Systems and methods for monitoring neural activity

Abstract

To provide improved systems for monitoring neural activity.SOLUTION: A method is provided, comprising: applying a plurality of electrical stimuli having differing amplitudes at a first position in the brain of a patient; for each of the plurality of electrical stimuli, detecting a resonant response from a target neural structure in the brain, each resonant response being evoked by each one of the electrical stimuli; determining a change in one or more common waveform characteristics between two or more of the detected resonant responses; and determining effectiveness in delivering a therapeutic stimulation at the first position on the basis of the change.SELECTED DRAWING: Figure 10

Description

[Technical field]

[0001] The present disclosure relates to deep brain stimulation (DBS), and more particularly to methods and systems for monitoring neural activity in response to DBS. [Background technology]

[0002] Deep brain stimulation (DBS) is an established therapy for movement disorders and other neurological disorders, including epilepsy, obsessive-compulsive disorder, and depression. DBS is typically performed on patients whose symptoms cannot be adequately controlled with medication alone. DBS involves surgically implanting electrodes in or near specific neural structures in the brain, typically the subthalamic nucleus (STN), globus pallidus interna (GPi), and / or the thalamus. The electrodes are connected to a neurostimulator, often implanted in the body, configured to deliver electrical pulses into the target area. This electrical stimulation is believed to disrupt abnormal brain activity that is causally related to the patient's symptoms. Stimulation parameters can be adjusted using a controller external to the body, remotely connected to the neurostimulator.

[0003] Although established DBS techniques have proven effective in reducing the symptoms of movement disorders, state-of-the-art devices have several limitations. In particular, established techniques for intraoperative testing of DBS electrodes to ensure accurate positioning in the brain, such as x-ray imaging, microelectrode recordings, and clinical evaluations, can be inaccurate. As a result, electrodes are often implanted in suboptimal locations, resulting in poorer treatment outcomes and unwanted side effects. DBS devices must be manually adjusted by clinicians after implantation. This typically involves the clinician adjusting parameters of stimulation based on a largely subjective assessment of immediate or short-term improvements in the patient's symptoms. Because treatment effects can be slow to appear, and because the DBS parameter space is large, the task of finding a preferred set of parameters can be time- and cost-inefficient and can result in suboptimal treatment outcomes. In addition, the constant and unchanging application of electrical stimulation with conventional DBS can also result in suboptimal treatment outcomes, including unwanted side effects, and reduced battery life of DBS stimulators. Summary of the Invention

[0004] According to a first aspect of the present disclosure, there is provided a method comprising: a) applying a plurality of electrical stimuli having different amplitudes at a first location in a patient's brain; for each of the plurality of electrical stimuli; b) detecting a resonant response from a target neural structure in the brain, each resonant response evoked by a respective individual electrical stimulus; c) determining a change in one or more common waveform characteristics between two or more of the detected resonant responses; and d) determining effectiveness of delivery of a therapeutic stimulus at the first location based on the change.

[0005] The one or more common waveform characteristics may include one or more of the following: amplitude of the resonant response; recurrence of the resonant response; frequency of the resonant response; time envelope of the resonant response; fine structure of the resonant response; rate of decay of the resonant response; delay time between the onset of the stimulus and the onset of the temporal characteristics of the resonant response amplitude.

[0006] Determining a change in one or more common waveform characteristics between two or more of the detected resonant responses may include determining a rate of change in the one or more common waveform characteristics between two or more of the detected resonant responses.

[0007] The method may further include selecting a different amplitude for therapeutic stimulation of the target neural structure based on the variation.

[0008] The method may further include selecting a first location for therapeutic stimulation of the target neural structure based on the change.

[0009] The method may further include repeating steps a-d at a second location in the brain different from the first location and comparing changes in one or more common waveform characteristics between the first and second locations. The method may further include selecting one of the first and second locations for therapeutic stimulation of the target neural structure based on the comparison. The method may further include applying therapeutic stimulation to the target neural structure at the selected location.

[0010] The plurality of electrical stimuli applied at the first location may be applied by a first electrode of the plurality of electrodes. The method may further include moving the first electrode between the first and second locations in the brain; and determining an optimal electrode location for therapeutic stimulation of the target neural structure based on a comparison of changes in one or more common waveform characteristics at the first and second locations.

[0011] The plurality of electrical stimuli applied at the second location may be applied by a second electrode of the plurality of electrodes.

[0012] In some embodiments, the plurality of electrical stimuli applied at the first location are applied by two or more electrodes at the first location. In some embodiments, at least one of the two or more electrodes is not located at the first location. In some embodiments, the resonant response is detected at the two or more electrodes.

[0013] The method may further include removing the plurality of electrodes from the first location in the brain and implanting one or more permanent electrodes in the brain at the first location.

[0014] The patient may be under general anesthesia during any or all of the above steps. In some embodiments, the patient may be under general anesthesia during at least steps a-d above.

[0015] According to another aspect of the present disclosure, there is provided a neurostimulation system including at least one electrode adapted for implantation in or near a target neural structure in a brain; a signal generator selectively coupled to one or more of the at least one electrode and configured to apply a plurality of electrical stimuli having different amplitudes via the at least one electrode at a first location in the patient's brain; a measurement device selectively coupled to one or more of the at least one electrode and configured to detect a resonant response from the target neural structure evoked by each of the plurality of electrical stimuli; and a processing unit coupled to the measurement device and configured to determine a change in one or more common waveform characteristics between two or more of the detected resonant responses and to determine effectiveness of delivery of a therapeutic stimulation at the first location based on the change.

[0016] Determining a change in one or more common waveform characteristics between two or more of the detected resonant responses may include determining a rate of change in the one or more common waveform characteristics between two or more of the detected resonant responses.

[0017] The one or more common waveform characteristics may include one or more of the following: amplitude of the resonant response; recurrence of the resonant response; frequency of the resonant response; time envelope of the resonant response; fine structure of the resonant response; rate of decay of the resonant response; delay time between the onset of the stimulus and the onset of the temporal characteristics of the resonant response amplitude.

[0018] The processing unit may be further configured to select a different amplitude for therapeutic stimulation of the target neural structure based on the variation.The processing unit may be further configured to select a first location for therapeutic stimulation of the target neural structure based on the variation.

[0019] The at least one electrode may include a first electrode at a first location and a second electrode at a second location different from the first location. The signal generator may be configured to apply a plurality of electrical stimuli to each of the first and second electrodes. The measurement device may be configured to detect a resonant response from the target neural structure evoked by each of the plurality of electrical stimuli applied to each of the first and second electrodes. The processing unit may be configured to determine a change in one or more common waveform characteristics among two or more of the detected resonant responses evoked by each of the plurality of electrical stimuli applied to the second electrode at the second location; to compare the change in the one or more common waveform characteristics of the two or more of the detected resonant responses at the first electrode with a change in the one or more common waveform characteristics among two or more of the detected resonant responses at the second electrode; and to select one of the first electrode and the second electrode for therapeutic stimulation of the target neural structure based on the comparison.

[0020] The signal generator may be further configured to apply therapeutic stimulation to the target neural structure via selected electrodes.

[0021] The plurality of electrical stimuli are applied to a first electrode of the at least one electrode. The signal generator may be configured to apply the plurality of electrical stimuli to the first electrode at the second location in response to movement of the first electrode to a second location in the brain different from the first location. The measurement device may be configured to detect a resonant response from the target neural structure evoked by each of the plurality of electrical stimuli applied to the first electrode at the second location. The processing unit may be configured to determine a change in one or more common waveform characteristics among two or more of the detected resonant responses evoked by each of the plurality of electrical stimuli applied to the first electrode at the second location; to compare the change in the one or more common waveform characteristics among two or more of the detected resonant responses detected at the first location with a change in the one or more common waveform characteristics among two or more of the detected resonant responses detected at the second location; and to determine an optimal electrode location for therapeutic stimulation of the target neural structure based on the comparison.

[0022] In some embodiments, the plurality of electrical stimuli applied at the first location are applied by two or more electrodes at the first location. In some embodiments, at least one of the two or more electrodes is not located at the first location. In some embodiments, the resonant response is detected at two or more of the at least one electrode.

[0023] It will be understood that throughout this specification the word "comprise" or variations such as "comprises" or "comprising" imply the inclusion of a stated element, integer or step, or group of elements, integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps.

