Systems and methods for symptom forecasting of neurological diseases and conditions

An IMD with brain monitoring and adaptive therapy delivery addresses the limitations of conventional methods by predicting neurological symptoms and adjusting interventions, enhancing patient safety and daily life management.

WO2026239194A1PCT designated stage Publication Date: 2026-11-19MEDTRONIC INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/IB2026/054307
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-14
Filing Date
2026-04-30
Publication Date
2026-11-19

AI Technical Summary

Technical Problem

Conventional methods for predicting symptom occurrence in neurological diseases, such as epilepsy and movement disorders, are limited to clinician-assisted inpatient monitoring and wearable devices with limited sensing and prediction capabilities, making it difficult to anticipate when interventions like electrical stimulation and medication are needed.

Method used

An implantable medical device (IMD) with electrodes and motion sensors is used to monitor bioelectrical and motion data from the brain, determining a risk parameter for symptom occurrence and adjusting electrical stimulation or medication accordingly to prevent or reduce symptoms.

Benefits of technology

Enables ambulatory symptom forecasting and adaptive therapy delivery, improving patient safety and reducing anxiety by providing timely interventions in daily life situations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure IB2026054307_19112026_PF_FP_ABST
    Figure IB2026054307_19112026_PF_FP_ABST
Patent Text Reader

Abstract

Systems and methods for symptom forecasting are disclosed herein. According to some embodiments, the present technology includes a method for symptom forecasting of a neurological disease or condition for a patient. The method can include receiving, via a processor, bioelectrical data indicative of bioelectrical activity in a subcortical region of the patient's brain. The method can further include receiving, via the processor, motion data indicative of patient movement. The method can also include determining, via the processor, a symptom risk parameter indicative of a likelihood of a symptom occurring for the neurological disease or condition. The symptom risk parameter may be based on the bioelectrical data and the motion data. The method can also include outputting, via the processor, the symptom risk parameter on a display. Further, the method may include adjusting stimulation parameters and / or adjusting instructions for a patient's medication protocol based on the symptom risk parameter.
Need to check novelty before this filing date? Find Prior Art

Description

MDT Ref. No. A0013188W001SYSTEMS AND METHODS FOR SYMPTOM FORECASTING OF NEUROLOGICAL DISEASES AND CONDITIONSCROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims the benefit of priority to U.S. Provisional Application No. 63 / 805,734 filed May 14, 2025, which is incorporated herein by reference in its entirety.TECHNICAL FIELD

[0002] The present technology generally relates to medical devices, and in particular, to systems and methods for symptom forecasting of neurological diseases and conditionsBACKGROUND

[0003] For patients with neurological diseases and conditions, the unpredictability of when and where a symptom (e.g., a seizure or tremor) will occur can pose a significant challenge in the patient’s day-to-day life. For example, the threat of a sudden unexpected seizure may discourage a patient from traveling or participating in recreational activities, and a sudden onset of tremors may directly impede a patient’s ability to eat or speak. This is further complicated by the sporadic nature of symptom occurrence — symptoms can happen at seemingly irregular intervals across multiple timescales (e.g., hours, days, and even weeks). Interventions such as electrical stimulation and medication are available and may be effective in providing symptom relief and / or prevention, but it is difficult to anticipate when such interventions would be helpful. Conventional techniques for predicting symptom occurrence are limited to clinician-assisted inpatient monitoring and wearable devices with limited sensing and prediction capabilities. Accordingly, there remains a need for improved symptom forecasting of neurological diseases and conditions.BRIEF DESCRIPTION OF THE DRAWINGS

[0004] Many aspects of the present disclosure can be better understood with reference to the following drawings. The components in the drawings are not necessarily to scale. Instead, emphasis is placed on illustrating clearly the principles of the present disclosure.

[0005] FIG. 1 is a conceptual diagram illustrating an example system that includes an implantable medical device (HMD) configured to deliver DBS to a patient, in accordance with the present technology.MDT Ref. No. A0013188W001

[0006] FIG. 2 is a block diagram of the example IMD of FIG. 1 for delivering DBS therapy, in accordance with embodiments of the present technology.

[0007] FIG. 3 is a block diagram of the external programmer of FIG 1 for controlling delivery of DBS therapy, in accordance with embodiments of the present technology.

[0008] FIG. 4A is a block diagram illustrating a representative example of a workflow including a model generation workflow and a model execution workflow for symptom forecasting of a neurological disease or condition for a patient, in accordance with embodiments of the present technology.

[0009] FIG. 4B illustrates an example phase-locked relationship between patient data and symptom occurrence, in accordance with embodiments of the present technology.

