Nerve sensing in implantable stimulator device during passive charge recovery

Abstract

The invention discloses nerve sensing in an implantable stimulator device during passive charge recovery. Techniques are provided for sensing a neural response, such as evoked compound action potential (ECAP), in an implantable stimulator device. The first treatment pulse phase is followed by a charge recovery phase comprising at least one high impedance passive charge recovery duration. The ECAP is sensed during a high impedance passive charge recovery duration. The period of passive charge recovery is extended, and the high impedance passive charge duration completely overlaps with the ECAP (i.e., neural response duration) at the sensing electrode.

Description

[0001] This application is a divisional application of the patent application with the application number of "202080069764.X", the application date of "August 4, 2020", and the title of "Neural Sensing in an Implantable Stimulator Device during Passive Charge Recovery". Technical Field

[0002] This application relates to implantable medical devices (IMDs), and more particularly to circuitry for assisting in sensing neural signals in an implantable stimulator device. Background Art

[0003] Implantable nerve stimulator devices are devices that generate and deliver electrical stimulation to body nerves and tissues to treat various biological disorders, such as pacemakers for treating arrhythmias, defibrillators for treating cardiac fibrillation, cochlear stimulators for treating deafness, retinal stimulators for treating blindness, muscle stimulators for generating coordinated limb movement, spinal cord stimulators for treating chronic pain, cortical and deep brain stimulators for treating movement and psychological disorders, and other nerve stimulators for treating urinary incontinence, sleep apnea, shoulder subluxation, etc. The following description typically focuses on the use of the present invention in a spinal cord stimulation (SCS) system, such as that disclosed in U.S. Patent No. 6,516,227. However, the present invention may be found applicable to any implantable nerve stimulator device system.

[0004] An SCS system typically includes Figure 1 the implantable pulse generator (IPG) 10 shown in. The IPG 10 includes a biocompatible device housing 12 that houses a battery 14 and circuitry for powering the operation of the IPG. The IPG 10 is coupled to tissue stimulation electrodes 16 via one or more electrode leads forming an electrode array 17. For example, one or more percutaneous leads 15 may be used, which have annular or split-ring electrodes 16 carried on a flexible body 18. In another example, paddle leads 19 provide electrodes 16 positioned on one of their generally flat surfaces. Lead wires 20 within the leads are coupled to the electrodes 16 and proximal contacts 21 that can be inserted into a lead connector 22 in a head 23 fixed to the IPG 10, which may include, for example, epoxy resin. Once inserted, the proximal contacts 21 connect to head contacts 24 within the lead connector 22, which in turn are coupled to the stimulation circuitry 28 within the housing 12 by feedthrough pins 25 through a housing feedthrough 26.

[0005] In the illustrated IPG 10, there are 32 electrodes (E1 - E32), separated among four percutaneous leads 15 or incorporated in a single paddle lead 19, and thus, the head 23 can include a 2x2 array of eight - electrode lead connectors 22. However, the type and number of leads in the IPG and the number of electrodes are application - specific and can thus vary. The conductive housing 12 can also include electrodes (Ec). In an SCS application, typically one or more electrode leads are implanted into the spine of the patient near the dura mater of the spinal cord, preferably spanning the left and right sides of the patient's spine. The proximal contacts 21 tunnel through patient tissue to a distant location, such as the buttock where the IPG housing 12 is implanted, at which point they are coupled to the lead connector 22. In other examples of IPGs designed for direct implantation at the site requiring stimulation, the IPG can be leadless, with its electrodes 16 alternatively appearing on the body of the IPG 10 for contacting patient tissue. In other solutions, one or more IPG leads can be integrated with and permanently connected to the IPG 10. The goal of SCS therapy is to provide electrical stimulation from the electrodes 16 to relieve the patient's symptoms, such as chronic back pain.

[0006] The IPG 10 can include an antenna 27a to allow two - way communication with a plurality of external devices for programming or monitoring the IPG, such as a handheld patient controller or a clinician programmer, as described, for example, in U.S. Patent Application Serial No. 16 / 210,794 filed on December 5, 2018. As illustrated, the antenna 27a includes a conductive coil within the housing 12, although a coil antenna 27a can also appear in the head 23. When the antenna 27a is configured as a coil, communication with external devices is preferably effected using near - field magnetic induction. The IPG 10 can also include a radio - frequency (RF) antenna 27b. In Figure 1 the RF antenna 27b is shown within the head 23, but it can also be within the housing 12. The RF antenna 27b can include a patch, slot, or wire and can operate as a monopole or dipole. The RF antenna 27b preferably uses far - field electromagnetic waves for communication and can operate according to any number of known RF communication standards such as Bluetooth, Zigbee, WiFi, MICS, and the like.

[0007] Stimulation in the IPG 10 is typically provided by pulses, each of which can include multiple phases, such as 30a and 30b, as Figure 2Aas shown in the example of. Stimulation parameters typically include amplitude (current I, although voltage amplitude V can also be used); frequency (F); pulse width (PW) of a pulse or its individual phases; the electrodes 16 selected to provide the stimulation; and the polarities of these selected electrodes, i.e., whether they act as anodes that pull (source) current from the tissue or cathodes that sink (sink) current into the tissue. These and possibly other stimulation parameters together comprise a stimulation program that the stimulation circuit 28 in the IPG 10 can execute to provide therapeutic stimulation to the patient.

[0008] In Figure 2A the example of, electrode E4 has been selected as the anode (during its first phase 30a), and thus provides a pulse that pulls a positive current of amplitude +A from the tissue. Electrode E5 has been selected as the cathode (also during the first phase 30a), and thus provides a pulse that sinks a corresponding negative current of amplitude -A into the tissue. This is an example of bipolar stimulation, where only two lead-based electrodes are used to provide stimulation to the tissue (one anode, one cathode). However, more than one electrode can be selected to act as the anode at a given time, and more than one electrode can be selected to act as the cathode at a given time.

[0009] As mentioned, the IPG 10 includes a stimulation circuit 28 to form a prescribed stimulation at the patient's tissue. Figure 3 An example of the stimulation circuit 28 is shown, which includes one or more current source circuits 40 i and one or more current sink circuits 42 i . The source circuit 40 i and the sink circuit 42 i can include digital-to-analog converters (DACs), and can be referred to as PDAC 40 i and NDAC 42 i respectively according to the positive (source, anode) current and negative (sink, cathode) current they emit. In the example shown, NDAC / PDAC 40 i / 42 i are paired and dedicated (hardwired) to specific electrode nodes ei 39. For reasons explained below, each electrode node ei 39 is connected to the electrode Ei 16 via a DC-blocking capacitor Ci 38. The stimulation circuit 28 in this example also supports selecting the conductive housing 12 as an electrode (Ec12), typically selected for monopolar stimulation. PDAC 40 i and NDAC 42 i can also include voltage sources.

[0010] For PDAC 40 i and NDAC 42 iAppropriate control allows any of the electrodes 16 to act as an anode or a cathode to generate a current through the patient tissue R, desiring good therapeutic effects. In the example shown ( Figure 2A ), and during the first phase 30a (where electrodes E4 and E5 are selected as the anode and cathode respectively), PDAC 404 and NDAC 425 are activated and digitally programmed to generate a desired current A with the correct timing (e.g., at a prescribed frequency F and pulse width PWa). During the second phase 30b (PWb), PDAC 405 and NDAC 424 will be activated to reverse the polarity of the current. More than one anode electrode and more than one cathode electrode can be selected simultaneously, and thus the current can flow through the tissue R between two or more of the electrodes 16.

[0011] The power for the stimulation circuit 28 is provided by the compliance voltage VH. As described in further detail in U.S. Patent Application Publication 2013 / 0289665, the compliance voltage VH can be generated by a compliance voltage generator 29, which can include circuitry for boosting the voltage (Vbat) of the battery 14 to a voltage VH sufficient to drive a prescribed current A through the tissue R. The compliance voltage generator 29 can include an inductor-based boost converter as described in the '665 publication, or can include a capacitor-based charge pump. Since the resistance of the tissue is variable, VH can also be variable and can be up to 18 volts in one example.

[0012] Other stimulation circuits 28 can also be used in the IPG 10. In an example not shown, a switch matrix can intervene between one or more PDAC 40 i and the electrode node ei 39, and intervene between one or more NDAC 42 i and the electrode nodes. The switch matrix allows one or more of the PDACs or one or more of the NDACs to be connected to one or more anode or cathode electrode nodes at a given time. Various examples of stimulation circuits can be found in U.S. Patent 6,181,969, 8,606,362, 8,620,436, U.S. Patent Application Publication 2018 / 0071520, and U.S. Patent Application Serial No. 16 / 131,809 filed on September 14, 2018. Figure 3 Most of the stimulation circuit 28 (including PDAC 40 i and NDAC 42 i, the switch matrix (if present) and the electrode nodes ei 39) can be integrated on one or more application specific integrated circuits (ASICs), as described in U.S. Patent Application Publications 2012 / 0095529, 2012 / 0092031, and 2012 / 0095519, which are incorporated herein by reference. As explained in these references, one or more ASICs can also include other circuits useful in the IPG 10, such as telemetry circuits (for interfacing with telemetry antennas 27a and / or 27b outside the chip), compliance voltage generators 29, various measurement circuits, etc.

