Automatic determination of inputs to closed-loop algorithms for stimulation parameter optimization

Abstract

Methods and systems for facilitating determining and setting stimulation parameters for programming an electrical stimulation system are disclosed. The disclosed systems and methods use an algorithm to identify a patient-specific metric to use as a feedback variable to optimize stimulation parameters for a patient. The patient-specific metric is determined by ranking multiple clinical indicators for patients with and without medical intervention to determine which clinical indicators are most responsive to the medical intervention. The most responsive clinical indicator can be used as a patient-specific metric for optimizing stimulation, or a composite patient-specific metric can be derived as a mathematical combination of multiple clinical indicators that respond well to intervention.

Description

Technical Field

[0001] The present application relates to methods and systems for providing neuromodulation, and in particular deep brain stimulation (DBS). More specifically, the present application relates to methods and systems for automatically identifying optimized variables for use as input to an algorithm for optimizing stimulation parameters. Background Art

[0002] Implantable electrical stimulator devices are devices that generate and deliver electrical stimulation to nerves and tissues for the treatment of 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 producing coordinated limb movements, spinal cord stimulators for treating chronic pain, cortical and deep brain stimulators for treating movement and psychological disorders, other neurostimulators for treating urinary incontinence, sleep apnea, shoulder subluxation, etc. The following description will focus primarily on the use of the disclosed technology within a deep brain stimulation (DBS) system, such as disclosed in U.S. Patent No. 2013 / 0184794. However, the disclosed technology may find applicability in the context of any implantable medical device and implantable medical device system.

[0003] like Figure 1 As shown, a DBS system typically includes an electrical stimulator, such as an implantable pulse generator (IPG) 10 (or more generally, an implantable medical device), which includes a biocompatible device housing 12 constructed of a metallic material such as titanium. The housing 12 typically includes two parts that are welded together, and it houses the circuitry necessary for the operation of the IPG 10 and a battery 14 ( Figure 2 ). The battery 14 may be either rechargeable or primary (non-rechargeable) in nature. The IPG 10 is coupled to the electrodes 16 via one or more electrode leads 18 (two of which are shown). The proximal ends of the leads 18 include electrode terminals 20, which are coupled to the IPG 10 at one or more connector blocks 22, which are fixed in the head 24, which may include, for example, epoxy resin. The contacts in the connector block 22 are in electrical contact with the electrode terminals 20 and communicate with the circuits within the housing 12 via feedthrough pins 26 passing through a sealed feedthrough 28 to allow such circuits to provide stimulation or monitoring electrodes 16 to various electrodes 16. The feedthrough assembly 28, which is typically a glass, ceramic or metal material, is fixed to the housing 12 at its edges to form an airtight seal. In the system shown, there are sixteen electrodes 16 separated between two leads 18, although the number of leads and electrodes is application specific and can vary accordingly.

[0004] like Figure 2As shown, assuming that the battery 14 is a rechargeable battery, the IPG 10 includes a charging coil 30 for wirelessly charging the battery 14 of the IPG using an external charging device 50. If the IPG 10 has a primary battery 14, the charging coil 30 and the external charger 50 in the IPG 10 can be omitted. The IPG 10 also includes a telemetry coil antenna 32 for wirelessly communicating data with an external controller device 40, which will be further explained below. In other examples, the antenna 32 may include a short-range RF antenna, such as a slot antenna, a patch antenna, or a wire antenna. The IPG 10 also includes control circuits, such as a microcontroller 34 and one or more application-specific integrated circuits (ASICs) 36, which may be described in, for example, USP 8,768,453. The ASIC 36 may include a current generation circuit for providing stimulation pulses at one or more electrodes 16, and may also include telemetry modulation and demodulation circuits for enabling two-way wireless communication at the antenna 32, battery charging and protection circuits that may be coupled to the charging coil 30, DC blocking capacitors in each current path leading to the electrode 16, and the like. The components within the housing 12 are integrated via a printed circuit board (PCB) 38 .

[0005] Figure 2 The external components referenced above that can be used to communicate with the IPG 10 are further shown in plan and cross-sectional views. An external controller 40 can be used to control and monitor the IPG 10 via a two-way wireless communication link 42 that passes through the patient tissue 5. For example, the external controller 40 can be used to provide or adjust a stimulation program for execution by the IPG 10 to provide stimulation to the patient. The stimulation program can specify a number of stimulation parameters, such as which electrodes are selected for stimulation; whether such active electrodes are used as anodes or cathodes; and the amplitude (e.g., current), frequency, and duration of the stimulation at the active electrodes, assuming that such stimulation includes typical stimulation pulses.

[0006] As is well known, communication on the link 42 may occur via magnetic inductive coupling between a coil antenna 44 in the external controller 40 and the telemetry coil 32 of the IPG 10. Typically, the magnetic field comprising the link 42 is modulated via frequency shift keying (FSK), etc., to encode the transmitted data. For example, data telemetry via FSK may occur near a center frequency of fc=125kHz, where a 129kHz signal represents the transmission of a logical '1' bit and a 121kHz signal represents the transmission of a logical '0' bit. However, transcutaneous communication on the link 42 need not be by magnetic induction, and may include short-range RF telemetry (e.g., Bluetooth, WiFi, Zigbee, MICS, etc.) if the antennas 44 and 32 and their associated communication circuitry are so configured. The external controller 40 is typically similar to a cell phone and includes a portable housing that can be held in a hand.

[0007] If the battery 14 of the IPG 10 is rechargeable, the external charger 50 provides power to recharge the battery 14. Such power transfer occurs by energizing a charging coil 54 in the external charger 50, which generates a magnetic field that includes a transcutaneous link 52, which may occur at a frequency (f2=80 kHz) different from the data communication on the link 42. The magnetic field 52 excites the charging coil 30 in the IPG 10, which is rectified, filtered, and used to recharge the battery 14. The link 52, like the link 42, may be bidirectional to allow the IPG 10 to report status information back to the external charger 50, such as by using the well-known Load Shift Keying. For example, once the circuitry in the IPG 10 detects that the battery 14 is fully charged, it may cause the charging coil 30 to signal this fact back to the external charger 50, so that charging may be stopped. As with the external controller 40, the external charger 50 typically includes a hand-held and portable housing.