[0024] Embodiments of the present disclosure will now be described by way of non-limiting examples with reference to the drawings, in which: [Brief description of the drawings]

[0025] [Figure 1]1 is a graph showing resonance from neural structures in response to a deep brain stimulation (DBS) signal. [Diagram 2] 1 is a graph showing resonance from neural structures in response to a patterned DBS signal. [Diagram 3] The graph shows the observed resonance from neural structures in response to patterned stimulation of 0.3 mA, 0.67 mA, 1.5 mA and 3.38 mA at different insertion depths of the electrode lead in the vicinity of the subthalamic nucleus of a patient. [Figure 4A] 13 is a graph showing evoked resonance beginning to bifurcate into two peaks in response to patterned DBS with an amplitude of 1.5 mA. [Figure 4B] 13 is a graph showing a bifurcated two-peak evoked resonance in response to patterned DBS with an amplitude of 2.25 mA. [Figure 4C] 1 is a graph showing two distinct evoked resonance peaks in response to patterned DBS with an amplitude of 3.375 mA. [Diagram 5] The graph shows the observed resonance from neural structures in response to patterned stimulation of 0.3 mA, 0.67 mA, 1.5 mA and 3.38 mA at different insertion depths of the electrode lead in the vicinity of the subthalamic nucleus of a patient. [Figure 6] The graph shows the observed resonance from neural structures in response to patterned stimulation of 0.3 mA, 0.67 mA, 1.5 mA and 3.38 mA at different insertion depths of the electrode lead in the vicinity of the subthalamic nucleus of a patient. [Figure 7] The graph shows the observed resonance from neural structures in response to patterned stimulation of 0.3 mA, 0.67 mA, 1.5 mA and 3.38 mA at different insertion depths of the electrode lead in the vicinity of the subthalamic nucleus of a patient. [Figure 7A] The graph shows the observed resonances from neural structures in response to 1.5 mA patterned stimulation at different insertion depths of an electrode lead near the subthalamic nucleus of a patient, and the associated spectrograms based on the wavelet transform for each insertion depth. [Figure 7B]The graph shows the observed resonances from neural structures in response to 3.38 mA patterned stimulation at different insertion depths of an electrode lead near the subthalamic nucleus of a patient, and the associated spectrograms based on the wavelet transform for each insertion depth. [Figure 8] FIG. 1 is a schematic diagram of the tip of an electrode lead for implantation in the brain. [Figure 9] FIG. 1 is a schematic diagram of an electrode lead implanted in the subthalamic nucleus of the brain. [Figure 10] FIG. 1 is a schematic diagram of a system for performing DBS. [Figure 11] 1 is a flow chart showing the process of placing DBS electrodes in the brain. [Figure 12] 1 is a flow diagram illustrating a process for monitoring and processing a resonance response at multiple electrodes in response to stimulation at the multiple electrodes. [Figure 13] 13 is a graphical illustration of the resonant response in response to a stimulation signal measured at different electrodes implanted in the brain according to the process shown in FIG. 12. [Figure 14] The transition between therapeutic and non-therapeutic stimulation periods is plotted against the resonant activity characteristic of the evoked response. [Figure 15A] 1 illustrates a patterned stimulation signal according to an embodiment of the present disclosure. [Figure 15B] 1 illustrates another patterned stimulation signal according to an embodiment of the present disclosure. [Figure 16] 3D reconstruction of electrode arrays implanted in the STN (smaller mass) and substantia nigra (larger mass) of a patient. [Figure 17] AC are MRI and CT fusion scans of the patient's brain showing the positioning of the electrode array. [Figure 18] The variation of ERNA amplitude with respect to electrode position is shown in the graph. [Figure 19A] ERNA frequency is plotted against DBS amplitude for 19 brain hemispheres. [Figure 19B] ERNA amplitude versus DBS amplitude for 19 brain hemispheres is graphed. [Figure 19C]Relative beta (RMS 13-30Hz / RMS 4-45Hz) versus DBS amplitude is graphed for 19 hemispheres. [Figure 19D] Relative beta (ρ=0.601, p<0.001) correlated with ERNA frequency for 19 brain hemispheres is shown in the graph. [Figure 20A] The graph shows ERNA frequency washout during successive 15 second periods after DBS and in the final 15 seconds before DBS for 19 hemispheres. [Figure 20B] ERNA amplitude washout is graphed for 19 brain hemispheres. [Figure 20C] Relative beta washout is graphed for 19 brain hemispheres. [Figure 20D] The graph shows washout in the 200-400 Hz HFO band for 19 brain hemispheres. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0026] SUMMARY OF THE DISCLOSURE Embodiments of the present disclosure relate to improvements in neurostimulation in the brain. Specifically, embodiments of the present disclosure improve upon techniques for placement of DBS electrodes in the brain by comparing evoked resonant neural activity (ERNA) in response to electrical stimulation applied to the DBS electrodes.

[0027] DBS devices typically apply constant amplitude stimuli to target regions of the brain at a constant frequency of 130 Hz. It has previously been found that application of such stimuli elicits neural responses from target regions of the brain, and that the neural responses include a resonant component. It has previously been reported that continuous DBS at conventional frequencies does not allow for a long enough time window to observe resonant activity. However, resonant activity can be monitored by monitoring the evoked neural responses in the brain after stimulation is discontinued (such as by patterning the stimulation signal). This is described in detail in International Application No. PCT / AU2017 / 050809, the contents of which are hereby incorporated by reference in their entirety. Such technology has applications both for reducing the physical effects associated with musculoskeletal disorders, and also for reducing the adverse effects of other neurological conditions, neuropsychiatric disorders, sensory disorders, and pain.

[0028] FIG. 1 is a graph showing responses from a neural circuit stimulated by a 130 Hz signal delivered from a neurostimulator via an electrode lead, e.g., a 3387 electrode lead manufactured by Medtronic (RTM), implanted in the subthalamic nucleus (STN) of a Parkinson's disease (PD) patient. Each response to a stimulation pulse contains an evoked compound action potential (ECAP) component along with a component of evoked resonant neural activity (ERNA) that occurs after the ECAP. The ECAP typically occurs within 1-2 milliseconds of the stimulation pulse. The graph shows responses to a series of the last three 60-second continuous stimulation pulses followed by a no-stimulation period. It can be seen that the evoked resonant responses to each of the first two stimulation pulses shown in FIG. 1 are interrupted by the onset of the next stimulation pulse such that only a single secondary peak is detectable. However, the evoked resonant response to the third (and last) pulse is able to resonate for a longer period and is therefore clearly visible in the form of a damped oscillation with at least seven peaks over a post-stimulation period of about 30 milliseconds.

[0029] It has been found that by controlling DBS parameters in a certain way, non-therapeutic stimulation can be administered that induces a resonant neural response (ERNA) in the patient without any therapeutic effects or undesired side effects. Such non-therapeutic stimulation can be used to reliably measure ERNA without causing lasting changes to the resonant neural circuitry or the symptomatic state of the patient. Non-therapeutic stimulation is preferably accomplished by administering stimulation that includes a short burst of pulses followed by a non-stimulation period, and ERNA is measured during this non-stimulation period. By doing so, the total charge or energy delivered to the patient is below the therapeutic threshold, and the measured ERNA provides information about the patient's natural state (without therapy). In an alternative embodiment, the overall charge or energy delivered to the patient can be reduced by reducing the amplitude of the stimulation signal below the therapeutic threshold. However, doing so may also result in a lower amplitude of the ERNA peak, which may be more difficult to observe.

[0030] Patterned stimulation may also be used to monitor and analyze induced resonant neural activity during therapeutic stimulation of a patient. By patterning the stimulation signal, therapeutic stimulation may be maintained while providing a time window for monitoring the resonant response after a first resonant peak, or more preferably after two or more resonant peaks.

[0031] FIG. 2 graphically illustrates an example therapeutic patterned DBS stimulation 20 and associated evoked resonance response according to an embodiment of the present disclosure. The patterned stimulation 20 is shown on top of the graph to show the correlation between stimulation and response. In the patterned stimulation, a single pulse is omitted from a continuous 130 Hz pulse train. Thus, the pulse train includes multiple bursts of continuous stimulation pulses, the bursts being separated by a first duration t1, and each of the multiple pulses being separated by a second duration t2. While continuation of stimulation before and after the omission of a pulse (or more than one pulse) maintains the therapeutic properties of DBS, the omission of a pulse allows the resonance of the ERNA to be monitored for several (three in this example) resonance periods before the next stimulation pulse disrupts this resonance.