[0013] Figure 3 Also shown is a DC-blocking capacitor Ci 38 in the electrode current path placed in series between each of the electrode nodes ei 39 and the electrodes Ei 16 (including the can electrode Ec 12). The DC-blocking capacitor 38 serves as a safety measure to prevent DC current injection into the patient, such as may occur if there is a circuit fault in the stimulation circuit 28. The DC-blocking capacitor 38 is typically located outside the chip (outside one or more ASICs), and alternatively can be provided in or on a circuit board within the IPG 10 for integrating its various components, as explained in U.S. Patent Application Publication 2015 / 0157861.

[0014] Although not shown, the circuits in the IPG 10 including the stimulation circuit 28 can also be included in an external trial stimulator (ETS) device that is used to simulate the operation of the IPG during a trial period and prior to implantation of the IPG 10. The ETS device is typically used after the leads in the electrode array 17 have been implanted in the patient. The proximal ends of the leads in the electrode array 17 pass through the patient's incision and are connected to an externally worn ETS, allowing the ETS to provide stimulation to the patient during the trial period. More details regarding the ETS device are described in USP9,259,574 and U.S. Patent Application Serial No. 16 / 210,794, filed December 5, 2018.

[0015] Referring again to Figure 2A, the stimulation pulses shown are biphasic, where each pulse at each electrode includes a first phase 30a, followed by a second phase 30b of opposite polarity. (Although not shown, it is well known that a short interphase period can intervene between phases 30a and 30b during which no current is actively driven by the DAC circuits 40 / 42, which allows the DAC circuits time to switch between phases). The biphasic pulses help to actively recover any charge that may be stored on capacitive components (such as the DC-blocking capacitor 38, the electrode / tissue interface, or within the tissue itself) in the electrode current path. To recover all charge (Vc4 = Vc5 = 0V) at the end of the second pulse phase 30b of each pulse, the first phase 30a and the second phase 30b are preferably charge-balanced at each electrode, where these phases include an equal number but opposite-polarity charges. In the example shown, for each of the pulse phases 30a and 30b, this charge balance is achieved by using the same pulse width (PWa = PWb) and the same amplitude (|+A| = |-A|). However, as is known, if the product of the amplitudes and pulse widths of the two phases 30a and 30b is equal, the pulse phases 30a and 30b can also be charge-balanced.

[0016] Figure 3 It is shown that the stimulation circuit 28 can include a passive recovery switch 41 i , which is further described in U.S. Patent Application Publications 2018 / 0071527 and 2018 / 0140831. The passive recovery switch 41 i can be attached to each of the electrode nodes 39 and is used to passively recover any charge remaining on the DC-blocking capacitor Ci 38 after the second pulse phase 30b is issued - that is, to recover the charge without using the DAC circuits to actively drive current. Passive charge recovery can be prudent because non-idealities in the stimulation circuit 28 may result in pulse phases 30a and 30b that are not fully charge-balanced. By closing the passive recovery switch 41 i , passive charge recovery typically occurs after the actively driven phases 30a and 30b have been completed and during at least a portion 30c of the quiet period between pulses ( Figure 2A ). As shown in Figure 3 , the other end of the switch 41 that is not coupled to the electrode node 39 i is connected to a common reference voltage V CM . The common reference voltage V CM can be, for example, VH / 2 or can be another voltage, such as Vbat. As explained in the references cited above, passive charge recovery tends to occur by placing the capacitor in parallel with the reference voltage (V CM) to balance the charge on the DC-blocking capacitor 38 and other capacitive components between the [device] and the patient tissue. Note that passive charge recovery is shown as a small exponential decay curve during 30c in Figure 2A and can be positive or negative, depending on whether the pulse phase 30a or 30b has a charge advantage at a given electrode. Summary of the Invention

[0017] Aspects of the present disclosure relate to stimulator devices. According to some embodiments, a stimulator device includes: a plurality of electrode nodes, each electrode node being configured to be coupled to one of a plurality of electrodes configured to contact patient tissue; a stimulation circuit configured to provide actively driven stimulation at at least one stimulation node selected from the plurality of electrode nodes, wherein the stimulation includes at least one pulse, and the at least one pulse includes at least a first phase; and a passive charge recovery circuit configured to provide passively driven passive charge recovery during a passive charge recovery duration, wherein the passive charge recovery circuit includes a resistance circuit, and the resistance circuit is configurable to adjust a recovery impedance during the passive charge recovery duration; and a sensing circuit configured to sense a nerve response at at least one sensing node selected from the plurality of electrode nodes during the passive charge recovery duration. According to some embodiments, the resistance circuit includes a variable resistance circuit. According to some embodiments, the variable resistance circuit is configured to provide a first recovery impedance for a high impedance portion of the passive charge recovery duration and a second recovery impedance for a low impedance portion of the passive charge recovery duration, wherein the first recovery impedance is greater than the second recovery impedance. According to some embodiments, the sensing circuit is configured to sense the nerve response during the high impedance portion of the passive charge recovery duration. According to some embodiments, the passive charge recovery circuit includes a plurality of switch circuits, wherein each of the plurality of switch circuits is coupled to a different one of the electrode nodes and is configured to provide a variable impedance between its corresponding electrode node and a common node when selected. According to some embodiments, the common node includes a reference voltage selected from the group consisting of a battery voltage, a compliance voltage, a portion of the compliance voltage, and ground. According to some embodiments, each of the plurality of switch circuits includes a plurality of switches, and the switches are selectable to change the resistance. According to some embodiments, the plurality of switches includes a plurality of transistors in parallel. According to some embodiments, the stimulator device further includes a control circuit configured with at least one algorithm, wherein the at least one algorithm is configured to determine the time at which the nerve response will occur at the sensing electrode node and to time the passive charge recovery duration such that the high impedance passive charge recovery portion will exactly overlap the time at which the nerve response will occur at the sensing electrode node. According to some embodiments, the stimulator device further includes a control circuit configured with at least one algorithm, wherein the at least one algorithm is configured to determine when the nerve response will occur at the sensing electrode node.

[0018] The present disclosure also discloses a method for operating a stimulator device, the stimulator device including a plurality of electrode nodes, each electrode node being configured to be coupled to one of a plurality of electrodes configured to contact patient tissue, the method including: providing actively driven stimulation at at least one stimulation node selected from the plurality of electrode nodes, wherein the stimulation includes at least one pulse, the at least one pulse including at least a first phase; providing passively driven passive charge recovery for a passive charge recovery duration, selecting at least one recovery impedance for the passive charge recovery duration, providing the selected recovery impedance using a resistive circuit of the stimulator device during the passive charge recovery duration, and sensing a neural response at at least one sensing electrode node selected from the plurality of electrode nodes during the passive charge recovery duration. According to some embodiments, the resistive circuit includes a variable resistive circuit. According to some embodiments, selecting the at least one recovery impedance includes selecting a first recovery impedance for a high impedance portion of the passive charge recovery duration and selecting a second recovery impedance for a low impedance portion of the passive charge recovery duration, wherein the first recovery impedance is greater than the second recovery impedance, and wherein sensing the neural response includes sensing the neural response during the high impedance portion of the passive charge recovery duration. According to some embodiments, the method further includes determining a time at which the neural response will occur at the sensing electrode node and timing the passive charge recovery duration such that the high impedance passive charge recovery portion will completely overlap with the time at which the neural response will occur at the sensing electrode node. According to some embodiments, providing actively driven stimulation includes providing at least one first one or more pulses and at least one second one or more pulses. According to some embodiments, providing passively driven passive charge recovery includes providing a first passive charge recovery duration after each of the first one or more pulses and providing a second passive charge recovery duration after each of the second one or more pulses. According to some embodiments, selecting at least one recovery impedance includes selecting a first recovery impedance for the first passive charge recovery duration and selecting a second recovery impedance for the second passive charge recovery duration, wherein the second recovery impedance is greater than the first recovery impedance. According to some embodiments, sensing the neural response includes sensing the neural response during the second passive charge recovery duration.