[0008] In DBS applications, as they are useful in treating neurological disorders such as Parkinson's disease, the IPG 10 is typically implanted under the patient's clavicle (neck bone), and leads 18 (or lead extensions) are passed through the neck and between the skull and scalp, with electrodes 16 implanted through holes drilled in the skull on the left and right sides of the patient's brain, such as Figure 3 Specifically, the electrode 16 may be implanted in the subthalamic nucleus (STN), the pedunculopontine nucleus (PPN), or the globus pallidus internal (GPi). The stimulation therapy provided by the IPG 10 has shown evidence of alleviating symptoms of neurological disorders including rigidity, bradykinesia, tremor, gait and rotation disorders, postural instability, freezing, arm swinging, balance disorders, and dystonia.

[0009] Although Figure 1 The electrodes 16 are generally shown as being arranged linearly along the lead 18, but electrode leads 18 for DBS applications typically include segmented electrodes that allow for directional control of the stimulation. Figure 4 The electrode lead 18 in the embodiment includes a plurality of circumferential (or annular) electrodes and a plurality of segmented electrodes. In particular, electrodes E1 and E8 are circumferential electrodes extending around the circumference of the lead 18, while electrodes E2-E7 are segmented electrodes. As used herein, a segmented electrode refers to an electrode that does not extend completely around the periphery of the electrode lead 18. In the illustrated embodiment, the segmented electrodes are arranged in three electrodes at a particular axial position, with each segmented electrode spanning an arc of approximately 90 degrees around the lead 18, with a spacing of approximately 30 degrees between adjacent segmented electrodes. Although Figure 4, but the type and arrangement of electrodes 16 along the lead are application specific and may vary accordingly. For example, the lead may include more or fewer segmented electrodes at a given axial position, and more or fewer circumferential electrodes in addition to the segmented electrodes. As will be appreciated, because the segmented electrodes are separated by non-conductive interruptions, electrical stimulation directed to the segmented electrodes propagates outward in the direction of the electrodes, rather than uniformly around the lead 18 as with the circumferential electrodes. Lead 18 additionally includes a marker 46 aligned with the segmented electrodes E2 and E5. The marker 46 provides a visual indication of the lead orientation prior to implantation and a radiographic indication of the lead orientation after implantation.

[0010] As described above, the electrical stimulation that the IPG 10 is capable of delivering is highly customizable with respect to the selected electrodes, current amplitude and polarity, pulse duration, pulse frequency, etc. Due to the uncertainty of the position of the electrodes relative to the neural targets, the patient's physiological response to the stimulation pattern, and the nature of the electrical environment in which the electrodes are located, the stimulation parameters that may provide effective stimulation therapy for a particular patient are typically determined using a trial and error method. Therefore, after the leads are implanted, an initial programming session is typically performed to customize the stimulation parameters provided by the IPG 10 to obtain the maximum benefit for the patient. Although uncommon in DBS applications due to the risk of having external leads or lead extension, in other applications, such as spinal cord stimulation (SCS), an initial programming session is typically performed after the leads are implanted using an external test stimulator that simulates the operation of the IPG 10, and the external test stimulator is coupled to the implanted leads 18 but is not implanted itself.

[0011] refer to Figure 5 , initial programming is typically performed by transferring different stimulation programs from the clinician's programming system (CP system) 200 to the IPG 10 and observing the patient's response to the IPG 10 executing the different programs. For DBS applications, the clinician can observe to what extent the current stimulation program reduces the effects of the patient's neurological disorder (e.g., to what extent the stimulation program reduces the degree of tremor) and any side effects caused by the stimulation program. As shown, the CP system 200 may include: a computing device 202, such as a desktop computer, a portable computer, a laptop computer, a tablet computer, a mobile smart phone, a personal digital assistant (PDA) type mobile computing device, etc. (hereinafter referred to as a "CP computer"). Figure 5 In the figure, the CP computer 202 is shown as a portable computer, which includes typical computer user interface devices, such as a screen 204, a mouse, a keyboard, a speaker, a stylus, a printer, etc., which are not all shown for convenience.

[0012] Figure 5Also shown is an accessory communication head 210 that can be coupled to a port of the CP computer 202, such as a USB port 206, and that is specific to the operation of the CP computer 202 as a neurostimulator controller. The communication between the CP system 200 and the IPG 10 can be a communication scheme such as Bluetooth, Bluetooth Low Energy (BTLE), etc., or can include a magnetic induction or short-range RF telemetry scheme (as described above with respect to the communication between the IPG 10 and the programmer 40), and in this regard, the IPG 10 and the CP computer 202 and / or the communication head 210 (which can be placed near the IPG 10) can include antennas that conform to the selected telemetry device. For example, the communication head 210 can include a coil antenna 212a, a short-range RF antenna 212b, or both. The CP computer 202 can also communicate directly with the IPG 10, for example using an integrated short-range RF antenna 212b, without using the communication head 210.

[0013] If the CP system 200 includes a short-range RF antenna (in the CP computer 202 or the communication head 210), such antenna can also be used to establish communications between the CP system 200 and other devices, and ultimately for larger communication networks such as the Internet. The CP system 200 can also typically communicate with such other networks via a wired link provided at an Ethernet or network port 208 on the CP computer 202, or use other wired connections (e.g., at a USB port 206) to communicate with other devices or networks.

[0014] To test different stimulation parameters during an initial programming session, the user interfaces with a clinician programmer graphical user interface (CP GUI) 94 provided on a display 204 of a CP computer 202. As will be appreciated by those skilled in the art, the CP GUI 94 may be presented by executing CP software 96 on the CP computer 202, which may be stored on a non-volatile memory 220 of the CP computer 202. Those skilled in the art will additionally appreciate that the execution of the CP software 96 in the CP computer 202 may be facilitated by control circuitry 222, such as a microprocessor, microcomputer, FPGA, other digital logic structure capable of executing a program in a computing device, etc. Such control circuitry 222, when executing the CP software 96, in addition to presenting the CP GUI 94, will cause the CP computer 202 to communicate the stimulation parameters to the IPG 10 in the communication head 210 or the CP computer 202 using the appropriate antenna 212a or 212b, as previously explained. The CP software 96 enables the user to select the type of electrode leads that have been implanted (e.g., from a list of leads configured in the software 96) and customize the stimulation parameters using the available electrodes on the implanted leads. In this way, the user can pass different stimulation parameters to the IPG 10 for execution to observe the effects of the various parameters and determine the appropriate settings for the patient.

[0015] After implantation, the clinician will program the implantable pulse generator using a clinician programmer, remote controller, or other programming device. According to at least some programming techniques, the clinician enters stimulator parameters for the stimulation program and the stimulation program is used to stimulate the patient. The clinician observes the patient's response. In at least some cases, the clinician asks the patient to describe, evaluate, or otherwise provide information about the effects of the stimulation, such as which part of the body is affected, how strong the stimulation effect is, whether there are side effects or negative effects, etc.