[0032] In summary, by patterning the non-therapeutic and therapeutic stimulation, evoked responses can be monitored over a longer period of time than with conventional non-patterned stimulation. Thus, the stimulation is preferably applied as multiple-pulse bursts, with the bursts separated by a first period of no stimulation t1 and the pulses separated by a second period of time t2. For example, the stimulation signal may include a series of 10-pulse bursts at 130 Hz. To increase the reproducibility of the results, the multi-pulse bursts may be repeated after a predefined period of no stimulation. For example, the multi-pulse bursts may be repeated every second. The duration of the first period t1 is longer than that of the second period t2. The ratio of the duration of the bursts to the duration between bursts may be selected to ensure that the important attributes of the ERNA can be easily and efficiently monitored. In some embodiments, the duration of each burst is selected to be 1-20% of the duration of no stimulation between the bursts.

[0033] Patterned stimulation can also be achieved by interleaving therapeutic stimulation (e.g., at a frequency of 130 Hz) with bursts of stimulation having a lower frequency (e.g., 90 Hz). The frequency of these interleaved bursts is preferably low enough to allow the observation of multiple ERNA peaks. Similarly, the frequency of these interleaved bursts is preferably high enough to fall within the therapeutic frequency range for DBS. The transition between frequencies may be abrupt, or the frequency change may be gradual. It may be beneficial to ramp the frequency of the pulses to avoid abrupt step changes in frequency.

[0034] In addition to or instead of adjusting the frequency of the applied stimulation, the amplitude of the pulse may be modulated over time. This may include ramping the pulse amplitude to increase over several pulses in a burst, and / or ramping the pulse amplitude to decrease over several pulses in a burst. To enhance monitoring of ERNA, it may be beneficial to apply a fixed amplitude to the pulse preceding the observation window, and if this amplitude differs from that applied at other times (e.g., to maximize therapeutic benefit), it may be beneficial to ramp the amplitude of the pulse to avoid abrupt step changes in amplitude.

[0035] Using the above patterning technique, ERNA can be monitored in response to stimulation with different amplitudes. From these observations, it has been found that there is a relationship between the effectiveness of therapeutic stimulation and the change in the common waveform characteristics of evoked resonance in response to stimulation with different amplitudes. For example, in electrodes that deliver effective therapeutic stimulation, increasing the stimulation amplitude leads to a change in the amplitude of evoked resonance. In contrast, in electrodes that are less effective for therapy, changes in stimulation amplitude have little effect on the amplitude of evoked resonance.

[0036] FIG. 3 graphically illustrates the ERNA observed in response to patterned stimulation of 0.3 mA, 0.67 mA, 1.5 mA, and 3.38 mA (from left to right) at different insertion depths of the electrode lead near the subthalamic nucleus of a patient. Stimuli were applied to a single electrode as bursts of 10 pulses at 130 Hz. At each amplitude, two bursts per second were applied for 4 seconds each (i.e., 8 bursts per amplitude). ERNA responses were extracted and detrended by filtering each response with a high-pass filter (3rd order Butterworth, fc=130 Hz) and a low-pass filter (1st order Butterworth, fc=1000 Hz). The depth scale on the x-axis of each graph in FIG. 3 represents the location of the electrode relative to the surgical target. Note that this target does not necessarily correspond to an ideal electrode location, but rather to a fixed reference location in the patient's brain that provides the surgeon with a target location during electrode lead implantation. For surgical implantation, the surgical target is selected based on anatomical landmarks, typically identified by stereotactic imaging or similar, and is typically approximately 2-4 mm deeper than the therapeutic target.

[0037] Referring to the leftmost graph in FIG. 3, which corresponds to ERNA in response to stimulation at an amplitude of 0.3 mA, it can be seen that the maximum amplitude of ERNA occurs at the surgical target, i.e., at a depth of 0. However, as the stimulation amplitude is increased, it can be seen that the amplitude of ERNA measured at the surgical target remains largely unchanged. In contrast, there is a significant increase in the amplitude of ERNA observed at -1, -2, and -3 mm (i.e., 1, 2, and 3 mm above the surgical target), with the maximum change occurring at a depth of -2 (i.e., 2 mm from the surgical target). It is believed that therapeutic DBS may be most effective at this location.

[0038] The above indicates that optimal electrode positioning for therapeutic stimulation can be achieved more efficiently and accurately by observing ERNA in response to stimulation applied at both different implantation locations and different amplitudes.

[0039] It has also been previously found that in some cases the evoked activity contains multiple resonances. Figures 4a, 4b and 4c show ERNA in response to continuous DBS at 1.5mA, 2.25mA and 3.375mA, respectively. At 1.5mA, the resonant ERNA starts as a single peak, which can be seen to begin to bifurcate slightly into two peaks. At 2.25mA, the dominance is switched to the later of the two peaks. Meanwhile, the earlier peak that was dominant at 1.5mA continues at a lower amplitude. At 3.375mA, there are two peaks, with the later peak dominating. These multiple resonant peaks are believed to correspond to activity in different neural circuits. The relative amplitude between these resonant responses (or other features, such as temporal or spectral characteristics) can be indicative of the treatment state.

[0040] It is hypothesized that the reemergence in ERNA is the result of multiple overlapping resonant responses. Figure 5 graphically illustrates the ERNA observed in response to the same patterned stimulation described above with reference to Figure 3 at different insertion depths of the electrode lead near the patient's subthalamic nucleus (from left to right) at 0.3 mA, 0.67 mA, 1.5 mA, and 3.38 mA. Figure 5 illustrates the ERNA observed over a longer period (75 ms). Reemergence is visible in the ERNA observed at a depth of -2 mm in response to the 3.38 mA stimulation.

[0041] Figures 6 and 7 graphically show ERNA observed in another patient in response to the same patterned stimulation as described above with reference to Figures 3 and 5. Figure 6 shows ERNA over 25 ms, while Figure 7 shows ERNA over a longer period of 75 ms. For this patient, the growth in ERNA amplitude is comparable at the surgical target and at -1 mm over the shorter period of 25 ms, more clearly shown in Figure 6. However, over the longer period of 75 ms, shown in Figure 7, there is a significant resurgence at -1 mm, which is believed to correspond to optimal electrode placement for therapeutic stimulation. Thus, there appears to be a correlation between resurgence in ERNA and optimal electrode placement.

[0042] Figures 7A and 7B show the observed ERNAs (left) shown in Figure 7 at depths of -2mm, -1mm from the surgical target and at the surgical target for stimulation amplitudes of 1.5mA and 3.38mA, respectively. On the right of each of Figures 7A and 7B are shown the associated spectrograms based on the wavelet transform corresponding to the ERNA depicted on the left, for each depth. Each spectrogram represents the average of the spectra of eight responses to eight bursts of stimulation (as above). The spectrograms were generated using the MatLab (RTM) cwt function ("Morse" wavelet, gamma=5, time bandwidth=120, 32 tones per octave).

[0043] The wavelet spectrogram supports the hypothesis that the reemergence is consistent with a superposition or resonance at two frequencies. For example, the wavelet spectrogram for the ERNA observed at an insertion depth of -1 mm appears to contain two frequency components between 0 and approximately 20 ms. From approximately 15 ms, the frequency components appear to merge into a single frequency component. The observed reemergence of ERNA appears to occur immediately after this merging of multiple frequency components.

[0044] It can also be seen from the observed ERNA graphs in Figures 7A-7B that the latency between stimulation and the observed ERNA onset varies with insertion depth. Specifically, the latency appears to increase as the placed electrode approaches the surgical target. It will therefore be appreciated that the ERNA onset latency can be used to determine the location of the electrode relative to the surgical target, the distance to the target, the orientation to the target (triangulating using multiple (i.e., three or more) electrodes), etc.

[0045] It will be appreciated that various characteristics of the ERNA waveform, such as time to first peak, delay between ERNA peaks, variable ERNA amplitude and recurrence, may be patient specific. Advantageously, any reliance on such data to control DBS stimulation will be based on an initial characterization of the ERNA waveform and movement disorder associated with each patient to generate a plan for stimulation control (e.g., closed loop control).

[0046] Identification of correlations between changes in the resonant behavior of stimulated neural circuits and patient disease symptoms provides several opportunities to improve aspects of DBS therapy, including, but not limited to, techniques for the initial implantation and subsequent repositioning of DBS electrodes, setting the parameters of DBS stimulation, and using feedback to adjust DBS parameters in real time while DBS therapy is ongoing.