[0010] FIGS. 5 A and 5B are schematic illustrations of portions of a workflow for symptom forecasting for a patient having epilepsy, in accordance with embodiments of the present technology.

[0011] FIG. 6A and 6B are schematic illustrations of portions of a workflow for symptom forecasting for a patient having a movement disorder, in accordance with embodiments of the present technology.

[0012] FIG. 7 is a partially schematic diagram of an external programmer, in accordance with embodiments of the present technology.

[0013] FIG. 8 is a partially schematic diagram of an external programmer, in accordance with embodiments of the present technology.

[0014] FIG. 9 is a flow diagram illustrating a method for symptom forecasting of a neurological disease or condition for a patient, in accordance with embodiments of the present technology.DETAILED DESCRIPTION

[0015] The present technology relates to symptom forecasting for neurological diseases and conditions, such as epilepsy and movement disorders (e.g., Parkinson’s disease, dystonia, essential tremor, etc.). Some embodiments of the present technology, for example, include systems for symptom forecasting utilizing data obtained by a deep brain stimulation device. In some examples, the system comprises an implantable medical device (IMD) including a plurality of electrodes configured to be implanted in and deliver electrical stimulation energy to a subcortical and / or subthalamic region of the brain. The IMD can further include a motion sensor (e.g., an accelerometer). The system may further include one or more processors and a memory operablyMDT Ref. No. A0013188W001 coupled to the one or more processors and storing instructions that, when executed by the one or more processors, cause the system to perform operations including receiving, from at least one of the electrodes, bioelectrical data indicative of bioelectrical activity (e.g., neural activity) in a subcortical region of the patient’s brain (e.g., the anterior nucleus of the thalamus). The operations may further include receiving, from the motion sensor, motion data indicative of patient movement. The operations may further include determining a risk parameter indicative of a likelihood of an occurrence of a symptom of the neurological disease or condition, based on the bioelectrical data and the motion data. The operations may further include outputting an indication of the risk parameter on a display. Optionally, therapy may be adjusted for the patient based on the risk parameter, such as by initiating or modifying electrical stimulation provided via the at least one stimulation electrode and / or modifying a medication protocol.

[0016] Some embodiments of the present technology are directed to methods for symptom forecasting of a neurological disease or condition for a patient. A method can include receiving, via one or more processors, bioelectrical data indicative of bioelectrical activity (e.g., neural activity) in a subcortical region of the patient’s brain (e.g., the anterior nucleus of the thalamus). The method can further include receiving, via the one or more processors, motion data indicative of patient movement. The method can also include determining, via the one or more processors a risk parameter indicative of a likelihood of an occurrence of a symptom of the neurological disease or condition. The risk parameter may be based on the bioelectrical data, the motion data, or a combination thereof. The method can also include outputting, via the one or more processors, an indication of the risk parameter on a display. Optionally, therapy for the patient, such as electrical stimulation and / or medication, can be adjusted based on the risk parameter to prevent symptom occurrence and / or reduce symptom severity.

[0017] In some embodiments, the determination of the risk parameter is not based on motion data. For instance, a system for symptom forecasting of a neurological disease or condition for a patient can include an implantable medical device (IMD) including a plurality of electrodes configured to be implanted in and deliver electrical stimulation energy to a subcortical and / or subthalamic region of the brain. The system may further include one or more processors and a memory operably coupled to the one or more processors and storing instructions that, when executed by the one or more processors, cause the system to perform operations including receiving, from at least one of the electrodes, bioelectrical data indicative of bioelectrical activity (e.g., neural activity) in a subcortical region of the patient’s brain (e.g., the anterior nucleus of the thalamus). The operations may further include determining a risk parameter indicative of a likelihood of an occurrence of a symptom of the neurological disease or condition, based on theMDT Ref. No. A0013188W001 bioelectrical data. The operations may further include outputting an indication of the risk parameter on a display. Optionally, therapy may be adjusted for the patient based on the risk parameter, such as by initiating or modifying electrical stimulation provided via the at least one stimulation electrode and / or modifying a medication protocol.