[0016] In at least some cases, the clinician may be remote from the patient. For example, the clinician may be in another room, treatment or care facility, city, state, or even country. This may be advantageous because it may allow a skilled clinician to interact with a patient who is remote from the clinician without requiring travel or wasted time by the clinician. As another example, the patient may be sent home after surgery, and the clinician may remotely program the device while the patient is at home.

[0017] However, such remote programming may encounter difficulties such as, for example, limited bandwidth, speech-impaired patients, deaf patients, patients who speak a different language than the clinician, noise from the telephone or the telephone or video connection between the patient and the clinician, patient hearing difficulties due to hearing aid devices, etc.

[0018] Furthermore, current techniques for programming electrical stimulation systems are tedious trial-and-error processes that explore only certain possible programming parameters. With the introduction of multiple independent sources of current and complex lead geometries, the exploration of which parameters to manipulate and how to program them becomes more difficult to express and visualize, as well as to set up and track. Therefore, there is a need to automate and / or simplify the programming process. Summary of the invention

[0019] Disclosed herein is a method for evaluating stimulation therapy in a patient with an electrical stimulator, the method comprising: (a) selecting one or more of a plurality of patient responses (which are examples of clinical indicators) to be used to evaluate the stimulation therapy by determining a first severity of a plurality of patient responses in the absence of an intervention, determining a second severity of a plurality of patient responses in the presence of an intervention, and selecting one or more patient responses using the first severity and the second severity, and (b) evaluating the stimulation therapy by (i) causing the electrical stimulator to apply a stimulation parameter set to stimulate the patient, (ii) in response to applying the stimulation parameter set, determining a value for each of the one or more selected patient responses, (iii) adjusting the stimulation parameter set based on the determined one or more values, and (iv) repeating steps (i)-(iii) until a stop condition is reached. In other words, the severity of the patient responses with and without the intervention is used to derive one or more patient-specific metrics to evaluate the stimulation therapy. According to some embodiments, the method also includes assigning a weight to each patient response based on the first severity and the second severity, and wherein selecting one or more patient responses is based on the assigned weight. According to some embodiments, selecting one or more patient responses includes determining a composite feedback variable (which is an example of a patient-specific metric) that is a mathematical combination of multiple patient responses, each standardized by its assigned weight. According to some embodiments, the composite feedback variable is a linear combination of more than one patient response, each standardized by its respective weight. According to some embodiments, the intervention includes providing a medicament to the patient. According to some embodiments, the medicament is levodopa. According to some embodiments, the intervention includes providing electrical stimulation to the patient. According to some embodiments, the multiple patient responses include one or more of speech, tremor, rigidity, finger tapping, toe tapping, bradykinesia, hypokinesia, agile posture, gate, or postural stability. According to some embodiments, one or more of the multiple patient responses are automatically determined based on one or more sensors associated with the patient. According to some embodiments, determining the first severity and the second severity includes ranking the severity of each patient response in the absence and presence of the intervention. According to some embodiments, the method also includes assigning a weight to each patient response based on the ranking of the severity of each patient response in the absence and presence of the intervention. According to some embodiments, the stimulation therapy includes deep brain stimulation. According to some embodiments, the set of stimulation parameters includes one or more of stimulation contact configuration, amplitude, frequency, or pulse width.

[0020] Also disclosed herein is a computing system for evaluating stimulation therapy in a patient with an electrical stimulator, the system comprising: a processor, and a non-transitory computer-readable medium comprising instructions that, when executed by the processor, configure the computing system to: (a) select one or more of the plurality of patient responses to be used for evaluating stimulation therapy by determining a first severity of the plurality of patient responses in the absence of an intervention, determining a second severity of the plurality of patient responses in the presence of an intervention, and selecting the one or more patient responses using the first severity and the second severity, and (b) evaluate the stimulation therapy by (i) causing the electrical stimulator to apply a stimulation parameter set to stimulate the patient, (ii) in response to applying the stimulation parameter set, determining a value for each of the one or more selected patient responses, (iii) adjusting the stimulation parameter set based on the determined one or more values, and (iv) repeating steps (i)-(iii) until a stop condition is reached. According to some embodiments, selecting one or more of the plurality of patient responses to be used for evaluating stimulation therapy also includes: assigning a weight to each patient response based on the first severity and the second severity, and wherein selecting the one or more patient responses is based on the assigned weight. According to some embodiments, selecting one or more patient responses comprises determining a composite feedback variable that is a mathematical combination of a plurality of patient responses, each patient response normalized by its assigned weight. According to some embodiments, the intervention comprises providing a medicament to the patient. According to some embodiments, the intervention comprises providing electrical stimulation to the patient. According to some embodiments, one or more of the plurality of patient responses are automatically determined based on one or more sensors associated with the patient. According to some embodiments, the stimulation therapy comprises deep brain stimulation.

[0021] Also disclosed herein is a method for optimizing stimulation for a patient with an implantable pulse generator (IPG), the method comprising: determining one or more patient-specific metrics for the patient based on a plurality of clinical indicators by: assigning a first ranking to each clinical indicator based on an evaluation of the patient's clinical indicators in the absence of medical intervention, providing the patient with medical intervention, assigning a second ranking to each clinical indicator based on an evaluation of the patient's clinical indicators in the presence of medical intervention, determining one or more patient-specific metrics based on the first ranking and the second ranking for each clinical indicator, and optimizing stimulation for the patient using the determined one or more patient-specific metrics. According to some embodiments, optimizing stimulation for the patient using the determined one or more patient-specific metrics comprises: (i) causing an electrical stimulator to apply stimulation to the patient according to a stimulation parameter set, (ii) determining a value for the one or more patient-specific metrics, (iii) adjusting the stimulation parameter set based on the value, and (iv) repeating steps (i) to (iii) until a stop condition based on the value of the patient-specific metrics is reached. According to some embodiments, the one or more patient-specific metrics include the selected one or more clinical indicators. According to some embodiments, the selected one or more clinical indicators are selected based on their respective first rankings and second rankings. According to some embodiments, the method further comprises assigning a weight to each clinical indicator based on the first and second rankings for each clinical indicator. According to some embodiments, the one or more patient-specific metrics comprise a mathematical combination of a plurality of the clinical indicators, each of which is normalized using its assigned weight. According to some embodiments, the one or more patient-specific metrics comprise a linear combination of a plurality of the clinical indicators, each of which is normalized by its respective weight. According to some embodiments, the medical intervention comprises providing a medicament to the patient. According to some embodiments, the medicament is levodopa. According to some embodiments, the medical intervention comprises providing electrical stimulation to the patient. According to some embodiments, the plurality of clinical indicators comprises one or more of speech, tremor, rigidity, finger tapping, toe tapping, bradykinesia, hypokinesia, agile posture, gate or postural stability. According to some embodiments, assigning the first and second rankings to at least one of the clinical indicators is based on measurements from one or more sensors associated with the patient. According to some embodiments, the stimulation therapy comprises deep brain stimulation. According to some embodiments, the stimulation parameters comprise one or more of stimulation contact configuration, amplitude, frequency or pulse width.