[0047] Several practical applications of the induced resonant neural activity will now be described with reference to several embodiments. In embodiments, one or more electrode leads may be used for stimulation of one or more neural structures in one or both hemispheres of the brain, and each lead may include one or more electrodes located near the tip of each lead. Each electrode may be used for stimulation, monitoring, or both stimulation and monitoring. One or more of these electrodes may be implanted. The implanted electrodes may be used independently or in addition to one or more electrodes placed external to the brain or skull.

[0048] A typical DBS electrode lead tip 70 as incorporated into the Medtronic (RTM) DBS lead model 3387 is shown in FIG. 8. The lead tip 70 includes a first electrode 72a, a second electrode 72b, a third electrode 72c, and a fourth electrode 72d. Each of the electrodes 72a, 72b, 72c, 72d, once implanted in the brain, can be used to apply stimulation to one or more neural structures or to monitor and optionally record evoked responses (including ERNA) from neural circuits to stimulation. In other embodiments, leads with more electrodes or electrodes having different sizes or topologies can be used. Additionally, one or more reference electrodes can be located at a remote site and used to complete an electrical circuit when one or more electrodes on the DBS lead are activated for stimulation or used for signal monitoring.

[0049] Target locations for the lead tips 70 vary depending on the neural structure. Examples of target structures include, but are not limited to, the subthalamic nucleus (STN), the substantia nigra pars reticulata (SNr), and the globus pallidus interna (GPi).

[0050] FIG. 9 shows a lead tip 70 implanted in a target structure in the brain, in this example the subthalamic nucleus (STN) 82. It will be appreciated that crossing the electrode tip 70 into the STN 82, which has a typical diameter of 5-6 mm, can be a very challenging surgical task. Currently, techniques such as stereotactic imaging, microelectrode recording, intraoperative x-ray imaging, and applying therapeutic stimulation while following up on the patient's symptoms are used to localize the electrode tip 70. However, these methods can lack precision. In addition, existing methods typically require the patient to be awake for the procedure, as a voluntary response from the patient can be used to confirm that the electrode is in the right place relative to the target structure in the brain. For this reason, many potential recipients of DBS therapy are uncomfortable having to be awake during a surgical procedure and reject the option.

[0051] The accuracy of the positioning of the electrode tip 70 within the target structure can be greatly improved by generating and measuring an evoked resonant response from the neural target using a series of patterned stimuli. Such techniques can eliminate the need for the patient to be awake during the implantation procedure, as the electrode can be placed much more precisely at the correct location in the brain relative to the target neural structure. This means that the patient can be sedated or under general anesthesia during the procedure, as no feedback from the patient is required to position the electrode with sufficient accuracy.

[0052] An example of a DBS delivery system 90 according to an embodiment of the present disclosure is shown in Figure 10. The system 90 includes the lead tip 70 of Figure 8 including integrated multiple electrodes 72a, 72b, 72c, 72d, along with a processing unit 92, a signal generator 94, measurement circuitry 96, and an optional multiplexer 98. The processing unit includes a central processing unit (CPU) 100, a memory device 102, and an input / output (I / O) bus 104 communicatively coupled to one or more of the CPU 100 and the memory device 102.

[0053] In some embodiments, a multiplexer 98 is provided to control whether the electrodes 72a, 72b, 72c, 72d are connected to the signal generator 94 and / or the measurement circuitry 96. In other embodiments, a multiplexer may not be required. For example, the electrodes 72a, 72b, 72c, 72d may instead be directly connected to both the signal generator 94 and the measurement circuitry 96. Although all of the electrodes 72a, 72b, 72c, 72d are shown connected to the multiplexer 98 in FIG. 10, in other embodiments, only one or some of the electrodes 72a, 72b, 72c, 72d may be connected.

[0054] The measurement circuitry 96 may include one or more amplifiers and digital signal processing circuitry including sampling circuitry for measuring neural responses including, but not limited to, ERNA. In some embodiments, the measurement circuitry 96 may also be configured to extract other information from the received signal, including local field potentials. The measurement circuitry 96 may also be used in conjunction with the signal generator 94 to measure electrode impedance. The measurement circuitry 96 may be external to the processing unit 92 or integrated therein. Communication between the measurement circuitry 96 and / or the signal generator 94 on the one hand and the I / O port on the other hand may be wired or by wireless coupling, e.g., via inductive coupling, WiFi (RTM), Bluetooth (RTM), etc. Power may be provided to the system 90 by at least one power source 106. The power source 106 may include a battery such that the elements of the system 90 can maintain power when implanted in a patient.

[0055] The signal generator 94 is coupled to one or more of the electrodes 72a, 72b, 72c, 72d via a multiplexer 98 and can be operated to deliver electrical stimulation to each electrode based on signals received from the processing unit 92. To this end, the signal generator 94, the multiplexer 98 and the processing unit 92 are also communicatively coupled so that information can be transferred therebetween. Although the signal generator 94, the multiplexer 98 and the processing unit 92 in FIG. 10 are shown as separate units, in other embodiments the signal generator 94 and the multiplexer may be integrated with the processing unit 92. Furthermore, any of the units may be implanted or located external to the patient's body.

[0056] The system 90 may further include one or more input devices 108 and one or more output devices 110. The input devices 108 may include, but are not limited to, one or more of a keyboard, a mouse, a touchpad, and a touchscreen. Examples of output devices include a display, a touchscreen, a light display device (LED), a sound generator, and a haptic generator. The input and / or output devices 108, 110 may be configured to provide feedback (e.g., visual, auditory, or haptic feedback) to the user, for example, regarding the characteristics of the ERNA or a later obtained indicator (e.g., the proximity of the electrode 70 to a neural structure in the brain). To this end, one or more of the input devices 108 may also be an output device 110, for example, a touchscreen or a haptic joystick. The input / output devices 108, 110 may also be wired or wirelessly connected to the processing unit 92. The input / output devices 108, 110 may be configured to provide the patient with control of the device (i.e., a patient controller) or to allow the clinician to program stimulation settings and receive feedback of the effect of stimulation parameters on the ERNA characteristics.

[0057] One or more elements of system 90 may be portable. One or more elements may be implantable within a patient. In some embodiments, for example, signal generator 94 and leads 70 may be implantable within a patient, and processing unit 92 may be external to the patient's skin and configured for wireless communication with the signal generator via RF transmission (e.g., induction, Bluetooth (RTM), etc.). In other embodiments, processing unit 92, signal generator 94, and leads 70 may all be implanted within the patient's body. In either case, signal generator 94 and / or processing unit 92 may be configured to wirelessly communicate with a controller (not shown) located outside the patient's body.

[0058] An embodiment of the present disclosure provides a system and method for localizing the lead tip 70 within a target structure of the brain using the measured ERNA. During the procedure to implant the lead tip 70 within the brain, instead of relying on less accurate positioning techniques such as those described above to estimate the location of the electrodes relative to neural structures within the brain, the system 90 can be used to provide real-time feedback to the surgeon based on characteristics such as the strength and quality of the evoked response signal received from one or more electrodes of the lead tip 70. This feedback can be used to estimate the location within the target structure in three dimensions and to inform decisions about whether the electrodes should be repositioned or removed and re-implanted along a different trajectory.

[0059] FIG. 11 shows a general example of such a process in which ERNA is measured at various electrode locations in response to stimulation at various amplitudes. The process begins at step 112 with the electrode lead tip as described with reference to FIG. 8 advanced intraoperatively along a predetermined trajectory toward the target neural structure. The stride size (or spatial resolution) at which the electrode lead is advanced may be selected by the surgeon and / or clinician. In some embodiments, the stride size is 1 mm. At step 112, patterned stimuli as described above are applied to the electrodes at amplitude A out of Y amplitudes, where Y is the total number of different amplitudes used for stimulation for the electrodes at the lead tip 70. The stimuli may be applied for the entire time the lead tip 70 is implanted. Alternatively, the patterned signal may be repeated a predetermined number of times, for example 10 times. The evoked responses may be measured at the same electrodes used to apply the stimuli or at one or more different electrodes. By doing so, a more accurate estimate of the location of each electrode relative to the target neural structure may be obtained.

[0060] After ERNA is measured in step 114, in step 115, if ERNA has been measured for all Y amplitudes of the patterned stimulation, the process continues to step 116. Otherwise, steps 112 and 114 are repeated at different amplitudes A through Y using the electrode(s) at the same location. The magnitude of the amplitude increments may be selected to provide the granularity of results required to allow accurate characterization of ERNA at various amplitudes, which may be patient-specific in nature. In some embodiments, the stimulation amplitudes are 0.3 mA, 0.67 mA, 1.5 mA, and 3.38 mA, as shown in FIG. 3 and FIGS. 5-7.