[0019] The present disclosure also discloses a stimulator device, comprising: a plurality of electrode nodes, each electrode node being configured to be coupled to one of a plurality of electrodes configured to contact patient tissue; a stimulation circuit configured to provide actively driven stimulation at at least one stimulation node selected from the plurality of electrode nodes, wherein the stimulation comprises at least one pulse, and the at least one pulse comprises at least a first phase; and a passive charge recovery circuit configured to provide passively driven passive charge recovery during a passive charge recovery duration; and a sensing circuit configured to sense a neural response at at least one sensing node selected from the plurality of electrode nodes during the passive charge recovery duration. According to some embodiments, the neural response is generated in response to the actively driven stimulation. According to some embodiments, the passive charge recovery is configured to recover the charge stored during the actively driven stimulation. According to some embodiments, the passive charge recovery circuit comprises a plurality of switching circuits, wherein each of the plurality of switching circuits is coupled to a different one of the electrode nodes and is configured to provide a variable resistance between its corresponding electrode node and a common node when selected. According to some embodiments, the common node comprises a reference voltage selected from the group consisting of a battery voltage, a compliance voltage, a portion of the compliance voltage, and ground. According to some embodiments, each of the plurality of switching circuits comprises a plurality of switches, wherein the switches are selectable to change the resistance. According to some embodiments, the plurality of switches comprises a plurality of transistors in parallel. According to some embodiments, the passive charge recovery duration comprises a high-impedance passive charge recovery duration and a low-impedance passive charge recovery duration, wherein: during the high-impedance passive charge recovery duration, the variable resistor is configured to provide a first impedance to the passive charge recovery, and during the low-impedance passive charge recovery duration, the variable resistor is configured to provide a second impedance lower than the first impedance to the passive charge recovery. According to some embodiments, the sensing circuit is configured to sense the neural response during the high-impedance passive charge recovery duration. According to some embodiments, the stimulator device further comprises a control circuit configured with at least one algorithm, wherein the at least one algorithm is configured to determine the time at which the neural response will occur at the sensing electrode node and to time the passive charge recovery duration such that the high-impedance passive charge recovery duration will exactly overlap with the time at which the neural response will occur at the sensing electrode node. According to some embodiments, the stimulator device further comprises a control circuit configured with at least one algorithm, wherein the at least one algorithm is configured to determine when the neural response will occur at the sensing electrode node. According to some embodiments, the stimulation circuit is further configured to provide actively driven active charge recovery.According to some embodiments, the sensing circuit includes a differential amplifier, and wherein the differential amplifier receives the sensing electrode node at a first input, and wherein the differential amplifier receives a reference electrode node selected from one of the electrode nodes at a second input. According to some embodiments, each electrode node is coupled to its associated electrode through a DC-blocking capacitor. According to some embodiments, the stimulator device includes an implantable pulse generator or an external trial stimulator.

[0020] The present disclosure also provides a method for operating a stimulator device including a plurality of electrode nodes, each configured to couple to one of a plurality of electrodes configured to contact patient tissue. The method includes providing actively driven stimulation at at least one stimulation node selected from the plurality of electrode nodes, where the stimulation includes at least one pulse including at least a first phase; providing passively driven passive charge recovery for a passive charge recovery duration and sensing a nerve response at at least one sense electrode node selected from the plurality of electrode nodes during the passive charge recovery duration. According to some embodiments, the nerve response is present at the at least one sense electrode node during a nerve response duration and the entire nerve response duration during which the nerve response is sensed occurs during the passive charge recovery. According to some embodiments, the method further includes using a control circuit in the stimulator device to determine when the nerve response will occur at the sense electrode node during the nerve response duration. According to some embodiments, the passive charge recovery duration is timed such that the passive charge recovery duration will fully overlap with the nerve response duration. According to some embodiments, providing passively driven passive charge recovery during the passive charge recovery duration includes providing high impedance passive charge for a high impedance passive charge recovery duration and providing low impedance passive charge recovery for a low impedance passive charge recovery duration. According to some embodiments, the nerve response is sensed during the high impedance passive charge recovery duration. According to some embodiments, the stimulator device includes a passive charge recovery circuit configured to provide passively driven passive charge recovery, where the passive charge recovery circuit includes a plurality of switch circuits, each of the plurality of switch circuits coupled to a different one of the electrode nodes and configured to provide a variable resistance between its corresponding electrode node and a common node when selected. According to some embodiments, during the high impedance passive charge recovery duration, the switch circuit is configured to provide a first impedance to the passive charge recovery, and during the low impedance passive charge recovery duration, the switch circuit is configured to provide a second impedance lower than the first impedance to the passive charge recovery. According to some embodiments, the method further includes using a control circuit in the stimulator device to determine when the nerve response will occur at the sense electrode node and timing the passive charge recovery duration such that the nerve response is sensed during the high impedance passive charge recovery duration. According to some embodiments, the method further includes using the low impedance passive charge recovery to recover stored charge not recovered by the high impedance passive charge recovery. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 An implantable pulse generator (IPG) according to the prior art is shown.

[0022] Figure 2A and Figure 2B shows an example of a stimulation pulse that can be generated by an IPG according to the prior art.

[0023] Figure 3 shows a stimulation circuit that can be used for an IPG according to the prior art.

[0024] Figure 4 shows an improved IPG that has the ability to sense neural responses and adjust stimulation depending on such sensing.

[0025] Figure 5A and Figure 5B shows a lead for generating stimulation and shows differential sensing of a neural response caused by the stimulation.

[0026] Figure 6A shows the neural response that is ideally sensed at the sensing electrode, while Figure 6B and Figure 6C shows how a stimulation artifact and passive charge recovery can interfere with sensing the neural response.

[0027] Figure 7 shows a sense amplifier circuit that can be used to sense neural responses.

[0028] Figure 8 shows stimulation and passive charge recovery, including sensing neural responses during passive charge recovery.

[0029] Figure 9 shows a timing algorithm that is operable to determine when a neural response starts and stops at the sensing electrode.

[0030] Figure 10 shows an adjustment algorithm that can be used in conjunction with the timing algorithm to ensure that the high-impedance passive charge recovery duration will overlap with the neural response at the sensing electrode.

[0031] Figure 11 shows aspects of a passive charge recovery circuit for providing an adjustable impedance during passive charge recovery.

[0032] Figure 12 shows the operation of the timing and adjustment algorithms in an external device that communicates with the IPG.

[0033] Figure 13A - Figure 13C shows different examples where a neural response can be sensed during the high-impedance passive charge recovery duration. Detailed Description

[0034] In a pulse generator system, and particularly in a spinal cord stimulator (SCS) pulse generator system, an increasingly interesting development is to increase the sensing capabilities to complement the stimulation provided by such a system. For example, and as explained in U.S. Patent Application Publication 2017 / 0296823, it may be beneficial to sense the neural response in nerve tissue that receives stimulation from an SCS pulse generator. One such neural response is the evoked compound action potential (ECAP). An ECAP comprises the cumulative response provided by nerve fibers recruited by the stimulation and essentially comprises the sum of the action potentials of the recruited fibers when they "fire". The ECAP is Figure 4 shown in, and comprises a plurality of peaks, which are conventionally labeled with P for positive peaks and N for negative peaks, where P1 comprises the first positive peak, N1 comprises the first negative peak, P2 comprises the second positive peak, and so on. Note that not all ECAPs will have the exact shape and number of peaks as Figure 4 shown in, as the shape of the ECAP is a function of the number and type of nerve fibers that are recruited and participate in its conduction. The ECAP is typically a small signal and may have an inter-peak amplitude on the order of tens of microvolts to tens of millivolts.

[0035] Figure 4 Also shown in is a circuit for the IPG 100 that is capable of providing stimulation and sensing the resulting ECAP or other neural response or signal. The IPG 100 includes a control circuit 102, which may include a microcontroller, such as a component number MSP430 manufactured by Texas Instruments, which is described in the data sheet at http: / / www.ti.com / lsds / ti / microcontroller / 16-bit_msp430 / overview.page?DCMP=MCU_other&HQS=msp430, which is incorporated herein by reference. Other types of controller circuits may also be used in place of the microcontroller, such as a microprocessor, FPGA, DSP, or a combination of these, etc. The control circuit 102 may also be formed, in whole or in part, in one or more application specific integrated circuits (ASICs), such as those described previously. The disclosed circuits and techniques may also be implemented in an ETS implantable stimulator, although this is not further discussed.

[0036] The IPG 100 also includes a stimulation circuit 28 to generate stimulation at the electrodes 16, which may include the stimulation circuit 28 shown previously ( Figure 3 ). A bus 118 supplies one or more PDACs 40 i or NDACs 42 iProvide a digital control signal from the control circuit 102 (and possibly from the ECAP algorithm 124, as described below) to generate a current or voltage of a specified amplitude (A) for the stimulation pulse and having the correct timing (PW, f). As previously mentioned, the DAC can be powered between the compliance voltage VH and ground. Also as previously mentioned but not shown in Figure 4 The switch matrix can intervene between the PDAC and the electrode node 39 and between the NDAC and the electrode node to route their outputs to one or more of the electrodes (including the conductive housing electrode 12 (Ec)). The control signal (if any) of the switch matrix can also be carried by the bus 118. Note that the current path to the electrode 16 includes the DC-blocking capacitor 38 described above, which provides safety by preventing the unintentional supply of DC current to the electrode and to the patient tissue. In Figure 4 In the embodiment shown in, the passive charge recovery switch introduced above is replaced by the passive charge recovery switch circuit 41 i As described above, the passive charge recovery switch circuit is used to passively recover any charge remaining on the DC-blocking capacitor by coupling the electrode node 39 to the common reference voltage V CM The passive charge recovery switch circuit is also used to provide a variable resistance path between the electrode node 39 and V CM For ease of discussion, the passive charge recovery circuit is shown as including the variable resistor 188. The passive charge recovery switch circuit 41 i will be described in more detail below, for example, with reference to Figure 11 . The capacitor C R can be set between V CM and ground (GND) to reduce the recovery impedance.