[0022] Also disclosed herein is a system for optimizing stimulation for a patient with an implantable pulse generator (IPG), the system comprising: a processor, and a non-transitory computer-readable medium comprising instructions, which, when executed by the processor, configure the system to: determine one or more patient-specific metrics for the patient based on a plurality of clinical indicators by: assigning a first ranking to each clinical indicator based on an evaluation of the patient's clinical indicators in the absence of medical intervention, assigning a second ranking to each clinical indicator based on an evaluation of the patient's clinical indicators in the presence of medical intervention, and determining one or more patient-specific metrics based on the first ranking and the second ranking for each clinical indicator, and optimizing stimulation for the patient using the determined one or more patient-specific metrics. According to some embodiments, optimizing stimulation for the patient using the determined one or more patient-specific metrics comprises: (i) causing an electrical stimulator to apply stimulation to the patient according to a stimulation parameter set, (ii) determining a value for the one or more patient-specific metrics, (iii) adjusting the stimulation parameter set based on the value, and (iv) repeating steps (i) to (iii) until a stop condition based on the value of the patient-specific metrics is reached. According to some embodiments, the one or more patient-specific metrics comprise the selected one or more clinical indicators. According to some embodiments, the instructions further configure the processor to assign a weight to each clinical indicator based on the first and second rankings for each clinical indicator. According to some embodiments, the one or more patient-specific metrics include a mathematical combination of a plurality of the clinical indicators, each of which is normalized using its assigned weight. According to some embodiments, the medical intervention includes providing a pharmaceutical agent to the patient. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 An implantable pulse generator (IPG) with an electrode array is shown.

[0024] Figure 2 Shows Figure 1 Figure 2. Cross section of an IPG implanted in a patient, along with external devices supporting the IPG, including an external charger and an external controller.

[0025] Figure 3 An IPG is shown implanted in a patient in a deep brain stimulation (DBS) application.

[0026] Figure 4 An electrode lead with segmented electrodes that can be used for DBS applications is shown.

[0027] Figure 5 Components of a clinician's programming system are shown, including components for communicating with a neurostimulator such as an IPG.

[0028] Figure 6A schematic block diagram of a feedback loop stimulation parameter control system for determining electrical stimulation programming settings is shown.

[0029] Figure 7 An example display screen of a user interface for visualizing stimulation parameter determination using a feedback loop stimulation parameter control system is shown.

[0030] Figure 8 An example workflow diagram for selecting stimulation programming settings using a feedback variable selection algorithm and a feedback loop stimulation parameter control system is shown.

[0031] Fig. 9 An example workflow diagram of a feedback variable selection algorithm is shown.

[0032] Fig.10 A list of patient responses (eg, symptoms) in therapy OFF and therapy ON states and the weights associated with the patient responses are shown.

[0033] Figures 11A-11C An example display screen for assessing patient symptoms is shown.

[0034] Fig.12 An example display screen for selecting symptoms to be used in evaluating stimulation therapy is shown.

[0035] Fig.13 An example display screen for ranking patient responses during an evaluation of stimulation therapy is shown.

[0036] Fig.14 An example workflow diagram for selecting stimulation programming settings using a feedback variable selection algorithm and a feedback loop stimulation parameter control system is shown.

[0037] Fig.15 A schematic block diagram of one embodiment of a computing system for programming an electrical stimulation system is shown. DETAILED DESCRIPTION

[0038] As described above, current techniques for programming electrical stimulation systems are a tedious trial-and-error process. The present disclosure relates to methods and systems for automating the programming of electrical stimulation systems, such as the IPG 10 (100) used for neuromodulation therapies such as DBS. Figure 1). U.S. Patent Publication No. US 2018 / 0104500 (the "'500 Application"), published on April 19, 2018, entitled "Systems and Methods for Closed-Loop Determination of Stimulation Parameter Settings for an Electrical Simulation System," the entire contents of which are hereby incorporated by reference herein, describes methods and systems for facilitating determination and setting of stimulation parameters for programming an electrical stimulation system using closed-loop programming. In the system described in the '500 Application, pulse generator feedback logic is executed by a processor to interface with control instructions of an implantable pulse generator by incorporating one or more machine learning engines to automatically generate a set of proposed stimulation parameter values, each of which affects a stimulation aspect of the implantable pulse generator, receive one or more patient responses (e.g., clinical responses), automatically generate a set of revised values ​​in view of the received patient responses, and repeat the automatic receipt of patient responses and adjustment of stimulation parameter values ​​in view of the patient responses until or unless a stop condition is reached or the therapeutic response indication is within a specified tolerance range.

[0039] Figure 6 An example feedback loop stimulation parameter control system (FLSPCS) as described in the '500 application is shown. In at least some embodiments, the FLSPCS 600 includes pulse generator feedback control logic 601 that determines and sends parameters (604) to configure a deep brain stimulator controller 605 (such as an IPG 10, Figure 1 The deep brain stimulator controller 605 then modifies the various leads (such as Figure 1 , 34 and 18) stimulation parameters, which may have different effects on the patient 606 implanted with the IPG. The patient 606 provides feedback to the pulse generator feedback control logic 601, which is reported as a patient response value 608. In some embodiments, the patient 606 automatically or semi-automatically provides feedback via a sensor 607, which is sent to the pulse generator feedback control logic 601 as a patient response value 608. According to some embodiments, the feedback is determined by a clinician based on the clinician's observation of the patient. The control logic 601 then adjusts the value of the stimulation parameter 603 based on the received feedback 608 and based on the model implemented by one or more machine learning engines 602. The control logic 601 then sends these adjusted (new or modified) stimulation parameter values ​​(604) to further configure the controller 605 to change the stimulation parameters of the leads implanted in the patient 606 to the adjusted values. The feedback control loop then continues until a result that is considered to be optimized, desired, or acceptable is reached, or until some other stop condition is reached, such as the number of iterations, the time spent in the programming session, etc. Results may be considered optimized, desired, or acceptable if they meet certain thresholds or tests (e.g., better outcomes for patients, faster device programming, extended battery life, and / or control of multiple independent current sources and directional leads).