[0061] In step 116, stimuli at all Y amplitudes are applied and each ERNA is recorded, at which point the electrode wire is advanced in step 117. Steps 112 to 116 are then repeated until the tip of the electrode wire is inserted to the maximum allowable depth that may be in the target neural structure or slightly beyond it.

[0062] By repeating steps 112-116, evoked response profiles or maps can be generated at different locations along the insertion trajectory for stimulation of different amplitudes. The evoked response profiles can include measurements from multiple electrodes or from just one electrode. The evoked response profiles at different depths and different amplitudes can be output to one or more output devices 110. The evoked response profiles can then be compared in step 118 to determine whether a preferred electrode location can be identified. Identification of a preferred electrode location can be based on different ERNA features including relative differences or spatial derivatives of amplitude, rate of decay, rate of change, and frequency at different insertion locations (e.g., locations that result in maximum resonance), particularly at different stimulation amplitudes.

[0063] Identification of preferred electrode locations may also be based on comparison with model ERNA activity, the model being derived from recordings from other patients. The evoked response profile may also be used to estimate the trajectory of the electrode lead 70 through the target neural structure, including the structure's boundaries and intersecting regions (e.g., trajectory through medial or lateral regions). The evoked response profile may also be used to estimate the approach to the target structure in the event that the target structure does not intersect with the insertion trajectory.

[0064] If a preferred electrode location can be identified in step 120, the electrode lead tip 70 may be repositioned in step 122 to place the electrode at the preferred location. Alternatively, for embodiments including an electrode lead tip with multiple electrodes, the electrode placed closest to the preferred location may be designated for use in subsequent application of therapeutic stimulation. If a preferred location cannot be identified in step 120, the surgeon and / or clinician may choose to remove and re-implant the electrode along a different trajectory.

[0065] In some embodiments, the lead tip 70 used to determine optimal electrode placement (as described above) may be a temporary lead used only for that procedure, and thus, after identifying the optimal location for stimulation, the lead tip 70 may be removed and replaced with a permanent electrode or electrode array for long-term DBS therapy.

[0066] Another embodiment of the present disclosure provides a system and method for determining the relative position of an array of electrodes with respect to a target neural structure and for subsequently selecting preferred electrodes to be used for application of therapeutic stimulation. This process may be performed during electrode implantation surgery to aid in electrode placement or to assist previously implanted electrodes when programming a device to deliver therapeutic stimulation. Stimulation may be applied to more than one of the electrodes of the array, for example, two or more of electrodes 72a, 72b, 72c, 72d in the case of electrode array 70. When using a patterned stimulation scheme, sequential bursts of stimulation patterns may be applied to each of electrodes 72a, 72b, 72c, 72d. Alternatively, a full stimulation pattern may be applied at one electrode followed by another full stimulation pattern at another electrode. By doing so, a determination can be made as to which electrodes of the electrode array are best positioned for providing therapeutic stimulation to one or more of the target neural structures; for example, which of electrodes 72a, 72b, 72c, 72d is best positioned within the target neural structure.

[0067] FIG. 12 illustrates an example of a process 130 for measuring evoked responses from an array of multiple electrodes. In step 132, a stimulus of amplitude A of Y different amplitudes is applied to a first electrode of an array of X electrodes (e.g., electrode 72a in the case of lead tip 70). The applied stimulus may be a burst patterned stimulus as described above. An evoked response from a target neural structure is measured in step 134 at one or more of the electrodes in the array. For example, in the case of lead tip 70, evoked responses may be measured at the second, third and fourth electrodes 72b, 72c, 72d when the first electrode 72a is stimulated. Additionally or alternatively, the evoked response may be measured at the same electrode as the electrode being stimulated. In some embodiments, the evoked response received at one or more of the electrodes may be stored in a memory device. Once an evoked response has been measured at one or more electrodes, another electrode is selected for stimulation. This may be accomplished by incrementing a counter as shown in step 138 after process 130 checks in step 136 to see if all electrodes in the array of X electrodes have been stimulated, i.e., the process has cycled through all of the electrodes in the system. If any electrodes remain unstimulated, the process repeats applying stimulation to the next selected electrode in the array. When all electrodes in the array have been stimulated and evoked responses to the stimulation measured and recorded, another stimulation amplitude of the Y amplitudes may be selected for stimulation. This may be accomplished, for example, by incrementing a counter as shown in step 137 after process 130 checks in step 139 to see if all amplitudes Y have been cycled through. Steps 132-138 may then be repeated with the new stimulation amplitude A. If step 139 finds that all electrodes have been stimulated at all amplitudes, the resulting measured evoked responses are then processed in step 140.

[0068] Processing the evoked response may include comparing different ERNA features, including amplitude, rate of decay, rate of change, and relative differences or spatial derivatives of frequency, between different combinations of electrodes used for stimulation and measurement, and between each of the different stimulation amplitudes used for stimulation. For example, processing may include identifying the electrode with the greatest change in evoked resonance amplitude in response to a change in stimulation condition (e.g., amplitude), or identifying the electrode with the greatest reinstatement or change in reinstatement measured through increasing stimulation amplitudes. Identification of preferred electrode locations may also be based on comparison with exemplar ERNA activity. The exemplar may be derived from recordings from other patients or from one or more models or simulations.

[0069] Based on the processing of the evoked responses, preferred electrodes to use for therapeutic stimulation may be selected in step 142. Results of the ERNA processing and preferred electrode suggestions may be output to one or more output devices 110. If the process is performed intraoperatively, the results of the ERNA processing may also be used to determine which electrodes are in the target neural structures and whether the electrode array should be repositioned. The results may also be used to generate one or more exemplars for future processing of evoked responses in the same or different patients.

[0070] The evoked response may be measured at one or two or any number of electrodes in any configuration. For example, ERNA may be measured and / or recorded from different combinations of electrodes. Additionally or alternatively, the measurement electrodes may be implanted and / or placed outside the brain or skull.

[0071] The inventors have discovered that by comparing the evoked responses measured at each electrode in response to stimulation at another electrode, a determination can be made first as to whether any of the electrodes are located within the target neural structure, second as to whether any of the electrodes are located at an optimal location within the target neural structure, and third as to the orientation and / or distance of a particular electrode from the target neural structure. In some embodiments, one or more of the presence, amplitude, natural frequency, damping, rate of change, envelope, recurrence, and fine structure of the evoked resonant response (or superposition of resonant responses) to stimulation can be used to identify the most effective electrode in the electrode array. Preferably, changes in one or more of these characteristics in response to patterned stimulation at different amplitudes can be used to identify the most effective electrode in the electrode array. In addition, it has been found that the evoked response (and changes in the characteristics of the evoked response in response to different stimulation amplitudes) vary depending on the location of the electrode used for stimulation, illustrating the relevance of using the process shown in FIG. 11 to localize electrodes within the target neural structure.

[0072] Process 130 of FIG. 12 and other processes described herein may be implemented by varying stimulation parameters other than amplitude (e.g., with different stimulation frequencies) or using more than one stimulation electrode in step 132 (e.g., applying stimulation in parallel with multiple electrodes on one or more electrode leads). For example, electrical stimulation may be directed to a target location in the brain by applying stimulation from two or more spaced electrodes. Such electrodes may be placed at a location near but not at the target location. Current steering may be used to direct electrical stimulation to a target location or location in the brain such that the epicenter of electrical stimulation is centered at the target location or location. To do so, an amplitude bias may be applied to one or more of the two or more spaced electrodes to direct the current in a particular direction. Such current steering may be fixed or may be variable, allowing for two-dimensional or three-dimensional (with three or more spaced electrodes) movement of the focal point of electrical stimulation.

[0073] The resonant response to electrical stimulation may be measured at two or more spaced electrodes. Doing so may provide directional information related to the location or neural network that may be the source of the response. The response characteristics obtained by one or more electrodes may be used to aid in current steering (e.g., setting the distribution of current across simultaneously operating electrodes to direct therapeutic or other stimulation to a target location) and the selection of the active electrode (e.g., which electrode to use for stimulation). For example, as discussed above, the response characteristics may be used to estimate the spatial spread of activation relative to the target region. Using this information, a stimulation profile may be shaped using two or more electrodes to direct stimulation toward specific regions of the brain, i.e., the target structure, and away from areas where the clinician does not want stimulation.