[0037] The IPG 100 also includes a sensing circuit 115, and one or more of the electrodes 16 can be used to sense a nerve response, such as the ECAP described above. In this regard, each electrode node 39 can also be coupled to the sensing amplifier circuit 110. Under the control of the bus 114, the multiplexer 108 can select one or more electrodes to operate as sensing electrodes by coupling one or more electrodes to the sensing amplifier circuit 110 at a given time, as further explained below. Although Figure 4Only one multiplexer 108 and sense amplifier circuit 110 are shown, but there may be more than one. For example, there may be four multiplexer 108 / sense amplifier circuit 110 pairs, each pair operable within one of four timing channels supported by the IPG 100 to provide stimulation. The analog waveform including the ECAP is preferably converted to a digital signal by one or more analog-to-digital converters (one or more ADCs) 112, which may sample the waveform at, for example, 50 kHz. One or more ADCs 112 may also reside within the control circuit 102, particularly if the control circuit 102 has an A / D input. The multiplexer 108 may also provide a DC reference voltage Vamp (e.g., GND) to the sense amplifier circuit 110, as this is useful in single-ended sensing mode.

[0038] To not bypass the safety provided by the DC-blocking capacitor 38, the input to the sense amplifier circuit 110 is preferably taken from the electrode node 39, and thus the DC-blocking capacitor 38 intervenes between the electrode 16 at which the ECAP is sensed and the electrode node 39. However, since the DC-blocking capacitor 38 will pass the AC signal while blocking the DC component, the AC ECAP signal will pass through the capacitor 38 and still be readily sensed by the sense amplifier circuit 110. In other examples, the ECAP may be sensed directly at the electrode 16 without passing through the intervening capacitor 38.

[0039] As shown, the ECAP algorithm 124 is programmed into the control circuit 102 to receive and analyze the digitized ECAP. Those skilled in the art will understand that the ECAP algorithm 124 may include instructions that can be stored on a non-transitory machine-readable medium, such as magnetic, optical, or solid-state memory within the IPG 100 (e.g., stored in association with the control circuit 102).

[0040] In Figure 4 the example shown, the ECAP algorithm 124 operates within the IPG 100 to determine one or more ECAP characteristics, which may include but are not limited to:

[0041] · The height of any peaks present in the ECAP (e.g., H_N1);

[0042] · The peak-to-peak height between any two peaks (such as H_PtoP from N1 to P2);

[0043] · The peak height ratio (e.g., H_N1 / H_P2);

[0044] · The peak width of any peak (e.g., the full width at half maximum of N1, FWHM_N1);

[0045] · The area under any peak (e.g., A_N1);

[0046] · The total area (A_tot), including the positive peak area minus or plus the negative peak area;

[0047] · The length of any part of the ECAP curve (e.g., the curve length from P1 to N2, L_P1toN2)

[0048] · Any time defining the duration of at least a part in the ECAP (e.g., the time from P1 to N2, t_P1toN2);

[0049] · The time delay from the stimulus to the emitted ECAP, which indicates the nerve conduction velocity of the ECAP and may vary in different types of nerve tissues;

[0050] · Any mathematical combination or function of these variables (e.g., H_N1 / FWHM_N1 typically specifies the quality factor of peak N1).

[0051] Once the ECAP algorithm 124 determines one or more of these features, it can adjust the stimulus provided by the IPG 100, for example, by providing new data to the stimulation circuit 28 via the bus 118. This is further explained in U.S. Patent Application Publication 2017 / 0296823 and U.S. Patent Application Serial No. 16 / 135,961 filed on September 19, 2018, the entire contents of which are incorporated herein by reference. In a simple example, the ECAP algorithm 124 can check the height of the ECAP (e.g., its inter-peak voltage) and adjust the amplitude of the stimulation current in a closed-loop manner to attempt to hold the ECAP at an expected value. The ECAP algorithm 124 can also include sub-algorithms, such as the timing algorithm 150 and the adjustment algorithm 170, which will be further described below.

[0052] An embodiment of the microcontroller can include a recovery logic / control block 402 that implements logic that emits a plurality of control signals for controlling passive charge recovery, including a control signal for controlling the resistor at which passive charge recovery occurs. Details of the recovery control and recovery logic for controlling the resistor at which passive charge recovery occurs are described in U.S. Patent Application Publication No. 2018 / 0071527, the entire contents of which are incorporated herein by reference. The recovery logic / control block 402 can receive data for adjusting aspects of passive charge recovery from the timing algorithm 150 and / or the adjustment algorithm 170, as described in more detail below.

[0053] Figure 5A and Figure 5B shows the percutaneous lead 15 (paddle leads 19 or other leads can also be used), and shows Figure 2AAn example of a stimulation program, where E4 and E5 are used to generate biphasic pulses in a bipolar stimulation mode, where (during the first phase 30a) E4 includes the anode and E5 includes the cathode, although other electrode arrangements (e.g., tripolar, etc.) can also be used. This stimulation generates an electromagnetic (EM) field 130 in the patient tissue volume around the selected electrodes. Some of the nerve fibers within the EM field 130 will be recruited and fired, particularly those close to the cathode electrode E5. It is hoped that the sum of the nerve fiber firings will mask the signals indicating pain in SCS applications, thereby providing the desired treatment. The total number of recruited nerve fibers generates an ECAP, which can travel rostrally towards the brain and caudally away from the brain. The ECAP is conducted through the spinal cord by nerve conduction, and its speed depends on the nerve fibers involved in the conduction. In one example, the ECAP can move at a speed of approximately 5 cm / 1 ms.

[0054] The ECAP is preferably sensed differentially using two electrodes, and Figure 5A and Figure 5B different examples are shown. In Figure 5A , a single electrode E8 on the lead 15 is used for sensing (S+), and another signal is used as a reference (S-). In this example, the sensing reference S- includes an electrode further away in the electrode array 17 or the case electrode Ec (as shown). The reference S- may also include a fixed voltage provided by the IPG 100, such as ground or Vamp ( Figure 4 ), in which case the sensing can be said to be single-ended rather than differential. In Figure 5B , two lead-based electrodes are used for sensing, where these electrodes are adjacent to each other or at least relatively close to each other. Specifically, in this example, the electrode E8 is again used for sensing (S+), and the adjacent electrode E9 provides the reference (S-). This can also be reversed, where E8 provides the reference (S-) for sensing (S+) at the electrode E9. Sensing a given ECAP at different electrodes can allow the ECAP algorithm 124 to understand the time difference between when the ECAP arrives at each of the electrodes. If the distance x between the electrodes is known, the ECAP algorithm 124 can calculate the speed of the ECAP. As mentioned above, the ECAP speed indicates the nerve fibers involved in nerve recruitment and conduction, which are of interest in their own right and may be useful to the ECAP algorithm 124 when adjusting the stimulation provided by the stimulation circuit 28.

[0055] Figure 6AShows the ECAP that is ideally sensed at the sense electrode S+ (e.g., E8). In this example, it is assumed that the distance between the sense electrode E8 and the stimulation electrode (e.g., E5) is d = 12 mm, and it is assumed that the electrodes in the array are spaced apart at a distance of x = 4 mm. If it is assumed that the ECAP travels at a speed of 50 mm / ms (again, this may vary depending on the neural tissue involved), then the ECAP will start passing through the sense electrode S+ at time t1 = 0.24 ms. The ECAP itself also has a time spread (t ECAP ). This duration is also variable, but in Figure 6A , it is assumed that the ECAP exists at the sense electrode S+ for one millisecond as a reasonable nominal value (i.e., t ECAP = 1 ms). Thus, in this example, the ECAP will finish passing through the sense electrode S+ at time t2 = t1 + t ECAP (e.g., t2 = 1.24 ms). The time t ECAP that the ECAP takes to pass through the sense electrode is referred to herein as the neural response duration.

[0056] Figure 6B Shows the waveform of the stimulation program and the signals that will appear in the tissue at the sense electrode E8 (S+). In addition to including the ECAP signal to be sensed (between times t1 and t2), the signal at the sense electrode S+ also includes a stimulation artifact 134. The stimulation artifact 134 includes the voltage formed in the tissue due to the stimulation, i.e., due to the EM field 130 generated at the stimulation electrodes E4 and E5. As described in U.S. Provisional Patent Application Serial No. 62 / 650,844, filed March 30, 2018, the entire content of which is incorporated herein by reference, the PDAC and NDAC for forming current in the tissue have a high output impedance. This can cause the voltage in the tissue to vary between ground and the compliance voltage VH used to power the DAC, which as mentioned earlier can be a high voltage (on the order of volts). Thus, the magnitude of the stimulation artifact 134 at a given sense electrode S+ or its reference S- can be high (e.g., from millivolts to volts) and significantly higher than the magnitude of the ECAP. The magnitude of the stimulation artifact 134 at the sense electrodes S+ and S- depends on many factors. For example, if the sense electrode is closer to the stimulation electrode (E4, E5), then the stimulation artifact 134 will be larger. The stimulation artifact 134 is typically also larger during the delivery of the pulse (during phases 30a and 30b), and due to the capacitive nature of the tissue, the stimulation artifact 134 may still be present even after the pulse (i.e., the last phase 30b of the pulse) has stopped, which prevents the electric field 130 from dissipating immediately. As shown, when the current reverses polarity, the polarity of the stimulation artifact 134 changes between phases 30a and 30b of the stimulation pulse. Although the sense artifact 134 and the ECAP are shown simply in Figure 6BShown separately in the middle, but in reality they will be superimposed (added) at the sensing electrode S+. Note that the sizes of the sensing artifact 134 and the ECAP are not necessarily drawn to scale; in particular, the sensing artifact 134 may be much larger.