[0040] Aspects of the FLSPCS 600 are described in detail in the incorporated '500 application and will not be repeated in detail here. Figure 7 It is to give a brief understanding of how the algorithm operates. Figure 7 An example of a user interface (UI) 700 is shown that is operable when a clinician executes FLSPCS 600 to optimize which electrode contacts to use for stimulation (i.e., where stimulation is to be performed) and the optimal amplitude at which stimulation is to be performed. Figure 7 The proposed parameters (604, Figure 6 ) includes a contact configuration and an amplitude value. UI 700 shows a grid 702 with contact numbers on the y-axis (contacts 1-8) and amplitude values ​​on the x-axis. The UI also includes controls 704 for inputting / adjusting those parameters. The algorithm (i.e., FLSPCS 600) is configured to suggest proposed parameter values ​​to try. The UI also includes controls 706 to input / record observed therapeutic responses elicited by the suggested parameters. Figure 7 In the present invention, the observed patient response is bradykinesia, which can be automatically measured using a sensor such as an accelerometer worn by the patient. For example, while wearing an accelerometer, the patient may be prompted to perform finger tapping exercises.

[0041] As the algorithm runs, the patient is stimulated with various proposed parameter sets, which are represented by black circles in the grid 702. The algorithm uses the observed / recorded patient responses to the stimulation and constructs a treatment map within the grid that represents ideal treatments (indicated by the shaded areas) and suboptimal treatments (indicated by the cross-hatched areas). As data is collected, the algorithm suggests further parameters to try. The next suggested parameter set is indicated by Figure 7 The algorithm continues until a result is reached that is considered optimal, desired, or acceptable, or until some other stopping criterion is reached, such as the number of iterations, time spent in a programming session, etc.

[0042] Please note that Figure 7 In, the FLSPCS algorithm uses a single patient response, bradykinesia, as a measurable feedback variable for optimizing stimulation. The feedback variable can be selected from different feedback variables. The inventors have observed that the selection of one or more appropriate feedback variables can greatly affect the ability of the algorithm to optimize treatment for a particular patient. For example, one patient may experience significant improvement in bradykinesia, but another patient may not. However, another patient may see significant improvement in other symptoms, such as rigidity or tremor. Therefore, the methods and systems disclosed in the present application can be used to select one or more appropriate feedback variables to optimize stimulation parameters. In other words, aspects of the present disclosure relate to determining one or more patient-specific metrics that can be used to optimize stimulation for a particular patient. As described in more detail below, the patient-specific metric can be a weighted combination of multiple patient responses.

[0043] Figure 8 An example of a workflow 800 as disclosed herein is shown. At step 802, as described in further detail below, a feedback variable selection algorithm (FVSA) is used to determine the best patient-specific metric to use as an input to a stimulation optimization algorithm (such as the FLPCS described in the '500 application). At step 804, the FLPCS is used to determine optimized stimulation parameters for treatment of the patient. Once the parameters are determined, the patient's system is programmed with the optimized stimulation parameters as step 806. According to some embodiments, Figure 8 Each of the steps shown in can be performed on an external system capable of programming / controlling the patient's electrical stimulator device. For example, the external system may include a clinician's programmer (e.g., Figure 5 200) and / or a remote controller (e.g., Figure 240). According to some embodiments, all three of these steps can be performed using a clinician's programmer. Alternatively, the external system can include an additional computing device. For example, steps 802 and / or 804 can be performed on a computing device other than the clinician's programmer, such as a tablet, a personal computer, a smart device, etc., and the clinician's programmer can be used for step 806. According to some embodiments, some aspects of any of steps 802, 804, and / or 806 can be performed by the patient's stimulator device (e.g., IPG 10, Figure 1 , 2 For example, according to some embodiments, aspects of FVSA and / or FLSPCS may be embodied in IPG to provide continuous optimization of stimulation parameters outside of a clinical setting.

[0044] Fig. 9 A workflow 900 of FVSA is shown for selecting one or more appropriate patient-specific metrics to use as input to an algorithm for optimizing stimulation parameters. According to some embodiments, the goal of FVSA is to identify one or more patient responses (also referred to herein as clinical indicators) or an appropriate combination of patient responses that are expected to respond well to treatment and therefore provide good feedback variables for optimizing stimulation settings. Clinical indicators can include patient symptoms, treatment responses, etc. In other words, the goal of FVSA is to identify one or more patient-specific metrics to optimize treatment for a particular patient.

[0045] Step 902 in the workflow includes identifying possible clinical indicators to be evaluated. According to some embodiments, clinical indicators can be selected from the standard battery of tests for diagnosing neurological disorders, such as the motor reactions specified by UPDRS (e.g., UPDRS-Part III) known in the art. Other examples may include movement disorder UPDRS (MDS-UPDRS), and / or readings from sensors with arbitrary scales. The example of clinical indicators that can be evaluated includes speech, tremor at different anatomical locations, rigidity at different anatomical locations, the ability to perform tasks such as finger tapping or toe tapping, bradykinesia / hypokinesia, agility, posture, gate, posture stability, etc. It should be appreciated that some clinical indicators are axial, i.e. they are not limited to the left or right side of the body. Other clinical indicators are left or right specific. Selected clinical indicators will constitute the scope of possible clinical indicators, and these clinical indicators will be evaluated as potential patient-specific metrics.

[0046] Once the range of clinical indicators is determined, the severity of each clinical indicator is assessed in the treatment-off state and the treatment-on state to determine how each clinical indicator responds to treatment. The treatment-off state is simply the absence of any therapeutic intervention. According to some embodiments, the treatment-on state can be the provision of stimulation (e.g., DBS). As an example of DBS, which can be provided as a treatment-on state, typically a monopolar check is performed as the first step in defining the initial stimulation parameters. The monopolar check involves independently activating each contact with a fixed pulse width and frequency (e.g., 60 μs and 130 Hz, respectively), and increasing the stimulation amplitude (current or voltage) until the side effect and treatment (usually tonic inhibition) thresholds are determined. The contact that produces the largest treatment window (the difference between the side effect and treatment threshold) is typically used for initial stimulation.

