part that includes the control device 5 and the drive circuit 7 in an analog system. Also, this signal generating unit is a part that includes the signal generating unit 5C and the drive circuit 7 in a digital system. Note that in the above-mentioned calculation method, the data of the calculation results obtained from each calculation formula may be stored in a storage device, and the data corresponding to the input frequency f _in may be read out from the storage device.

As explained above, the above-mentioned deep brain stimulation device 100 comprises an EEG detector 1, a drive signal generating device 10 which extracts signal components of a predetermined frequency band from the output signal of the EEG detector 1, determines the frequency of waveforms among the extracted signal components whose intensity is equal to or greater than a reference value and whose duration is equal to or greater than a reference time, and generates a drive signal _SD having a frequency f and amplitude A corresponding to the frequency, and a deep brain stimulation element 8 to which the drive signal _SD output from the drive signal generating device 10 is applied.

The drive signal generating device 10 can also extract signal components of a predetermined frequency band from the output signal of the electroencephalogram detector 1, determine the frequency of waveforms whose intensity is equal to or greater than a reference value and whose duration is equal to or greater than a reference time among the extracted signal components, and generate a drive signal S _D having a frequency f or amplitude A according to the frequency. The signal components of the predetermined frequency band are frequency bands that correlate with movement, and are preferably the above-mentioned frequency bands. The drive signal generating device 10 can preferably generate a drive signal such that the higher the frequency, the higher the frequency of the drive signal, and the higher the frequency, the larger the amplitude of the drive signal. Increasing the frequency and amplitude of the drive signal increases the power supplied to the deep brain stimulation element 8 and increases the amount of stimulation. Although a therapeutic effect can be expected even when only one of these parameters is increased, it is more effective to increase both parameters.

The above-mentioned deep brain stimulation device 100 can be applied to a variety of subjects. Applicable subjects include primates (e.g., humans, chimpanzees, and monkeys), but the principle of extracting and processing frequencies from the motor cortex can also be applied to other mammals (e.g., dogs, cats, mice, guinea pigs, rats, birds, horses, pigs, cows, etc.). It is also possible to use it in conjunction with a therapeutic drug for Parkinson's disease.

The deep brain stimulation device 100 can restore functional disorders in the loop circuits in the brain through stimulation. Therefore, it is believed that the use of this deep brain stimulation device can be effective in treating movement disorders other than Parkinson's disease that have similar functional disorders.

In addition, in the above-mentioned deep brain stimulation device 100, after extracting signal components with a central wavelength of 80 Hz to 200 Hz by a filter, the frequency of the waveform (see FIG. 2) in which the intensity and duration of the signal components exceed the standard is detected. If the frequency is high, the frequency and amplitude of the drive signal are increased to increase the power of the stimulation. The lower limit frequency that passes through the filter is at least higher than 30 Hz, but as described above, this lower limit frequency can also be 40 Hz, 60 Hz, 80 Hz, 100 Hz, or 120 Hz. The central frequency is basically the frequency that is the object of observation, but it will naturally be higher than the lower limit frequency. Therefore, for example, in order to detect brain waves in the gamma 2 band, the lower limit frequency can be set to 100 Hz or more, and the central frequency can be set to 100 Hz or more and 200 Hz or less. Although such a frequency band is clearly different from that of the conventional non-patent document 1, as described above, the extraction frequency of the deep brain stimulation device 100 can also be set to other frequency bands. Non-Patent Document 1 discloses a device that utilizes the increased activity of the gamma 1 band in dyskinesia, a side effect of Parkinson's disease treatment, to suppress the side effect by suppressing gamma 1 band activity. On the other hand, a deep brain stimulation device according to an embodiment utilizes the increased activity of the gamma 2 band when starting to move, and rather enhances gamma 2 band activity to assist the movement. Although these technologies appear similar at first glance, the underlying principles are completely different, and the frequency bands that are recorded are also different.

The drive signal generating device extracts signal components from a frequency band that includes a center frequency selected from 80 Hz to 200 Hz and has a lower limit frequency at least higher than 40 Hz, determines the frequency of waveforms among the extracted signal components that have an intensity equal to or greater than a reference value and a duration equal to or greater than a reference time, and generates a drive signal having a frequency and amplitude according to the frequency.