[0057] The background stimulation artifact 134 of a relatively large signal makes it difficult to resolve and sense the small signal ECAP at the sensing amplifier circuit 110. To ameliorate this concern, it may be beneficial to use a sensing electrode S+ that is far from the stimulation electrode. See, for example, U.S. Provisional Patent Application Serial No. 62 / 768,617, filed Nov. 16, 2018, the entire contents of which are incorporated herein by reference. This may be beneficial because the stimulation artifact 134 will be smaller at a more-distant sensing electrode and because the ECAP will pass through the distant sensing electrode at a later time when the stimulation artifact 134 may have dissipated. However, using a distant sensing electrode is not always feasible or practical. On the one hand, the electrode array 17 simply may not be large enough and thus no electrode can be appropriately far enough from the stimulation electrode to operate ideally as a sensing electrode. Similarly, the magnitude of the ECAP also diminishes as the distance from the stimulation electrode increases and thus while the stimulation artifact 134 will be smaller at a more-distant sensing electrode, so will the ECAP, again making sensing difficult.

[0058] Then assume E8 is still Figure 6BThe sensing electrode in. In this example, it is assumed that the pulse phases 30a and 30b have relatively long pulse widths, where PWa of the first phase 30a and PWb of the second phase 30b are both equal to 0.25 ms. In total, the pulse is actively driven by the DAC circuit (40 / 42) from 0 to 0.5 ms, and thus the stimulus artifact 134 dominates during this time period. (This time period may include an interphase period of short duration between phases 30a and 30b, although this is not shown for simplicity). Unfortunately, this stimulus artifact 134 overlaps with the ECAP at the sensing electrode S+ in time, which occurs again between 0.24 (t1) and 1.24 ms (t2). This makes it difficult to sense the ECAP at the sensing electrode S+. First, the stimulus artifact 134 may be significantly larger than the small-signal ECAP. In addition, the stimulus artifact 134 changes significantly during the time when the ECAP is present at the sensing electrode S+. In particular, at 0.25 ms, the stimulus artifact 134 changes polarity (from phase 30a to 30b), swinging from a negative value to a positive value. In addition, the stimulus artifact 134 drops from a positive value to 0 at 0.5 ms (at the end of phase 30b), which in this example occurs in the middle of the ECAP. Since the ECAP is superimposed on the stimulus artifact 134, this makes it difficult to resolve the ECAP at the sense amplifier circuit 110.

[0059] Figure 6C This sensing problem is alleviated to some extent by making the pulse width smaller. In this example, PWa and PWb have each been reduced to 0.12 ms. Like this, when the ECAP first starts to appear at the sensing electrode S+ (at t1), the stimulus artifact 134 is mainly substantially over at 0.24 ms (at the end of phase 30b). Like this, there is no significant overlap between the ECAP and the stimulus artifact 134 at the sensing electrode S+, and this is even more so if the pulse width is further reduced. However, this solution may not be ideal. First, adjusting the pulse width may not be simply feasible because they may not be sufficient for providing adequate stimulation treatment to the patient. In addition, if the ECAP travels relatively fast, it may not be possible to simply reduce the pulse width to avoid overlap with the ECAP.

[0060] In addition, although in Figure 6C the ECAP may no longer significantly overlap with the stimulus artifact 134, the ECAP still overlaps during the time period 30c, during which it may be desirable to provide passive charge recovery after the active pulse phases 30a and 30b are completed. As mentioned previously, passive charge recovery involves turning off the passive charge recovery switch circuit 41i ( Figure 4 ), which connects the electrode node ei 39 to a reference potential (such as V CM)。Even if the switching circuit 418 at the sense electrode E8 is not closed, the effect of closing some switches will cause current to flow passively in the tissue, which also results in variable voltage artifacts in the tissue (not shown). See, for example, U.S. Patent Application Publication 2018 / 0140831. In short, passive charge recovery makes it difficult to sense ECAP because it - like the stimulation artifact 134 - can produce a voltage that varies over time in the tissue that is significantly greater than the ECAP. Note that Figure 6B the provision of the passive recovery 30c in

[0061] is also problematic because there the passive charge recovery period 30c overlaps with the ECAP to some extent in time. As mentioned previously, it is preferable to use the electrodes S+ and S- to differentially sense the ECAP, both of which are exposed to the tissue, thus allowing the artifacts in the tissue (i.e., the stimulation artifact 134 or the artifact associated with passive charge recovery) to be subtracted from the ECAP measurement at least to some extent. Figure 7Shown therein is a sense amplifier circuit 110 that provides differential sensing. The sense amplifier circuit 110 includes a differential amplifier 111. Also shown is an example of the circuitry within the differential amplifier 111, but it should be noted that there are many different differential amplifier circuits and they can also be used in the sense amplifier circuit 110. The sense electrode S+ and the sense reference electrode S- are coupled through a DC-blocking capacitor 38 (if used) to derive signals X+ and X- at the electrode node 39, and these signals are presented to the positive input terminal and the negative input terminal of the differential amplifier 111. The signals X+ and X- will be substantially the same as S+ and S- present at the selected sense electrode, but with the DC signal component removed. X+ and X- are provided to the gates (control terminals) of transistors M+ and M- in the differential amplifier 111. The drains of transistors M+ and M- are connected to the differential output terminals D+ and D-, which in turn are coupled to the supply voltage Vdd of the amplifier via resistors R+ and R-. The sources of transistors M+ and M- are connected to ground as another supply voltage through a common bias transistor Mb, which sets the total current Ib, which can be totaled through each of the branches (I+, I-) of the differential amplifier. Resistors R+ and R- are equal and are represented as simple resistors, although active devices (such as PMOS transistors) can also be used. The output Vo of the amplifier 111 is equal to the voltage difference at the output terminals D+ and D-, which in turn is affected by the signal difference present at X+ and X-. If the signals X+ and X- are different (e.g., if an ECAP is present at S+), transistors M+ and M- will be turned on to different extents, resulting in different currents I+ and I- flowing through each branch. This will create different voltage drops across resistors R+ and R-, and thus different voltages at D+ and D-. In short, Vo = D+ - D- = A1(X+ - X-), where A1 is the gain of the amplifier 111.

[0062] The differential amplifier 111 can provide its output to various processing circuits 147 before being presented to the control circuit 102 and the ECAP algorithm 124. For example, the differential output Vo of the differential amplifier 111 can be provided to the input of another differential amplifier 146 and to yet another series of differential amplifiers, etc. This can help increase the gain of the detected ECAP signal, as the gain of each amplifier stage will multiply (A1 * A2, etc.). A follower circuit or buffer can also be used in series as part of the processing circuit 147 between the differential amplifier 111 and the ADC 112, but such a stage is not shown. Additionally, the processing circuit 147 can include a low-pass filter (LPF) 148 to remove high-frequency components in the ECAP signal that are not of interest or are inconsistent with the rate at which the ADC 112 will sample the signal. In one example, the LFP 148 removes frequency components of 25 kHz or higher. The processing circuit 147 can include a part of the control circuit 102.

[0063] To prevent damage or improper operation of the differential amplifier 111 (i.e., the first differential amplifier in series), the inputs X+ and X- can be respectively provided with clamping circuits 142+ and 142-. In the illustrated example, the clamping circuit 142+ includes diodes 144a and 144b connected in series, which are forward-biased between a low clamping reference voltage (Vcl) and a high clamping reference voltage (Vch), and the signal X+ is connected to the node between the diodes. Vcl and Vch preferably include ground and the supply voltage Vdd (e.g., 3.3V). In this example, it is assumed that the diodes 144a and 114b have a forward bias threshold voltage (Vtd) of 0.6V. If the voltage at X+ is less than -0.6 volts, the diode 144a will conduct (turn on). Since this conductance has a very low resistance, X+ is effectively clamped to a minimum value of Vmin = -0.6 volts. If it is assumed that Vdd = 3.3V, then if X+ is greater than 3.9V volts, the diode 144b will conduct, which will clamp X+ to a maximum value of Vmax = 3.9V. If the voltage at X+ is at or between -0.6 and 3.9 volts, neither of the diodes 144a and 144b in the clamping circuit 142+ will conduct. The clamping circuit 142- is similar but is connected to the signal X-, and thus similarly clamps X- to a voltage at or between -0.6 and 3.9 volts. The clamping circuits 142+ and 142- can be modified to adjust the window of allowable voltages at which clamping does not occur. For example, Vcl and Vch may be generated by their own generator circuits to produce unique values different from ground and Vdd; different numbers of diodes may be used; Zener diodes can be used for breakdown and thus clamp X+ or X- at a specified reverse bias voltage; and so on.