[0074] In further embodiments, ERNA can also be used to optimize stimulation parameters used to target various medical conditions. For example, once an electrode array, such as lead tip 70, is precisely positioned within a target neural structure, measuring ERNA can aid in setting stimulation parameters for therapeutic DBS by improving accuracy and time and cost efficiency, and reducing undesirable side effects.

[0075] In addition to improving the accuracy of positioning of DBS electrodes in the brain, selecting electrode placement for stimulation, and optimizing stimulation parameters, ERNA can be used to generate feedback to control the stimulation of the electrodes. In some embodiments, the feedback can be achieved using a system 90 shown in FIG.

[0076] In one embodiment, the system 90 may use waveform exemplars corresponding to preferred patient conditions. The exemplars may be generated using previous ERNA recordings in patients with reduced symptoms. For example, ERNA exemplars recorded from medicated patients or patients undergoing effective stimulation treatment may be used. Alternatively, ERNA exemplars recorded from healthy patients, e.g., patients without movement disorders, may be used. The exemplars may be constructed from an average of many recordings from one patient or several patients. In some embodiments, selected features of the ERNA waveform may be used instead of the complete exemplar. For example, parameters of the ERNA, e.g., dominant frequency and amplitude components, and / or temporal features, may be used to allow improved electrode placement and control of therapeutic stimulation. In some embodiments, preferred ranges of different ERNA characteristics may be defined (e.g., stimulation is controlled so that the ERNA frequency remains within the range of 250-270 Hz).

[0077] In some embodiments, bursts of stimulation as described above may be used in combination with ERNA monitoring to identify therapeutic resonance states (e.g., states that correlate with good symptom suppression with minimal side effects and / or minimal power consumption). From this information, the therapeutic stimulation parameters required to produce the desired therapeutic state may be identified. In some embodiments, these parameters may be used to apply continuous therapeutic DBS to the target neural structures.

[0078] To reassess the resonant state, therapeutic DBS may be interspersed with probe bursts to identify resonant activity. These probe bursts may be performed on a periodic basis, for example, once every 10 seconds. In one embodiment, a probe burst may be applied for 1 second every 10 seconds (e.g., 10 pulses at 130 Hz) and ERNA may be evaluated. Therapeutic stimulation parameters may then be adjusted or maintained based on ERNA. For example, if there is a change in ERNA relative to the last probe burst, stimulation parameters may be adjusted so that ERNA is similar to the previously measured ERNA and / or exemplar ERNA and / or ERNA profile falls within a desired range.

[0079] FIG. 14 is a graphical comparison of a stimulation regime 178 (comprising one or more bursts of stimulation) including a patterned therapeutic stimulation signal 182 followed by a non-therapeutic patterned stimulation signal 184, and the corresponding ERNA characteristics (e.g., resonant frequency) 180 fluctuating between a preferred state 186 and a less preferred state 188.

[0080] There are various ways to achieve the patterned signals of the embodiments described herein. Figures 15a and 15b show two exemplary patterned profiles. In Figure 15a, the patterned profile includes a continuous stimulation block 190 followed by a no-stimulation period 192, which is followed by a burst of pulses 194 and another no-stimulation period 196. During the no-stimulation period, ERNA can be measured and the therapeutic stimulation signal can be adjusted (if necessary).

[0081] In an alternative embodiment, the system may monitor the ERNA after the last pulse 198 of the continuous stimulation, as shown in FIG. 15b. After the monitoring period 200, the therapeutic stimulation may be adjusted for a period 202, after which the therapeutic stimulation 198 may be applied with the adjusted parameters. This regimen may also be considered as continuous stimulation with periodically missing pulses. For this purpose, the continuous stimulation may be considered as a burst of pulses, and the no-stimulation period may be considered as the first period t1 described above with reference to FIG. 2.

[0082] In other embodiments, instead of omitting stimulation during the monitoring period 200, stimulation may be maintained but with parameter changes. For example, a conventional therapeutic stimulation at a frequency of, for example, 130 Hz may be applied during the therapeutic stimulation period 198. Then, during the monitoring period 200, a stimulation having a different frequency may be applied. The stimulation applied during the monitoring period may be lower than that applied during the stimulation period. For example, the stimulation frequency during this period may be in the region of 90 Hz. The frequency of this stimulation during the monitoring period 200 may be low enough to allow for observing multiple ERNA peaks. Similarly, the frequency of the stimulation applied during the monitoring period 200 may be high enough to fall within the therapeutic frequency range for DBS. As mentioned above, the transition between frequencies may be abrupt, or the change in frequency may be gradual. It may be beneficial to ramp the frequency of the pulses to avoid abrupt step changes in frequency.

[0083] Additionally or alternatively, the amplitude of the stimulation applied during the monitoring period 200 may be different than that applied during the treatment period 198. For example, the amplitude of the stimulation applied during the monitoring period 200 may be less than that applied during the therapeutic stimulation 198. To avoid abrupt step changes in amplitude, the stimulation may be transitioned with an amplitude ramp between the treatment period 198 and the monitoring period 200 through several pulses.

[0084] In some embodiments, characteristics of the stimulation other than amplitude and frequency may vary between the stimulation period 198 and the monitoring period. Examples of such characteristics include, but are not limited to, stimulation frequency, amplitude, pulse width, net charge, electrode placement or configuration.

[0085] While the above embodiments use a single electrode array for both stimulating and recording evoked neural responses, in other embodiments, electrodes may be distributed over multiple probes or leads in one or more target structures in one or both brain hemispheres. Similarly, electrodes either implanted or placed outside the brain may be used to stimulate or record, or both stimulate and record, evoked neural responses. In some embodiments, a combination of both microelectrodes and macroelectrodes may be used in any foreseeable manner.

[0086] In further applications of embodiments of the present invention, ERNA measurements may be recorded and tracked over time to monitor the progression or remission of a disease or syndrome, or may be used as a diagnostic tool (e.g., to classify a patient's neurological condition). Such embodiments may also be used to provide a medical alert to a patient, caregiver, or clinician should the patient's condition (as determined by ERNA) deteriorate into an undesirable or serious condition (e.g., a parkinsonian attack).

[0087] In yet another application, ERNA can be used to monitor the effects of medication over time, including effects such as adjusting medication doses. Such embodiments can also be used to provide medication alerts to patients to remind them when a dose is needed or when a dose is missed. Tracking medication effects using ERNA can also provide clinicians with information regarding whether medication is being taken as prescribed, or whether the medication is becoming less effective and needs to be adjusted.

[0088] Further analysis of ERNA and explanation of results DBS induces resonant neural activity The neural activity produced by the DBS pulses was investigated to determine whether there were any evoked responses that could be feasibly used as biomarkers. To preserve the evoked activity, we used a wide recording bandwidth and a symmetric biphasic pulse for stimulation instead of the conventional asymmetric pulse with a very long second phase to minimize the temporal duration of the stimulation artifact.

[0089] Because PD is the primary application of DBS, recordings were made from DBS electrodes immediately implanted within the STN of patients with PD who were still awake on the operating table. Moreover, the role of the STN in regulating movement, limb, and associated functions makes it a neural target with important connections to several different applications, including DBS treatment of dystonia, essential tremor, epilepsy, and obsessive-compulsive disorder.

[0090] STN-DBS was found to induce a large peak after each pulse, typically around 4 ms. By examining the activity after the last pulse before discontinuing DBS, it was discovered that this peak was the first in a series with decreasing amplitude. Because this response has a shape similar to a damped oscillation, we describe it as evoked resonant neural activity (ERNA).

[0091] To further investigate ERNA, we patterned standard 130Hz DBS in time so that multiple peaks could be observed. We employed two novel patterns: one that thinned out one pulse per second, and one that applied bursts of 10 pulses per second. The "thinned pulse" pattern was expected to have a therapeutic effect comparable to standard 130Hz DBS, since it caused only a 0.77% decrease in the total number of pulses delivered over time. In contrast, the "burst" pattern was expected to have the smallest therapeutic effect compared to continuous DBS, since only 7.7% of the pulses were delivered, making it a useful probe to investigate activity in the absence of therapy. Evoked responses tend to increase in amplitude and become more intense between successive pulses of the burst, reaching a steady state with longer durations of stimulation.