[0047] Additionally (or alternatively), the treatment onset state may be the provision of a drug intervention, such as levodopa. For example, it is generally recognized that clinical indicators that respond well to levodopa generally also respond well to DBS. Typically, patients who are considered possible candidates for DBS implantation undergo a "levodopa surge" to determine how their clinical indicators respond to levodopa. Thus, the patient's levodopa surge results may be used to determine appropriate clinical indicators to use as patient-specific metrics.

[0048] The example workflow shown includes first evaluating clinical indicators in the therapy off state (step 904 below), and then evaluating clinical indicators in the therapy on state (step 906 below). However, it should be understood that these steps can be performed in the reverse order.

[0049] At step 904 of workflow 900, the severity of each clinical indicator is evaluated and ranked in the treatment off state. In other words, the severity of each clinical indicator is evaluated when the patient is not subjected to any intervention (such as stimulation or drug intervention) to determine the baseline ranking of each clinical indicator. Please note that "ranking" does not necessarily mean "sorting" the clinical indicators from the lightest to the most severe, but more typically means: assigning each indicator its own independent ranking value based on the severity of the clinical indicator. According to some embodiments, clinical indicators can be numerically ranked based on the severity of the clinical indicator. For example, as in UPDRS-Part III, clinical indicators can be ranked on a scale of 0 to 4, where 0=normal; 1=mild; 2=mild; 3=moderate; and 4=severe. Depending on the specific implementation of workflow 900, other ranking levels can be used. Clinical indicators can be ranked based on patient feedback and / or clinical observations and based on sensor measurements. According to some embodiments, a combination of measurement criteria is used. For example, the severity of some clinical indicators can be ranked based on observations, while the severity of other clinical indicators can be ranked based on sensor readings. According to some embodiments, an assessment of clinical indicators of treatment off may be obtained from a patient's preoperative levodopa surge evaluation.

[0050] At step 906 of workflow 900, the severity of each clinical indicator is evaluated and ranked in the treatment on state. Likewise, the treatment on state may include providing stimulation, such as DBS. Alternatively (or in addition), the treatment on state may include providing medication, such as levodopa. According to some embodiments, the evaluation of the treatment on clinical indicators may be obtained from the patient's preoperative levodopa challenge evaluation.

[0051] At step 908 of workflow 900, a weight is assigned to each clinical indicator based on the difference between the ranking of the clinical indicator in the treatment-off state and the treatment-on state. In other words, the weight assigned to the clinical indicator indicates how the clinical indicator responds to the treatment. According to some embodiments, the clinical indicator can be weighted based on a mathematical analysis such as principal component analysis PCA. According to other embodiments, other weighting algorithms can be used.

[0052] Fig.10 An example of clinical indicators measured and ranked in the therapy-off and therapy-on states and the weights assigned to each clinical indicator is shown. Fig.10 In the , weights are assigned to clinical indicators according to the following rules: (1) If the ranking of a clinical indicator in the treatment-off state is 0, the clinical indicator is ignored (i.e., no rating is assigned); (2) If the ranking of a clinical indicator in the treatment-off state is > 0, the weighting factor W is calculated by W = (R T-OFF –R T-ON) 2 / (X*R T-OFF ) is determined, where R T-OFF and R T-ON are the rankings of the clinical indicator in the treatment-off state and the treatment-on state, respectively, and X is the value of the most severe possible ranking of the clinical indicator. In the illustrated example, ranking 4 corresponds to the most severe clinical indicator, and therefore, in the illustrated example, X=4. This weighting formula takes into account the magnitude of improvement and the end state of the clinical indicator when treatment is present. In other words, according to the weighting formula, an improvement that results in the absence of the clinical indicator should be ranked higher than an improvement that results in the presence of the clinical indicator. For example, please note that Fig.10 Speech and facial expressions in the symptom score were each assigned a weight of 0.333, which increased from a rank of 3 in the absence of treatment to 1 in the presence of treatment (an overall increase of 2 ranking points). However, facial tremor was given a weight of 0.500, which increased from a rank of 2 to 0 (also an overall increase of 2 ranking points). Facial tremor was assigned a higher weight because the treatment resulted in the symptom being completely absent.

[0053] According to some embodiments, a user (e.g., a clinician) can manually modify / edit or assign weighting factors associated with clinical indicators. For example, a particular clinician may prefer certain clinical indicators as feedback variables and may choose to weight those clinical indicators higher than the weighting factors determined by the formula. According to some embodiments, if the improvement found during the stimulation setting optimization session is greater than the improvement measured before the session, the weight or clinical indicator can be increased. For example, if the ankylosing score is 4 in the absence of treatment, and after treatment, the score is 1, this will result in a weight of 0.75. However, if the ankylosing score of the stimulation setting reaches 0 during the stimulation setting optimization session, this will result in a weight of 1. Therefore, the weights of all selected symptoms should be updated to reflect the potential benefit of the stimulation setting for ankylosing.

[0054] Please note, Fig.10 The list of clinical indicators shown in includes four clinical indicators that are measured using sensors worn by the patient (such as accelerometers) (in Fig.10 Likewise, any number of clinical indicators may be considered, and they may be sensor-based, observation-based, or a combination of both.

[0055] Reference again Fig. 9 At step 910, the workflow 900 may select which clinical indicators are to be used as patient-specific metrics. The selection may be determined based on the weights assigned in step 908. For example, all clinical indicators whose weights exceed a predefined threshold may be selected. Alternatively, a user may manually select clinical indicators to be used as patient-specific metrics.

[0056] At step 912, workflow 900 can determine a composite feedback parameter based on the selected clinical indicator. For example, the composite feedback parameter can be determined as a linear combination of the selected clinical indicators, where each indicator is weighted by their assigned weight. The determined composite feedback parameter can be used as a patient-specific metric for optimizing stimulation parameters, for example, using FLSPCS, as described in the '500 application (step 912).

[0057] According to some embodiments, the disclosed methods and systems provide a menu-driven UI to assess a patient's clinical indicators in both therapy-off and therapy-on states. Fig.11A and 11B Page 1100 of a UI is shown, which allows clinicians to rank the clinical indicators of patients in both treatment / drug off and on states. The UI page provides a list of various clinical indicators to be evaluated ( Fig. 9 As described above, one or more sensors (such as accelerometers, etc.) can be used to automatically measure one or more patient responses. Therefore, the system can provide a UI page such as Fig. 11C As shown, it is used to select which clinical indicators are to be measured using the sensor.