[0064] Figure 7 Blanking switches 141+ and 141- are also shown, which are respectively used to transfer the signals at X+ and X- to the differential amplifier. The blanking switches 141+ and 141- can be used to protect the differential amplifier 111, and in particular to protect the amplifier 111 from receiving excessive voltages at the signals X+ and X-. (Note, however, that the clamping circuits 142+ and 142- that limit the voltages at X+ and X- can somewhat alleviate the need for the blanking switches 141+ and 141-). As further described below, the blanking switches can be used in combination with the disclosed technology. Note that the blanking switches 141+ and 141- can include logic switches for routing the electrode node 39 to the sense amplifier circuit 110. For example, the blanking switches 141+ and 141- can include switches within the multiplexer 108( Figure 4 ), or they can include independent switches.

[0065] The sense amplifier circuit 110 may also include DC level shifting circuits 143+ and 143- to set the signals X+ and X- to a DC voltage reference consistent with the input requirements of the differential amplifier 111. The differential amplifier 111 can operate reliably only if the magnitudes of the signals X+ and X- cause currents I+ and I- to flow in each branch of the amplifier. In this regard, to sense a small signal ECAP, X+ and X- should be higher than the threshold voltages of the amplifier input transistors M+ and M- (e.g., greater than Vtt = 0.7V). Further preferably, X+ and X- do not exceed the supply voltage Vdd of the differential amplifier (e.g., Vdd = 3.3V) for proper amplifier operation. Thus, the signals provided to the differential amplifier 111 are preferably referenced to a DC voltage reference within this operating range. The reference can include 1 / 2Vdd (e.g., 1.65V), which is the midpoint between Vdd and ground. More preferably, the DC voltage reference can include 1 / 2(Vdd - Vtt)+Vtt (e.g., 2.0V), because this value will be the midpoint within the operating range of 0.7V and 3.3V, and thus allows X+ and X- to swing symmetrically from the reference by + / -1.3V, while still providing an input magnitude suitable for operating the differential amplifier 111. Although such a circuit can take different forms, in the example shown, the DC level shifting circuits 143+ and 143- include a resistor ladder, including resistors Ra and Rb serially biased between Vdd and ground, where the signals X+ and X- are connected to the node between the resistors. This sets the DC voltage reference of X+ and X- to Ra / (Ra + Rb)*(Vdd - ground). Thus, by appropriately setting the values of Ra and Rb, the DC voltage reference can be set to any desired value between Vdd and ground, such as 2.0V. The AC signals then coupled to X+ and X- through a capacitor 38 (such as an ECAP and / or a stimulus artifact 134) then reference (and ride on top of) this DC voltage reference. Generally, this allows the differential amplifier 111 to be affected by the ECAP at X+, because the superposition of the ECAP and the DC voltage reference will cause a change in the current I+. Preferably, Ra and Rb are large resistors, such as 1 megohm or higher.

[0066] Because the stimulus artifact 134 is present at both the sense electrode S+ and the reference electrode S-, the differential amplifier 111 will ideally subtract the artifact in the tissue (i.e., the stimulus artifact 134 and the artifact associated with passive charge recovery) from the output (Vo) as a common-mode voltage, leaving only the ECAP to be sensed. However, it is not surprising that the magnitude of such artifacts may not be exactly the same at the sense electrodes S+ and S-, because each must be located at a different distance from the stimulating electrode. Therefore, the common-mode rejection of this artifact by the differential amplifier 111 may not be perfect. In addition, when the artifact is relatively large and time-varying, it is difficult to design the differential amplifier 111 to resolve the ECAP. This is a particular problem in the scenarios discussed earlier Figure 6B and Figure 6C where the ECAP significantly overlaps in time with the stimulus artifact 134 and the passive recovery artifact at the sense electrode S+.

[0067] As described previously, conventional wisdom teaches that it is not desirable to sense ECAP during the delivery of pulses to tissue. Again, this is because the stimulus artifact - caused by either active stimulation or passive charge recovery that may follow - may be large or vary during these periods. However, contrary to this conventional wisdom, the inventors have designed a new ECAP sensing strategy that is capable of sensing ECAP during the active and / or passive charge recovery phases. The ECAP sensing strategy for sensing ECAP during the delivery of active stimulation / active charge recovery is described in U.S. Provisional Patent Application Serial No. 62 / 825,982, filed on March 29, 2019, the entire content of which is incorporated herein by reference. This disclosure describes ECAP sensing during passive charge recovery.

[0068] Figure 8 Shows the waveform of a single-phase pulse 802 delivered at electrode E4. Passive charge recovery is used to recover the charge after the pulse. Note that in Figure 8 while only the waveform of a single stimulating electrode E4 is shown, it will be understood that another electrode, such as the case electrode or one of the other lead electrodes, is used as a counter electrode. Also note that while Figure 8 a simple single-phase pulse is shown for clarity in

[0069] the passive charge recovery can also be used in combination with biphasic pulses and other more complex waveforms. The ECAP strategy described herein can be used with any stimulation waveform that includes passive charge recovery. R,i The single-phase waveform 802 includes a single-phase stimulation pulse with an amplitude of +Aa, and is followed by a passive charge recovery phase. The passive charge recovery phase has an initial amplitude of -A CThe voltage accumulation divided by the tissue impedance R t ; that is, -A R,i ≈V C / R t .

[0070] The amplitude of the passive charge recovery phase decays exponentially over the time period t R according to the time constant τ, which is equal to the capacitance of the capacitor multiplied by the tissue impedance R t . The time period tR for substantially all the charge to be recovered is approximately 5τ. As further explained below, the shape of the decay curve of the passive charge recovery can be controlled / altered by including a variable or selectable resistor in series with the passive charge recovery switch 41 (see, for example Figure 3 and Figure 4 ).

[0071] Still referring to Figure 8 , signals 804 and 806 that appear in the tissue at the sensing electrodes (E8 and E9) are also shown. Signals 804 and 806 include a stimulus artifact, which includes a voltage formed in the tissue due to the stimulus, i.e., due to the EM field generated at the stimulus electrode E4. Signals 804 and 806 also include a recovery artifact, which includes a voltage formed in the tissue due to passive charge recovery. Similar to the passive charge recovery of the stimulus waveform 802, the recovery artifacts of signals 804 and 806 are each represented as an exponential decay curve. The time constant of each of the recovery artifact signals is a function of the time constant of the passive charge recovery portion of the stimulus waveform 802 from which the artifact is generated. In other words, changes in the passive charge recovery of the stimulus waveform 802 will be reflected in the artifact signals 804 and 806.

[0072] From Figure 8 it can be seen that if the ECAP signal to be sensed at the sensing electrodes overlaps with the recovery artifacts at those electrodes, then for the same reasons described above regarding sensing an ECAP signal that overlaps with the actively driven charge recovery, sensing of the ECAP signal will be difficult. The inventors have found that an ECAP signal that overlaps with the recovery artifacts can be better resolved by implementing a high-impedance passive charge recovery. For example, a high recovery impedance can be used during the initial passive recovery phase during the time period when the ECAP signal is expected at the sensing electrodes. The impedance of the passive charge recovery phase can be controlled by including a variable or optional resistor in the passive charge recovery switch circuit, as explained in more detail below.

[0073] Figure 8 The waveform 808 of R,i, whose magnitude is lower than -A (of waveform 802) R,i , because now the initial amplitude is substantially equal to V C / (R t +R R ), where R R is the increased recovery impedance.

[0074] High-impedance passive charge recovery decays according to the time constant τ', which is a function of the tissue resistance R t plus the increased recovery resistance R R (i.e., τ' is the product of the capacitance multiplied by the sum of R t and R R ). The duration of the high-impedance phase is denoted as t HZ , and can be controlled by the system. Note that during high-impedance passive charge recovery, the system is in a near-steady state, i.e., the passive recovery current decays slowly and the decay curve is relatively flat. Also note that during the high-impedance duration, not all charges may be recovered, i.e., the decay curve may not decay completely to the baseline. Therefore, after the high-impedance duration, charge balance can be achieved by using lower-impedance passive charge recovery, thereby achieving more aggressive (i.e., fast) charge recovery. Waveform 808 includes an intermediate-impedance duration implemented for the time period t IZ and a low-impedance duration implemented for the time period t LZ . Note that the intermediate impedance and low impedance recoveries have time constants τ” and τ”' respectively (although not indicated in the figure). The total passive charge recovery duration t R ' is the sum of the durations of the high-impedance phase, the medium-impedance phase, and the low-impedance phase. It should be noted that the number of different impedance phases included during charge recovery may be more or less than three (which are shown only for illustrative purposes). According to some embodiments, the impedance during the high-impedance phase can be on the order of 10 kΩ; the intermediate impedance phase can be 300–1800 Ω, and the low impedance phase can be 50–100 Ω. These values are merely exemplary; other impedances can be used.

[0075] The benefits of using high-impedance passive charge recovery during the time when an ECAP is expected to be present at the sensing electrode are shown in the signals 810 and 812 sensed at the sensing electrode. As shown, the period of passive charge recovery is extended, and the period of the high-impedance duration t HZ completely overlaps with the ECAP (i.e., the neural response duration) at the sensing electrode. This is beneficial for ECAP sensing in several ways. First, although the recovery artifact still exists when the ECAP is present at the sensing electrode, the artifact is smaller because its magnitude has been reduced. This helps the differential amplifier 111( Figure 7) Sensing is performed because the differential amplifier can more easily handle (i.e., subtract) a smaller common-mode voltage that is closer to the magnitude of the ECAP.