[0092] We applied burst stimulation to the STN of 12 PD patients (n=23 hemispheres) undergoing DBS implantation and found ERNA with similar morphology in all cases, indicating that it is a robust and reliable signal that can be measured across patient populations. As a control to ensure that ERNA was not a spatial artifact, we also applied burst stimulation to three essential tremor patients (n=6 hemispheres) with electrodes implanted in the posterior subthalamic area (PSA), a white matter region medial to the STN. No ERNA was found in the PSA, suggesting that it is an electrophysiological response that can be localized to the STN.

[0093] ERNA can be localized to the STN To establish that ERNA varies with electrode location relative to the STN, we applied consecutive 10-second bursts of stimulation to each of the four DBS electrodes while recording from three non-stimulating electrodes. The DBS implantation procedure was aimed at placing two of the four electrodes within the STN, one electrode in the dorsal STN, where DBS benefits are usually greatest, and the other in the ventral STN. This variation in electrode location facilitated the comparison of ERNA responses from different regions of the STN with those outside the nucleus. In eight STN-PD patients (n=16 hemispheres), we found that both ERNA amplitude and morphology varied with the location of the stimulating and recording electrodes. Figure 14 shows an example of ERNA from the last pulse of each burst in one hemisphere, but the response with the largest amplitude and the clearest decaying oscillatory morphology occurred at the two middle electrodes at the target STN location. As a control, we also applied stimulation to two essential tremor patients (n=4 hemispheres) using implanted tracks intended to place a distal electrode in the PSA and a proximal electrode in the ventral intermediate thalamic nucleus, another target for tremor previously shown not to elicit evoked activity beyond approximately 2 ms.

[0094] Variations in ERNA amplitude were the clearest feature, so we used them for further analysis. Because not all recordings were in the STN and contained unique resonant activity, we calculated the root mean square (RMS) voltage from 4 to 20 ms to quantify ERNA amplitude. To estimate the implanted electrode location relative to the STN, a 3D reconstruction (FIG. 16) was created using a fusion of the postoperative CT scan and preoperative MRI (FIG. 17). Electrodes were classified as being superior, inferior, or internal to the STN based on blind measurements relative to the red nucleus. Electrodes within the STN were then further classified as being either dorsal or ventral.

[0095] ERNA amplitudes varied significantly with electrode position (Kruskal-Wallis, H(4)=45.73, p<0.001), with only the inferior electrodes not significantly different from the PSA area (p=0.370) (FIG. 18). Although the dorsal electrodes tended to be higher compared to the ventral and superior ones, they did not differ significantly from each other in amplitude. To account for the variation in amplitudes between patients due to variability in medial-lateral and anterior-posterior electrode positioning, as well as underlying differences in patient physiology, we reanalyzed the recordings from the STN-DBS electrodes after normalizing the responses to total ERNA amplitude across each hemisphere. After normalization, post-hoc comparisons revealed significant differences in amplitude between the dorsal and ventral STN (Kruskal-Wallis, H(3) = 14.94, p = 0.002; post-hoc Dunn's test, ventral vs. dorsal: p = 0.043, superior vs. dorsal: p = 0.081, inferior vs. dorsal: p = 0.002).

[0096] These results indicate that ERNA can be localized to the STN and varies across the STN, establishing its use as a feedback signal to guide electrode implantation to the most beneficial sites for stimulation. Furthermore, although variation in amplitude was the most obvious feature, other ERNA traits, such as frequency, latency, and rate of change, also have potential utility in discriminating STN regions.

[0097] ERNA is modulated by DBS To investigate whether ERNA was modulated by therapeutically effective DBS, we applied thinned-pulse stimulation of increasing current amplitude (range 0.67-3.38 mA) in 60-90 s blocks to 10 STN-PD patients (n=19 hemispheres). Overall, we found that the second and subsequent peaks in ERNA were consistently observed to spread apart over time and with increasing stimulation amplitude, with an asymptotic increase in latency consistent with a decrease in the frequency of resonant activity (Figs. 5, 6, 21a-21e). In many cases, the latency of the first peak also increased, although this change was not consistent across all recordings. The amplitude of the peaks was also generally variable, usually being larger at the beginning of each stimulation block and gradually weakening thereafter.

[0098] To quantify these effects, we calculated the inverse of the difference in latency between the first and second peaks as a representative measure of ERNA frequency. We also calculated the amplitude difference between the first peak and first valley as a representative measure of ERNA amplitude. We then used the average of the 45-60 second period for each condition as an estimate of the asymptotic ERNA value for analysis.

[0099] ERNA frequency was significantly decreased between conditions (one-way repeated measures (RM) ANOVA, F(4,94) = 45.79, p < 0.001). Post-hoc comparisons (Holm-Sidak) showed that ERNA frequency was significantly decreased with each step increase in DBS amplitude, except from 2.25 mA to 3.38 mA (p = 0.074) (Figure 19A). The median frequency was 256 Hz at 3.38 mA, approximately twice the stimulation rate of 130 Hz. It was proposed that STN-DBS may act by producing a pacing effect in the internal pallidal segment at twice the stimulation rate due to the time course of immediate excitation of STN axons and inhibition / recovery of STN somas.

[0100] ERNA amplitudes also differed significantly between conditions (Friedman, x2 (4)=41.31, p<0.001). Tukey's post-hoc comparisons showed that ERNA amplitude initially increased with DBS amplitude before plateauing at levels above 1.5 mA (FIG. 19B). Although these effects may be related to the therapeutic effects of DBS, they may alternatively or even be due to saturation of neural firing in the STN. For this reason, we next focused on ERNA frequency for correlation with therapeutic effects.

[0101] ERNA correlates with therapeutic efficacy The clinical efficacy of stimulation was confirmed by assessing limb bradykinesia and rigidity according to the Unified Parkinson's Disease Rating Scale (UPDRS, items 22 and 23) immediately before and 60 seconds after stimulation at 2.25 mA. Both clinical signs improved significantly at 2.25 mA, indicating that DBS was therapeutic (Wilcoxon signed rank, bradykinesia: Z=-3.62, p<0.001, rigidity: Z=-3.70, p<0.001).

[0102] However, time constraints prevented clinical testing at each level of stimulation intensity. Therefore, to correlate ERNA modulation with patient status, we used spontaneous LFP activity in the beta band (13-30 Hz). Excessive synchronization of oscillations in the beta band has been implicated in the pathophysiology of PD, and its suppression correlated with improvements in bradykinesia and rigidity.

[0103] Using a short-time Fourier transform, we calculated "relative beta," the RMS amplitude in the 13-30 Hz band divided by that in the 5-45 Hz band, as a representative measure of beta activity. We then averaged the values ​​over the 45-60 second period for each stimulation amplitude condition and found that relative beta varied significantly (one-way RM ANOVA, F(4,89)=18.11, p<0.001). Post-hoc tests (Holm-Sidak) showed significant suppression at 3.38 mA compared to all other conditions, and at 2.25 mA compared to 0.67 and 1 mA (Figure 19C). These results are consistent with previous studies showing that suppression of beta activity by DBS correlates with improvements in clinical signs, supporting the idea that stimulation was therapeutically effective above approximately 2.25 mA.

[0104] To further correlate ERNA frequency with beta activity and therefore therapeutic efficacy, we compared the averages of 15-s non-overlapping blocks between conditions (FIG. 19B). ERNA frequency was significantly correlated with relative beta (Pearson product moment, ρ=0.601, n=90, p<0.001).

[0105] These results indicate that ERNA is a clinically significant biomarker. Moreover, its large amplitude, ranging from 20 to 681 μVp-p (median 146 μVp-p), is an order of magnitude larger than spontaneous beta LFP activity, whose absolute values ​​range from 0.9 to 12.5 μV RMS (median 2.2 μV RMS). The robustness of ERNA and its unique graded modulated morphology (Figures 5, 6, 7, 21) contrasts with the inherent interpatient variability in the beta band and the noisy, intermittent nature of the beta band signal. Thus, ERNA is a more manageable signal for use with fully implanted DBS devices compared to noisy, low amplitude LFP measures.

[0106] ERNA modulation washes out subsequent DBS We also applied 60-s bursts of stimulation immediately before and after the therapeutic thinned-pulse stimulation (hereafter referred to as pre- and post-DBS conditions) to monitor changes in activity as the therapeutic effects were washed out.