[0058] Clinicians can use page 1100 ( Fig.11A and 11B ) to evaluate the patient's clinical indicators in the treatment-off and treatment-on states (steps 904 and 906, Fig. 9 After completing both assessments, the system can assign weights to each clinical indicator based on the improvement between the two states (step 908, Fig. 9 ), similar to Fig.10 As described above, the weights can be determined based on one or more mathematical operations, such as PCA or other mathematical formulas. The weight list determined for each clinical indicator can be printed as a report (similar to Fig.10 The list shown in ) may be reflected on the screen of the UI.

[0059] Once the weights are determined for each clinical indicator, the UI may provide a screen 1200 such as Fig.12As shown, the screen allows a user (e.g., a clinician) to select which clinical indicators are to be included in a composite patient-specific metric for optimizing stimulation settings. According to some embodiments, different clinical indicators can be selected as patient-specific metrics for programming different electrode leads (e.g., for programming left and right hemisphere leads). According to some embodiments, one or more clinical indicators can be automatically populated into screen 1200, for example, if the weight of the clinical indicator exceeds a certain threshold. According to some embodiments, for example, a user can select a clinical indicator from a drop-down menu to be included in the composite patient-specific metric. According to some embodiments, a user can select as many or as few clinical indicators as desired.

[0060] According to some embodiments, screen 1200 may also reflect the weights assigned to each clinical indicator. As described above, the weights may be mathematical and / or determined from PCA or other methods and may be automatically populated into screen 1200. According to some embodiments, the user may be able to adjust the assigned weights based on user experience or other factors.

[0061] Once the user has selected clinical indicators to include in the composite patient-specific metric for optimizing stimulation settings, the UI may present screen 1300, such as Fig.13 For example, assume that the user (i.e., clinician) has selected clinical indicator 1302 as a component clinical indicator to be included in the composite patient-specific metric to program a lead in the left hemisphere of the patient's brain. The clinician will then use a feedback loop stimulation parameter control system (FLSPCS) (as described in the '500 application) to determine candidate stimulation settings for programming stimulation at the lead, as described below.

[0062] Fig.14 A workflow 1400 is shown that a user can implement to determine stimulation settings using screen 1300 of the FVSA algorithm as described above, the FLSPCS described in the '500 application, and the stimulator control UI options of a clinician's programmer. As described above, the FVSA algorithm, the FLSPCS algorithm, and the associated UI options can be implemented as modules on the same computing device, such as a clinician's programmer. Alternatively, one or more of the FVSA / FLSPCS can be implemented on a separate computing device, and the clinician can transmit the stimulation parameters determined by the FVSA / FLSPCS to the implanted stimulator device using only the clinician's programmer.

[0063] At step 1402, the FLSPCS receives an indication of a plurality of stimulation parameters to be set. For example, the set of stimulation parameters being adjusted may include contact location, amplitude, frequency, pulse width, etc. Directional leads may include additional parameters such as x, y, and z coordinates, angles, and shapes. Other parameters and parameterizations are discussed in the incorporated '500 application.

[0064] At step 1404, the FLSPCS generates an initial proposed set of stimulation parameter values. These include values ​​for each stimulation parameter that is being adjusted. For example, when the stimulation parameter set includes contact position and amplitude, the initial proposed set may be "contact = 3" and "amplitude = 2mA", etc. In one example embodiment, the initial proposed set of stimulation parameter values ​​is arbitrarily selected because data for the patient has not yet been collected. In another example embodiment, the logic is extended to use a distribution of optimized settings (e.g., amplitude and / or contact position) from previous patients to pick starting settings for a new patient. This allows the clinician to begin a programming session using settings that other patients have found beneficial. In addition, a new patient can be matched to a previous patient via external information, such as a diagnosis, imaging data, or baseline clinical indicator severity.

[0065] At step 1406, the user forwards the proposed set of stimulation parameters to the implanted electrical stimulator device, typically using a clinician's programmer. At step 1408, the user forwards the proposed set of stimulation parameters to the implanted electrical stimulator device, for example, using a screen such as screen 1300 ( Fig.13 ) screen ranks the component clinical indicators of the composite patient-specific metric and provides the ranking value to the FVSA. According to some embodiments, the clinician can wait for a period of time before ranking the component clinical indicators to account for any "wash-in" period of the applied stimulation.

[0066] At step 1410, the FVSA may determine a composite patient-specific metric value based on the determined ranking of each component clinical indicator and the weight assigned to each clinical indicator. At step 1412, the determined composite patient-specific metric is provided to the FLSPCS. At step 1414, the logic of the FLSPCS determines whether an end condition is reached. For example, the logic may determine whether the composite patient-specific metric is within a specified threshold. In some embodiments, other stop conditions are specified, such as a certain number of iterations have been performed, the treatment response changes by less than x% with y number of iterations, or the user is satisfied with the results. Other stop conditions are also possible. If the stop condition is reached, the process stops.

[0067] If the stopping condition has not been reached, then at step 1416, the FLSPCS generates a modified set of proposed stimulation parameter values. The process defined in steps 1406-1412 is then repeated. The workflow 1400 may be iteratively repeated until an appropriate stopping condition is reached.

[0068] As the workflow 1400 is iteratively repeated, the FLSPCS can construct and display a treatment map similar to Figure 7 , . Recall that in treatment map 702, a single treatment response (bradykinesia in case 702) is mapped as a function of stimulation settings. In embodiments using FVSA, such as workflow 1400, the treatment map is based on a determined composite patient-specific metric response, which can be mapped as a function of stimulation settings. According to some embodiments, a map for a single clinical indicator can also be constructed, although this embodiment is not shown. According to some embodiments, based on the treatment map, stimulation settings that take into account different target areas can be recommended. For example, if spasticity is best treated when electrode 1 is activated at 3 mA, but this setting is not optimal for finger tapping, for which activation of a current distribution of 40% electrode level 3 and 60% electrode level 4 at a total current of 2 mA is optimal, then the FLSPCS can recommend as a stimulation setting a combination of activating electrode 1 and levels 3 and 4.

[0069] As described above, methods, systems, and algorithms (such as FVSA) for determining appropriate patient-specific metrics (such as composite patient-specific metrics for optimizing stimulation settings) can be implemented on a computing device, which can be a clinician's programmer or can be separate from the clinician's programmer. Fig.15 One embodiment of a computing device 1500 for running a software, hardware, or firmware embodiment of a FVSA is shown. According to some embodiments, the computing device may also be configured to run a software, hardware, or firmware embodiment of a stimulation parameter feedback loop control system, such as described in the '500 application. The computing device 1500 includes a processor 1502 and a memory 1504, a display 1506, and an input device 1508. Optionally, the computing device 1500 may be connected to or otherwise integrated into an electrical stimulation system 1510 (which may include an Figure 1 , 2 5). For example, in some embodiments, computing device 1500 is part of electrical stimulation system 1510, such as clinician programmer 200 ( Figure 5 )、remote controller 40( Figure 2 ) or part of an external test stimulator (not shown). In other embodiments, computing device 1500 is separate from electrical stimulation system 1510.