[0076] Second, by extending the duration of the passive charge recovery phase, this phase no longer starts or ends during the ECAP at the sense electrode (in the middle). This also simplifies sensing because the recovery artifact is relatively constant (i.e., flat) during the ECAP at the sense electrode.

[0077] In addition, it should be noted that the stimulation treatment of the patient is not significantly changed. The monophasic stimulation pulse (or the first phase of the biphasic pulse) produces a significant therapeutic effect on the patient, and thus the amplitude Aa and the pulse width PWa are typically tailored to the patient. In this example, these pulse parameters Aa and PWa are unchanged. The passive charge recovery phase generally has no therapeutic significance and can thus be changed without having a major impact on the patient.

[0078] Note that Figure 8 the solution may have practical limitations. For example, if the pulses have a high frequency F, there may not be enough time between subsequent pulses to accommodate the extended passive charge recovery. However, this problem can be simply alleviated by not providing subsequent pulses until the ECAP measurement is complete. This should not pose a major problem for the treatment of the patient because ECAP measurements are not typically performed after each pulse, but only occasionally; an occasional delay or missed therapeutic pulse will not significantly affect the stimulation treatment.

[0079] In the disclosed technique, as Figure 8 shown in, blanking should not occur during the passive charge recovery phase when the ECAP is sensed. That is, the switches to the input of the differential amplifier 111, such as 141+ and 141- ( Figure 7 ) should be closed to allow the sensed signals S+ and S- (and X+ and X-) to reach the input of the amplifier. During the stimulation phase, blanking may occur - i.e., the switches 141+ and 141- can be opened. This can help protect the differential amplifier 111 from saturation, which may occur if the stimulation artifact is large during the stimulation phase. That being said, blanking is not strictly required during the stimulation phase, especially if the clamping circuits 142+ and 142- are used to limit the voltages on X+ and X-.

[0080] To implement high-impedance passive charge recovery to assist in sensing neural responses, several timing aspects need to be considered. First, it is generally desirable (although not always necessary) that all the charge stored on the DC-blocking capacitor during the actively driven stimulus be recovered during the stimulus. In other words, if not all the charge is recovered during that period, the charge may continue to accumulate on the capacitor over multiple periods. Additionally, it is generally desirable to sense the neural response during the high-impedance duration and that the high-impedance duration completely overlap with the neural response duration. Thus, according to embodiments of the disclosed neural sensing strategy, charge recovery is provided and timed such that the neural response is sensed during the high-impedance passive charge recovery phase and additional charge recovery (lower-impedance passively driven passive charge recovery and / or actively driven active charge recovery) is provided to recover any additional remaining charge.

[0081] Figure 9 And Figure 10 Alternative algorithms 150 and 170 are disclosed, which can be used to adjust passive charge recovery to assist in ECAP sensing. Figure 9 Disclosed is a timing algorithm 150 for determining when (t1) and when (t2) an ECAP will begin and end (see Figure 6A ) at the sensing electrodes S+ and S- that have been selected for ECAP sensing. In other words, the algorithm determines the neural response duration. The timing algorithm 150 can receive as input, or be programmed with, one or more selected sensing electrodes (e.g., S+ = E8; S- = E9) and one or more stimulation electrodes (e.g., E4) and the distance (x) between the electrodes in the electrode array (e.g., 4 mm). Thereby, the algorithm 150 can calculate the distance (d) between the stimulation electrode and the sensing electrode (e.g., 12 mm). The timing algorithm 150 can also receive an expected ECAP speed (e.g., 5 cm / ms), or be programmed with it. Note that this speed can be an estimated speed or a speed actually measured by the IPG. Based on the distance d and the ECAP speed, the time t1 when the ECAP will first appear at the sensing electrode can be determined (e.g., d / speed = 0.24 ms). The timing algorithm 150 can also receive an expected ECAP duration (t ECAP = 1 ms), or be programmed with it. Again, this value can also be measured in the IPG. This allows calculation of the time t2 when the ECAP will finish appearing at the electrode (e.g., t2 = t1 + t ECAP ). If necessary, t1 and t2 can also be adjusted to provide additional margins - for example, t1 can be slightly decreased and t2 can be slightly increased to ensure that t1 and t2 are suitable for detecting the ECAP (using the following adjustment algorithm 170).

[0082] Although not shown, the timing algorithm 150 can also use measurements alone to determine t1 and t2. For example, a short test pulse with a low pulse width that is unlikely to produce significant artifacts can be used, where the resulting ECAP is measured by the sense amplifier circuit 110. Thus, t1 and t2 can be determined empirically.

[0083] Once t1 and t2 are determined using the timing algorithm 150 - the start and end of the ECAP at one or more sense electrodes S+ / S-, the adjustment algorithm 170 can use these values to determine how to adjust the prescribed pulse and passive charge recovery, as Figure 10 shown. In particular, the adjustment algorithm 170 can adjust the passive charge recovery to ensure that the high impedance duration of the passive charge recovery is long enough to overlap with the ECAP at the sense electrodes S+ / S-. In this regard, the pulse parameters of the received prescribed pulse - which may be determined to provide adequate patient treatment, which in this example includes the parameters for a single-phase pulse having an amplitude Aa and a pulse width PWa. As mentioned above, although a single-phase pulse is used as an example for simplicity, the methods described herein can be used with pulses of any shape.

[0084] As a first step, the adjustment algorithm 170 can optionally evaluate the timing of the pulse phase (PWa) to determine if it is less than t1. As previously discussed with respect to Figure 6B and Figure 6C if PWa is not less than t1, this can be problematic for ECAP sensing because the ECAP will appear at the sense electrode S+ when the stimulus artifact 134 changes (e.g., between phases 30a and 30b or between 30a and passive charge recovery). If PWa is not less than t1, the adjustment algorithm 170 can take certain actions, such as adjusting PWa to be less than t1 (even if this may change the treatment provided by the first phase 30a), or selecting one or more new sense electrodes S+ / S- that may be further from the stimulation electrode. If it is known a priori that PWa < t1 and thus the ECAP should not overlap with the first phase 30a of the pulse, this step may not be necessary. Note that selecting a new sense electrode changes the timing t1 and t2 at which the ECAP will start and end at that new sense electrode, and thus one or more newly selected sense electrodes S+ / S- can be passed back to the timing algorithm 150 ( Figure 9 ), such that t1 and t2 can be re-determined and the adjustment algorithm 170 repeated.

[0085] If PWa < t1, the adjustment algorithm 170 can continue by evaluating whether the durations of both the active pulse phase and the high impedance charge recovery phase are less than t2, i.e., if PWa + t HZ>than t2. If this is true, then the ECAP at one or more sense electrodes should fall entirely within the high impedance charge recovery phase, as desired. In this case, the IPG can simply provide stimulation and sense the ECAP during the high impedance phase, similar to Figure 8 as shown in

[0086] If PWa + t HZ is not greater than t2, this means that the high impedance passive charge recovery phase ends somewhere in the middle of the time when the expected ECAP appears at one or more sense electrodes. In other words, the high impedance duration does not entirely overlap with the nerve response duration. The adjustment algorithm 170 can thus adjust the timing of the high impedance passive charge recovery duration by increasing t HZ such that PWa + t HZ is now greater than t2. At this point, the ECAP at the sense electrodes should fall entirely within the high impedance passive charge recovery duration.

[0087] According to some embodiments, the adjustment algorithm 170 can also check to confirm that the passive charge recovery parameters allow charge recovery to be completed during the pulse period, i.e., before the next stimulation pulse. The adjustment algorithm can adjust the timing and / or resistance used during the intermediate impedance and / or low impedance phases of the passive charge recovery. For example, if the high impedance passive charge recovery is longer, the intermediate and / or low impedance phases (i.e., with lower impedance) can be applied more aggressively to ensure complete charge recovery before the subsequent stimulation pulse is issued. After such an adjustment to the passive charge recovery phase, the IPG can provide stimulation and sense the ECAP during the adjusted high impedance passive charge recovery duration.

[0088] Although algorithms 150 and 170 are described as separate for ease of illustration, they can be combined into a single algorithm.

[0089] As Figure 4 shown, the timing and adjustment algorithms can be part of the ECAP algorithm 124 operable in the control circuit 102 ( Figure 4 ). As mentioned above, the timing and adjustment algorithms can provide data to the recovery logic / control 402, which issues control signals to the passive charge recovery circuit to implement the specified impedance and timing adjustments to the passive charge recovery phase. The impedance of the passive charge recovery phase is adjusted using the variable resistor 188 (i.e., adjustment is made using the variable resistance of the passive charge recovery switch circuit 41, Figure 4 ). Thus, the recovery switch circuit 41 can be considered an example of a resistance circuit for controlling the recovery impedance.

[0090] Figure 11Shows a subset of electrodes including electrodes E1 - E3 and a housing electrode Ec. The passive charge recovery switch circuit 41 i is connected between the electrode node ei 39 and the common reference voltage V CM . As mentioned above, V CM can be VH / 2, Vbat, ground (GND), or some other reference voltage value.