[0107] Overall, ERNA remained relatively stable before DBS, indicating that the modulating effects of burst stimulation were minimal. However, immediately after DBS, ERNA peaks occurred at longer latencies and then gradually returned to its pre-DBS state. To quantify these effects, we averaged post-DBS ERNA frequency and amplitude across 15-second non-overlapping blocks and compared them to the final 15 seconds before DBS. Differences in ERNA frequency were observed for all hemispheres examined (n=19) (Friedman, χ 2 (4) = 70.23, p < 0.001), and frequencies were significantly reduced compared to pre-DBS frequencies at all time points except for the final 45–60 s block (Tukey, p = 0.73) (Figure 20A). ERNA amplitudes also varied significantly between time points (Friedman, χ 2 (4)=31.37, p<0.001), the difference between the amplitudes at the post-DBS time points indicated washout of the amplitude suppression caused by the therapeutic stimulation (FIG. 20B). Because the applied burst stimulation was constant between the pre-DBS and post-DBS conditions, the observed fluctuations in ERNA frequency and amplitude can be directly attributed to changes in the state of the STN neural circuitry as the DBS effect is washed out. Thus, therapeutically effective DBS modulated both ERNA frequency and amplitude, indicating that ERNA has multiple attributes that can be feasibly used as a biomarker and as a tool to explore mechanisms of action.

[0108] We next assessed relative beta activity pre- and post-DBS and found it to be significantly different between time points (Friedman, χ 2(4) = 24.55, p < 0.001). Consistent with previous reports, relative beta was significantly decreased immediately after DBS, washing out to pre-DBS levels after 30 seconds (Fig. 20C). Supporting the results of the thinning pulse, ERNA frequency was significantly correlated with relative beta pre- and post-DBS (Pearson product moment, ρ = 0.407, n = 152, p < 0.001). ERNA amplitude was also correlated with relative beta (Pearson product moment, ρ = 0.373, n = 152, p < 0.001), suggesting that it too may have clinical and mechanistic importance.

[0109] We further analyzed spontaneous LFP activity in the high frequency oscillation (HFO) band (200-400 Hz), which overlaps with the observed ERNA frequency. Changes in the HFO band have been implicated in the mechanism of action of DBS, along with motor status and effective pharmacological therapy, especially beta activity. The use of burst stimulation allowed for simultaneous ERNA and HFO analysis, since the data could be split to include only the activity between bursts, resulting in LFP epochs that were free of stimulation artifacts that could otherwise obscure the integrity of the HFO band.

[0110] Because HFO activity was generally characterized by broadband peaks in frequency, we calculated multitaper spectral estimates and then determined the frequency and amplitude of peaks occurring between 200 and 400 Hz. By comparing means in non-overlapping 15-s blocks (Figure 20D), we found that after DBS, HFO peak frequency was significantly reduced all the way through the final 45-60-s block (Tukey, p=0.077) (Friedman, χ 2(4)=45.18, p<0.001). This washout trend was consistent with that for ERNA frequency, albeit at lower frequencies, and there was a significant correlation between the two (Pearson product moment, ρ=0.546, n=152, p<0.001). The median HFO peak frequency immediately after DBS was 253 Hz, which was similar to the ERNA frequency (256 Hz) in the therapeutic 3.38 mA condition, suggesting that HFO activity occurs at the same frequency as ERNA during longer, consecutive thinned-pulse stimulation.

[0111] No significant differences were observed in HFO peak amplitude (Friedman, χ 2 (4) = 2.11, p < 0.72), but it was significantly correlated with ERNA amplitude (Pearson product moment, ρ = 0.429, n = 152). The very small amplitude of the HFO peak (< 1 μV) may have meant that any modulation effects were obscured by noise in the recordings.

[0112] It will be appreciated by those skilled in the art that numerous variations and / or modifications may be made to the above-described embodiments without departing from the broad overall scope of the present disclosure, and the present embodiments are therefore to be considered in all respects as illustrative and not restrictive. [Explanation of symbols]

[0113] 72a 1st electrode 72b 2nd electrode 72c 3rd electrode 72d 4th electrode 82 Subthalamic nucleus (STN) 90 DBS Delivery System 92 Processing Unit 94 Signal Generator 96 Measurement circuit 98 Multiplexer 100 CPU 102 Storage device 104 Output (I / O) Bus 106 Power supply 108 Input Devices 110 Output Device

Claims

1. A non-therapeutic method comprising: a. applying a plurality of electrical stimuli having different amplitudes via a first electrode at a first location in the patient's brain, wherein a total energy delivered to the patient by the plurality of electrical stimuli is below a therapeutic threshold; b. detecting a resonant response from a target neural structure in the brain for each of the plurality of electrical stimuli; each resonant response being evoked by a respective one of said electrical stimuli; c. determining a change in one or more common waveform characteristics between two or more of the detected resonant responses; and d. determining the effectiveness of delivery of a therapeutic stimulus at the first location based on the change; Non-therapeutic methods, including:

2. The one or more common waveform characteristics include: i) the amplitude of the resonant response; ii) the re-emergence of said resonant response; iii) the frequency of the resonant response; iv) the time envelope of said resonant response; v) the fine structure of the resonant response; vi) the rate of decay of the resonant response; vii) a delay time between application of each of the plurality of electrical stimuli and the onset of a temporal characteristic of the resonant response; 2. The non-therapeutic method of claim 1, comprising one or more of the following:

3. 1. A neurostimulation system comprising: at least one electrode adapted for implantation in or near a target neural structure in the brain; a signal generator selectively coupled to one or more of the at least one electrode and configured to apply a plurality of electrical stimuli having different amplitudes via a first one of the at least one electrodes at a first location in the patient's brain; a measurement device selectively coupled to one or more of the at least one electrode and configured to detect a resonant response from the target neural structure evoked by each of the plurality of electrical stimuli; and a processing unit coupled to the measurement device and configured to determine a change in one or more common waveform characteristics among two or more of the detected resonant responses and to determine effectiveness of delivery of a therapeutic stimulus at the first location based on the change; The neurostimulation system.

4. The one or more common waveform characteristics include: i) the amplitude of the resonant response; ii) the re-emergence of said resonant response; iii) the frequency of the resonant response; iv) the time envelope of said resonant response; v) the fine structure of the resonant response; vi) the rate of decay of the resonant response; vii) a delay time between application of each of the plurality of electrical stimuli and the onset of a temporal characteristic of the resonant response; The system of claim 3 , comprising one or more of:

5. The system of claim 4 or 5, wherein the processing unit is further configured to select an amplitude of the different amplitudes for therapeutic stimulation of the target neural structure based on the variation.

6. The system of any one of claims 3 to 5, wherein the processing unit is further configured to select the first location for therapeutic stimulation of the target neural structure based on the change.

7. the at least one electrode includes a first electrode at the first location and a second electrode at a second location different from the first location; the signal generator configured to apply the plurality of electrical stimuli to each of the first and second electrodes; the measurement device is configured to detect the resonant response from the target neural structure evoked by the plurality of electrical stimuli applied to each of the first and second electrodes; The processing unit comprises: determining a change in one or more common waveform characteristics among two or more of the detected resonant responses evoked by each of the plurality of electrical stimuli applied to the second electrode at the second location; comparing the change in the one or more common waveform characteristics among the two or more of the detected resonant responses at the first electrode to a change in one or more common waveform characteristics among two or more of the detected resonant responses at the second electrode; and selecting one of the first electrode and the second electrode for therapeutic stimulation of the target neural structure based on the comparison; It is composed of A system according to any one of claims 3 to 6.

8. 8. The system of claim 6 or 7, wherein the signal generator is further configured to apply therapeutic stimulation to the target neural structure via the selected electrodes.

9. the signal generator is configured to apply the plurality of electrical stimuli to the first electrode at a second location in the brain that is different from the first location in response to movement of the first electrode to the second location; the measurement device is configured to detect the resonant response from the target neural structure evoked by each of the plurality of electrical stimuli applied to the first electrode at the second location; The processing unit comprises: determining a change in one or more common waveform characteristics among two or more of the detected resonant responses evoked by each of the plurality of electrical stimuli applied to the first electrode at the second location; comparing the change in the one or more common waveform characteristics between the two or more of the detected resonant responses detected at the first location to a change in the one or more common waveform characteristics between the two or more of the detected resonant responses detected at the second location; and determining optimal electrode locations for therapeutic stimulation of the target neural structure based on said comparison; It is composed of A system according to any one of claims 3 to 5.

10. The system of any one of claims 3 to 9, wherein the resonant response is detected at two or more of the at least one electrode.

Images