The drive signal generating device extracts signal components from a frequency band that includes a center frequency selected from 80 Hz to 200 Hz and has a lower limit frequency at least higher than 60 Hz, determines the frequency of waveforms among the extracted signal components that have an intensity equal to or greater than a reference value and a duration equal to or greater than a reference time, and generates a drive signal having a frequency and amplitude according to the frequency.

The drive signal generating device extracts signal components from a frequency band that includes a center frequency selected from 80 Hz to 200 Hz and has a lower limit frequency at least higher than 80 Hz, determines the frequency of waveforms among the extracted signal components that have an intensity equal to or greater than a reference value and a duration equal to or greater than a reference time, and generates a drive signal having a frequency and amplitude according to the frequency.

The drive signal generating device extracts signal components in a frequency band that includes a center frequency selected from 100 Hz to 200 Hz and has a lower limit frequency at least higher than 100 Hz, determines the frequency of waveforms among the extracted signal components that have an intensity equal to or greater than a reference value and a duration equal to or greater than a reference time, and generates a drive signal having a frequency and amplitude according to the frequency.

As described above, the deep brain stimulation device includes an EEG detector, an analog or digital filter to which the output signal from the EEG detector is input, a drive circuit that outputs a drive signal having a frequency corresponding to the frequency of waveforms in which the intensity of the signal component extracted by the filter is equal to or greater than a reference value and the duration is equal to or greater than a reference time, and a deep brain stimulation element connected to the drive circuit.

To clarify the detection function of the gamma 2 band, the lower limit frequency of the frequency extracted by this filter may be set to 80 Hz or higher.

To clarify the detection function of the γ2 band, the lower limit frequency of the frequency extracted by this filter may be set to 90 Hz or higher.

To clarify the detection function of the gamma 2 band, the lower limit frequency of the frequency extracted by this filter may be set to 100 Hz or higher.

Although various exemplary embodiments have been described above, various omissions, substitutions, and modifications may be made without being limited to the above-described exemplary embodiments. Elements in different embodiments may be combined to form other embodiments. It will be understood from the above description that the various embodiments of the present disclosure have been described herein for illustrative purposes, and that various modifications may be made without departing from the scope and spirit of the present disclosure. Accordingly, the various embodiments disclosed herein are not intended to be limiting, with the true scope and spirit being indicated by the appended claims.

1...EEG detector, 2...filter, 3...window discriminator, 5...control device, 5A...digital filter, 5B...measuring unit, 5C...signal generating unit, 7...driving circuit (analog isolator), 8...deep brain stimulation element, 10...driving signal generating device, 100...deep brain stimulation device, BOX...sealed container, S _C ...driving control signal, S _D ...driving signal.

Claims (25)