[0091] As mentioned above, the passive charge recovery switch circuit 41 i provides a variable resistance path between the electrode node ei and the common reference voltage V CM (the variable resistance is represented by the variable resistor 188 in Figure 4 ). Each in the switch circuit 41 i receives each of the four resistance control signals RZ[4:1] from the recovery logic / control 402, and thus can assert any of these control signals to set the resistance of the switch circuit 41 i . The recovery impedance of each electrode can be independently controlled such that different impedances can be used on different electrodes at any given time. Thus, according to some embodiments, charge can be quickly and fully recovered on some electrodes while still presenting a high impedance (slower recovery) to other electrodes.

[0092] In the embodiment shown in Figure 11 , the switch circuit 412 includes four resistive transistors 189 connected in parallel between the electrode node e2 and the common reference voltage (e.g., V CM ). Each of the resistance control signals RZ[4:1] is received at the gate of one of the resistive transistors 189. Each of the resistive transistors 189 preferably has a different size to provide different resistances. For example, this size difference can be achieved by constructing each of the resistive transistors 189 with different lengths (e.g., 100x, 300x, 1800x, and 10000x). Similarly, the transistor width can also be sized to provide different resistances. In the example shown, RZ1 controls the 100 - ohm resistive transistor 189; RZ2 controls 300 ohms; RX3 controls 1800 ohms; and RZ4 controls 10000 ohms. It should be understood that these resistance values are merely exemplary and other resistance values can be selected. As mentioned above, other aspects of changing impedance during passive charge recovery are described in the incorporated U.S. Patent Application Publication No. 2018 / 0071527.

[0093] As Figure 4 shown, the timing and adjustment algorithm can include in the control circuit 102 ( Figure 4) as part of the ECAP algorithm 124 operable therein. However, these algorithms can also be operated in whole or in part in an external computer device 158 used to program the IPG (see, e.g., Figure 12 ), such as an external controller for the patient or a clinician programmer. Such an external device 158 typically communicates wirelessly with the IPG 100 and is described in U.S. Patent Application Publication 2019 / 0046800, which is incorporated herein by reference in its entirety. Figure 12 A graphical user interface (GUI) 160 presented on such an external device 158 is shown in

[0094] . The GUI 160 shows user-selectable options 162 for setting stimulation parameters of the pulses or pulse phases of the IPG, such as amplitude (A), pulse width (PW), and frequency (F), as well as whether certain electrodes operate as anodes or cathodes, and the percentage of amplitude (X%) applied to that electrode. In fact, the GUI 160 and 162 may be much more complex than shown. Figure 8 The GUI 160 may include an option 164 to modify a separately prescribed pulse into a pulse more suitable for ECAP sensing, such as by adding a multi-impedance passive charge recovery phase, or modifying an already prescribed passive charge recovery phase such that the high-impedance phase of the passive charge recovery overlaps with the ECAP at the sensing electrode, as shown in Figure 9 - Figure 10 . The selection of the option 164 can use the

[0095] timing and adjustment algorithm or other algorithms to determine the pulse / recovery parameters required to achieve that goal and send these parameters to the IPG. If necessary, the IPG can transmit the ECAP test measurement results back to the algorithm, such as the ECAP start and end times t1 and t2 measured at the sensing electrodes S+ / S-. The algorithms can be stored in a non-transitory machine-readable medium in the external device 158, such as magnetic, optical, or solid-state memory, which can be stored in association with the control circuit 166 of the external device 158, and the control circuit 166 can include one or more microcontrollers, microprocessors, FPGAs, DSPs, etc. In one example, the control circuit 166 can include one of the i5 series microprocessors manufactured by Intel Corporation. Figure 8The examples shown focus on providing a single-phase stimulation pulse, followed by multi-impedance passive charge recovery, which includes three impedance durations stepping from high impedance to low impedance. Generally, for the reasons explained above, it is preferred to sense the ECAP during the high impedance duration of the passive charge recovery. As mentioned above, other more complex waveforms can be implemented according to the sensing / passive charge recovery strategies disclosed herein. Figure 13A - Figure 13C Some examples of such more complex waveforms are shown that utilize a high impedance passive charge recovery phase during ECAP sensing.

[0096] Figure 13A Waveform 1302 with a stimulation pulse is shown, which has a stimulation phase 30a applied at a stimulation electrode (e.g., E4), followed by an active charge recovery phase 30b applied at the same electrode. Note that the active charge recovery phase 30b is not sufficient to recover all the charge stored during the stimulation pulse. Before the ECAP reaches the sensing electrodes S+ / S- (e.g., E8 / E9), the charge recovery switches from active recovery to passive recovery using high impedance passive charge recovery to recover the remaining stored charge, and the neural response is sensed during the high impedance passive charge recovery duration.

[0097] Figure 13B Waveform 1304 is shown, where a stimulation phase 30a is followed by a high impedance passive charge recovery interval during the time the ECAP is present at the sensing electrodes. After the high impedance passive charge recovery interval and sensing of the ECAP, a further active charge recovery phase 30b is used to recover the remaining charge.

[0098] Figure 13C Waveform 1306 is shown, where a stimulation phase 30a is followed by an initial low impedance passive charge recovery duration. During the time the ECAP is measured, the passive charge recovery switches to a high impedance passive charge recovery duration. Further passive charge recovery (e.g., intermediate impedance) can be used to recover the remaining charge after ECAP sensing. It should be noted that waveforms 1302, 1304, and 1306 are just a few examples of the waveforms contemplated by the present disclosure. Generally, according to the present disclosure, any combination of active and passive charge recovery can be used, where it is preferred to sense the ECAP during the high impedance passive charge recovery duration during the passive charge recovery phase.

[0099] As mentioned previously, the ECAP is just one example of a neural response that can be sensed using the disclosed techniques.

[0100] Although specific embodiments of the present invention have been shown and described, the foregoing discussion is not intended to limit the present invention to these embodiments. It will be apparent to those skilled in the art that various changes and modifications can be made without departing from the spirit and scope of the present invention. Accordingly, the present invention is intended to cover alternatives, modifications, and equivalents that may fall within the spirit and scope of the present invention as defined by the claims.

Claims

1. A stimulator device, comprising: A plurality of electrode nodes, each electrode node being configured to: couple to one of a plurality of electrodes configured to contact patient tissue; A stimulation circuit configured to provide actively driven stimulation at at least one stimulation node selected from the plurality of electrode nodes, wherein the stimulation comprises at least one pulse, and the at least one pulse comprises at least a first phase; And A passive charge recovery circuit configured to provide passively driven passive charge recovery during a passive charge recovery duration; A control circuit configured with at least one algorithm, wherein the at least one algorithm is configured to determine when a neural response will occur at a sensing electrode node; And A sensing circuit configured to: sense a neural response at at least one sensing node selected from the plurality of electrode nodes during the passive charge recovery duration.

2. The stimulator device according to claim 1, wherein, The neural response is generated in response to the actively driven stimulation.

3. The stimulator device according to claim 1, wherein, The passive charge recovery is configured to recover charge stored during the actively driven stimulation.

4. The stimulator device according to any one of claims 1-3, wherein the passive charge recovery circuit comprises a plurality of switching circuits, each of the plurality of switching circuits being coupled to a different one of the electrode nodes and configured to: provide a variable resistance between its corresponding electrode node and a common node when selected.

5. The stimulator device according to claim 4, wherein, The common node comprises a reference voltage selected from the group consisting of a battery voltage, a compliance voltage, a portion of the compliance voltage, and ground.

6. The stimulator device according to claim 4 or 5, wherein each of the plurality of switching circuits comprises a plurality of switches, and the switches are selectable to change the resistance.

7. The stimulator device according to claim 6, wherein, The plurality of switches comprises a plurality of transistors in parallel.

8. The stimulator device according to any one of claims 4-7, wherein the passive charge recovery duration comprises: A high impedance passive charge recovery duration and a low impedance passive charge recovery duration, wherein: During the high impedance passive charge recovery duration, the variable resistor is configured to provide passive charge recovery with a first impedance; and During the low impedance passive charge recovery duration, the variable resistor is configured to provide passive charge recovery with a second impedance lower than the first impedance.

9. The stimulator device according to claim 8, wherein, The sensing circuit is configured to sense the neural response during the high impedance passive charge recovery duration.

10. The stimulator device according to claim 8 or 9, wherein, The at least one algorithm is configured to time the passive charge recovery duration such that the high impedance passive charge recovery duration will exactly overlap with the time when the neural response will occur at the sensing electrode node.

11. The stimulator device according to any one of claims 1-10, wherein, The stimulation circuit is further configured to provide actively driven active charge recovery.

12. The stimulation device according to any one of claims 1-11, wherein the sensing circuit comprises a differential amplifier, and wherein the differential amplifier receives the sensing electrode node at a first input, and wherein the differential amplifier receives a reference electrode node selected from one of the electrode nodes at a second input.

13. The stimulator device according to any one of claims 1-12, wherein each electrode node is coupled to its associated electrode through a DC blocking capacitor.

14. The stimulator device according to any one of claims 1-13, wherein, The stimulator device includes an implantable pulse generator or an external trial stimulator.

Images