[0070] Computing device 1500 may be a computer, tablet, mobile device, or any other suitable device for processing information. Computing device 1500 may be local to a user, or may include components that are not local to a computer, including one or both of processor 1502 or memory 1504 (or portions thereof). For example, in some embodiments, a user may operate a terminal connected to a non-local processor or memory.

[0071] Other examples of computing devices may include watches, wristbands, smart phones, etc. Such computing devices may communicate wirelessly with other components of the electrical stimulation system, such as the clinician programmer 200 ( Figure 5 )、remote controller 40( Figure 2 ), external test stimulator (not shown) or IPG 10 ( Figure 1 ). Such devices may also be used to collect patient information such as general activity level or present questions or tests to the patient to identify or score pain, depression, stimulation effects or side effects, cognitive abilities, etc. For example, the device may require the patient to undergo regular testing of cognitive abilities (e.g., daily) to monitor, for example, Alzheimer's disease. Such devices may also be used to sense clinical indicators of treatment such as, for example, tremor, heart rate, etc. The device may be used with a clinician programmer 200 ( Figure 5 )、remote controller 40( Figure 2 ), external test stimulator (not shown) or IPG 10 ( Figure 1 ) to guide changes in stimulation parameters using a closed-loop algorithm, as described below, which may be on the device or the IPG. The device may be given to the patient to wear only during programming; alternatively, it may be worn all the time and continually adjust stimulation parameters. The algorithm may be on the patient's phone, which is connected to the IPG and possible evaluation devices (such as a wristband or watch) to change stimulation parameters. These devices may also be used to record information and send it to the clinician.

[0072] Computing device 1500 may utilize any suitable processor 1502, including one or more hardware processors that may be local to a user or non-local to a user or other components of the computing device. Processor 1502 is configured to execute instructions provided to the processor.

[0073] Any suitable memory 1504 may be used for computing device 1500. Memory 1504 illustrates a computer-readable medium, i.e., a computer-readable storage medium. Computer-readable storage media may include, but are not limited to, non-volatile, non-temporary, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules, or other data. Examples of computer-readable storage media include RAM, ROM, EEPROM, flash memory or other storage technology, CD-ROM, digital versatile disk ("DVD") or other optical storage, magnetic tape, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computing device.

[0074] The communication method provides another type of computer-readable media; namely, communication media. Communication media typically contains computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave, data signal, or other transport mechanism, and includes any information delivery media. The terms "modulated data signal" and "carrier signal" include a signal that has one or more characteristics set or changed in such a manner as to encode information, instructions, data, etc. in the signal. By way of example, communication media includes wired media such as twisted pair, coaxial cable, optical fiber, waveguide, and other wired media, and wireless media such as acoustic, RF, infrared, Bluetooth, and the like. TM , near field communications and other wireless media.

[0075] Display 1506 can be any suitable display or presentation device, such as a monitor, screen, display, etc., and can include a printer. Input device 1508 can be, for example, a keyboard, a mouse, a touch screen, a trackball, a joystick, a voice recognition system, or any combination thereof, etc. Another input device 1508 can be a camera from which the clinician can observe the patient. Yet another input device 1508 is a microphone in which the patient or clinician can provide responses or queries.

[0076] Although specific embodiments of the present invention have been shown and described, it should be understood that the above 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 may be made without departing from the spirit and scope of the present invention. Therefore, the present invention is intended to encompass substitutions, modifications and equivalents that may fall within the spirit and scope of the present invention as defined by the claims.

Claims

1. A system for optimizing stimulation therapy for a patient having an implantable pulse generator (IPG), the system comprising: processor, and A non-transitory computer-readable medium comprising instructions that, when executed by the processor, configure the system to: Determine one or more metrics for the patient based on multiple clinical indicators by: assigning a first ranking to each clinical indicator based on an assessment of the clinical indicator for the patient in the absence of medical intervention, assigning a second ranking to each clinical indicator based on the evaluation of the clinical indicator for the patient in the presence of a medical intervention, and The one or more metrics are determined based on the first ranking and the second ranking for each clinical indicator, and Stimulation is optimized for the patient using the determined one or more metrics.

2. The system according to claim 1, wherein: Optimizing stimulation for the patient using the determined one or more metrics includes: (i) causing an electrical stimulator to apply stimulation to the patient according to a set of stimulation parameters, (ii) determining values ​​for the one or more metrics, (iii) adjusting the set of stimulation parameters based on the value, and (iv) Repeating steps (i) to (iii) until a stopping condition based on the value of the metric is reached.

3. The system according to claim 1, wherein: The one or more metrics include selected one or more clinical indicators.

4. The system according to claim 3, wherein: The selected one or more clinical indicators are selected based on their respective first rankings and second rankings.

5. The system according to claim 4, wherein: The instructions further configure the processor to assign a weight to each clinical indicator based on the first ranking and the second ranking for each clinical indicator.

6. The system according to claim 5, wherein: The one or more metrics include a mathematical combination of a plurality of the clinical indicators, each of which is normalized using its assigned weight.

7. The system according to claim 5, wherein: The one or more metrics include a linear combination of a plurality of the clinical indicators, each of which is normalized by its corresponding weight.

8. The system according to claim 1, wherein: The medical intervention includes providing a medicament to the patient.

9. The system according to claim 8, wherein: The agent is levodopa.

10. The system according to claim 1, wherein: The medical intervention includes providing electrical stimulation to the patient.

11. The system according to claim 1, wherein: The plurality of clinical indicators include one or more of speech, tremor, rigidity, finger tapping, toe tapping, bradykinesia, hypokinesia, agile posture, gate or postural stability.

12. The system of claim 1, wherein: Assigning a first ranking and a second ranking to at least one of the clinical indicators is based on measurements from one or more sensors associated with the patient.

13. The system of claim 1, wherein: The stimulation therapy includes deep brain stimulation.

14. The system of claim 2, wherein: The stimulation parameters include one or more of stimulation contact configuration, amplitude, frequency or pulse width.

Images