A brainwave detector;
a drive signal generating device which receives an output signal from the electroencephalogram detector, extracts signal components in a frequency band having a center frequency selected from 80 Hz to 200 Hz and a lower limit frequency at least higher than 30 Hz, determines a frequency of waveforms having an intensity equal to or greater than a reference value and a duration equal to or greater than a reference time among the extracted signal components, and generates a drive signal having a frequency and amplitude according to the frequency;
a deep brain stimulation element to which the drive signal output from the drive signal generating device is applied;
A deep brain stimulation device comprising: The drive signal generating device includes:
The higher the frequency, the higher the frequency of the drive signal;
The higher the frequency, the larger the amplitude of the drive signal.
generating the drive signal such that
The deep brain stimulation device according to claim 1 . The drive signal generating device includes:
an analog filter that receives the output signal of the electroencephalogram detector and extracts signal components in the frequency band;
a window discriminator which receives the output signal of the analog filter, generates a pulse voltage every time a waveform having an intensity equal to or greater than a reference value and a duration equal to or greater than a reference time is detected, and outputs a pulse signal;
a control device that receives the pulse signal output from the window discriminator and generates a drive control signal having a frequency and amplitude corresponding to the frequency of the pulse signal;
a drive circuit that generates the drive signal having a frequency and an amplitude corresponding to the drive control signal output from the control device;
Equipped with
The deep brain stimulation device according to claim 1 . The drive signal generating device includes:
a control device to which an output signal from the electroencephalogram detector is input as a digital signal;
a drive circuit that generates the drive signal having a frequency and amplitude corresponding to a drive control signal output from the control device;
Equipped with
The control device includes:
a digital filter that extracts a signal component in the frequency band from the digital signal;
a measurement unit that measures a frequency of the waveform from the extracted signal components;
a signal generating unit that generates the drive control signal having a frequency and an amplitude according to the frequency measured by the measuring unit;
Equipped with
The deep brain stimulation device according to claim 1 . The drive circuit includes:
an input unit to which the drive control signal is input;
an output section that generates the drive signal insulated from the input based on the drive control signal;
An analog isolator comprising:
The deep brain stimulation device according to claim 3 or 4. The drive signal generating device is housed in a sealed container,
a signal line extending to the outside of the sealed container is connected to an output terminal of the drive signal of the drive signal generating device;
The signal line is connected to the deep brain stimulation element.
A deep brain stimulation device according to any one of claims 1 to 4. the number of the output signals of the electroencephalogram detector is more than one;
The drive signal generating device includes:
determining the frequency for each of the plurality of output signals, and when any one of the frequencies exceeds a threshold, generating the drive signal having a frequency and an amplitude according to an input frequency corresponding to the frequency;
A deep brain stimulation device according to any one of claims 1 to 4. the number of the output signals of the electroencephalogram detector is more than one;
The drive signal generating device includes:
determining the frequency for each of the plurality of output signals, weighting and adding the frequencies to generate the drive signal having a frequency and amplitude according to an input frequency corresponding to a value obtained by weighting and adding the frequencies;
A deep brain stimulation device according to any one of claims 1 to 4. The electroencephalogram detector includes:
A detection electrode for detecting brain waves;
A plurality of reference electrodes arranged around the detection electrode;
a plurality of resistors each having one end connected to the plurality of reference electrodes;
an amplifier including a reference potential input terminal to which a reference potential of a node to which the other ends of the plurality of resistors are connected is input, and a detection potential input terminal to which a detection potential from the detection electrode is input;
Equipped with
A deep brain stimulation device according to any one of claims 1 to 4. The drive signal generating device includes a signal generating unit that generates the drive signal having a frequency f and an amplitude A according to an input frequency f _in corresponding to the frequency.
3. A deep brain stimulation device according to claim 1 or 2. The frequency f and the amplitude A are
In the condition of 0≦f _in ≦f _TH , the following relationship is satisfied:
f=a ₁ ×f _in +b ₁ ,
A=a ₂ ×f _in +b ₂ ,
a ₁ , b ₁ , a ₂ , and b ₂ are coefficients greater than 0;
f _TH is an input frequency f _in that gives an upper limit value of the frequency f,
The deep brain stimulation device according to claim 10. The frequency f is
A plurality of input frequency intervals are defined under the condition of 0≦f _in ≦f _TH , and each input frequency interval is f _N ≦f _in ≦f _N+1 and satisfies the following relationship:
f=a _N ×f _in +b _N ,
_aN > _aN+1 ,
_bN <_bN+1;
0<f _N+1 −f _N ,
The amplitude A is
A plurality of input frequency intervals are defined under the condition of 0≦f _in ≦f _TH , and each input frequency interval is f _M ≦f _in ≦f _M+1 and satisfies the following relationship:
A=a _M ×f _in +b _M ,
_aM > _aM+1 ,
_bM <_bM+1;
0<f _M+1 −f _M ,
_aN , _bN , _aM , and _bM are coefficients greater than 0, and N and M are natural numbers;
f _TH is an input frequency f _in that gives an upper limit value of the frequency f,
The deep brain stimulation device according to claim 10. The frequency f is
A plurality of input frequency intervals are defined under the condition of 0≦f _in ≦f _TH , and each input frequency interval is f _N ≦f _in ≦f _N+1 and satisfies the following relationship:
f= _bN ,
_bN <_bN+1;
0<f _N+1 −f _N ,
The amplitude A is
A plurality of input frequency intervals are defined under the condition of 0≦f _in ≦f _TH , and each input frequency interval is f _M ≦f _in ≦f _M+1 and satisfies the following relationship:
A= _bM ,
_bM <_bM+1;
0<f _M+1 −f _M ,
b _N and b _M are coefficients greater than 0, and N and M are natural numbers;
f _TH is an input frequency f _in that gives an upper limit value of the frequency f,
The deep brain stimulation device according to claim 10. The frequency f is a function _of the input frequency f in which the slope decreases with increasing input frequency _f in the condition of 0≦f _in ≦f _TH ,
f _TH is an input frequency f _in that gives an upper limit value of the frequency f,
The deep brain stimulation device according to claim 10. The frequency f is a function of the input frequency _f in which the slope increases with increasing input frequency f _in under the condition of 0≦f _in ≦f _TH ,
f _TH is an input frequency f _in that gives an upper limit value of the frequency f,
The deep brain stimulation device according to claim 10. A brainwave detector;
a drive signal generating device that extracts signal components of a predetermined frequency band from the output signal of the electroencephalogram detector, determines a frequency of waveforms of which the intensity is equal to or greater than a reference value and whose duration is equal to or greater than a reference time among the extracted signal components, and generates a drive signal having a frequency and amplitude corresponding to the frequency;
a deep brain stimulation element to which the drive signal output from the drive signal generating device is applied;
A deep brain stimulation device comprising: A brainwave detector;
a drive signal generating device that extracts signal components of a predetermined frequency band from the output signal of the electroencephalogram detector, determines a frequency of waveforms of which the intensity is equal to or greater than a reference value and whose duration is equal to or greater than a reference time among the extracted signal components, and generates a drive signal having a frequency or amplitude corresponding to the frequency;
a deep brain stimulation element to which the drive signal output from the drive signal generating device is applied;
A deep brain stimulation device comprising: The drive signal generating device includes:
extracting signal components in a frequency band including a center frequency selected from 80 Hz to 200 Hz and having a lower limit frequency at least higher than 40 Hz, determining the frequency of waveforms among the extracted signal components that have an intensity equal to or greater than a reference value and a duration equal to or greater than a reference time, and generating a drive signal having a frequency and amplitude according to the frequency;
The deep brain stimulation device according to claim 1 . The drive signal generating device includes:
extracting signal components in a frequency band including a center frequency selected from 80 Hz to 200 Hz and having a lower limit frequency at least higher than 60 Hz, determining the frequency of waveforms having an intensity equal to or greater than a reference value and a duration equal to or greater than a reference time among the extracted signal components, and generating a drive signal having a frequency and amplitude according to the frequency;
The deep brain stimulation device according to claim 1 . The drive signal generating device includes:
extracting signal components in a frequency band including a center frequency selected from 80 Hz to 200 Hz and having a lower limit frequency at least higher than 80 Hz, determining the frequency of waveforms having an intensity equal to or greater than a reference value and a duration equal to or greater than a reference time among the extracted signal components, and generating a drive signal having a frequency and amplitude according to the frequency;
The deep brain stimulation device according to claim 1 . The drive signal generating device includes:
extracting signal components in a frequency band including a center frequency selected from 100 Hz to 200 Hz and having a lower limit frequency at least higher than 100 Hz, determining the frequency of waveforms having an intensity equal to or greater than a reference value and a duration equal to or greater than a reference time among the extracted signal components, and generating a drive signal having a frequency and amplitude according to the frequency;
The deep brain stimulation device according to claim 1 . A brainwave detector;
an analog or digital filter to which the output signal from the electroencephalogram detector is input;
a drive circuit that outputs a drive signal having a frequency corresponding to the frequency of waveforms in which the intensity of the signal component extracted by the filter is equal to or greater than a reference value and the duration is equal to or greater than a reference time;
a deep brain stimulation element connected to the drive circuit;
Equipped with
Deep brain stimulator. The lower limit frequency of the frequency to be extracted by the filter is 80 Hz or more.
23. The deep brain stimulation device of claim 22. The lower limit frequency of the frequency to be extracted by the filter is 90 Hz or more.
23. The deep brain stimulation device of claim 22. The lower limit frequency of the frequency to be extracted by the filter is 100 Hz or more.
23. The deep brain stimulation device of claim 22.

Images