[0018] The present technology can provide many advantages compared to conventional devices and techniques for monitoring and / or treating neurological diseases and conditions. For instance, electroencephalography (EEG) is commonly used for detecting ongoing or imminent symptom occurrence and determining potential sources. However, this type of invasive monitoring typically occurs in the clinic and is not accompanied by concurrent therapy. Less invasive means, such as wearable devices, have been tested for detecting symptom occurrence with prominent motor events (e.g., convulsions), but these devices require frequent recharging and high patient adherence and fail to capture symptom occurrence without concurrent motor events. In contrast, the present technology can be employed in ambulatory settings (e.g., while the patient is mobile), and in some embodiments, is adaptive to patient symptom risk. For instance, the present technology includes IMDs that are configured to forecast symptom occurrence and modify stimulation parameters and / or medical dosage for the patient to prevent symptom occurrence and / or reduce symptom severity. Further, the present technology may communicate the patient’s symptom risk to the patient, such as by displaying a risk parameter on a display (e.g., the patient programmer, the patient’s mobile phone, etc.). This may alleviate certain patient anxieties and assist the patient in day-to-day decision-making.I. Overview of Example Treatment Systems and Devices

[0019] FIG. 1 is a conceptual diagram illustrating an example system 100 that includes an implantable medical device (IMD) 106 configured to deliver continuous and / or adaptive deep brain stimulation (DBS) to a patient 112. DBS may be adaptive (aDBS) in the sense that the IMD 106 may adjust, increase, or decrease the value of one or more stimulation parameters that define the DBS in response to changes in patient activity or movement, a severity or frequency of one or more symptoms of a disease of the patient, a presence of one or more side effects due to the DBS, or one or more sensed signals of the patient, etc. For example, the system 100 may use one or more sensed signals of the patient as a control signal such that the IMD 106 adjusts the magnitude of the one or more parameters of the electrical stimulation in response to the magnitude or change in magnitude of the one or more sensed signals. This process enables system 100 to automatically adjust stimulation therapy in response to changes to the patient condition, such as changes to brain activity indicative of a level of therapy efficacy. Additionally, the system 100 may use one or more sensed signals of the patient to switch between different therapy programs for the IMD 106. The therapyMDT Ref. No. A0013188W001 programs may be predefined (e.g., selected by a clinician) to correspond with forecasted risk parameters for the patient.

[0020] In some embodiments, the system 100 includes medical device programmer 104, IMD 106, lead extension 110, and leads 114A and 114B with respective sets of electrodes 116, 118. In the example shown in FIG. 1, the electrodes 116, 118 of the leads 114A, 114B are positioned to deliver electrical stimulation to a tissue site within the patient’s brain 120, such as a deep brain site under the dura mater of the brain 120. In some examples, delivery of stimulation to one or more regions of the brain 120, such as the subcortical nucleus, globus pallidus or thalamus, may be an effective treatment to manage movement disorders, such as Parkinson’s disease, and / or neurological diseases or conditions, such as epilepsy. Some or all of the electrodes 116, 118 may additionally or alternatively be positioned to sense bioelectrical brain signals within the brain 120. In some examples, a first subset of the electrodes 116, 118 may be configured to sense bioelectrical brain signals and a second subset of the electrodes 116, 118 may be configured to deliver continuous and / or adaptive electrical stimulation to the brain 120. Alternatively, all of the electrodes 116, 118 may be configured to both sense bioelectrical brain signals and deliver continuous and / or adaptive electrical stimulation to the brain 120.

[0021] In some embodiments, the IMD 106 includes a therapy module (e.g., which may include processing circuitry, signal generation circuitry or other electrical circuitry configured to perform the functions attributed to the IMD 106) that includes a stimulation generator configured to generate and deliver electrical stimulation therapy to the patient 112 via a subset of the electrodes 116, 118 of the leads 114A and 114B, respectively. The subset of the electrodes 116, 118 that are used to deliver electrical stimulation to the patient 112, and, in some examples, the polarity of the subset of the electrodes 116, 118, may be referred to as a stimulation electrode combination. As described in further detail below, the stimulation electrode combination can be selected for a particular patient 112 and target tissue site (e.g., selected based on bioelectrical signal information and the patient’s condition). In some embodiments, the subset of the electrodes 116, 118 includes at least one electrode and can include a plurality of electrodes. In some examples, the electrodes 116 and / or 118 may have a complex electrode geometry such that two or more electrodes are located at different positions around the perimeter of the respective lead 114A or 114B.

[0022] In some embodiments, the system 100, via the IMD 106, delivers electrical stimulation therapy defined by one or more parameters, such as voltage or current amplitude, adjusted in response to a signal deviating from a range defined by a homeostatic window (e.g., a window defined by one or more thresholds for a brain signal, such as a lower threshold and upper threshold). The homeostatic window may be used as part of an adaptive stimulation mode forMDT Ref. No. A0013188W001 adjusting stimulation therapy over time. In some examples, the system 100 may change other parameters in response to sensed signals such as stimulation pulse frequency, pulse burst duration, pulse burst frequency, duty cycle, and / or electrode combination. Furthermore, the adjustable stimulation parameters may have parameter limits (lower and / or upper limits) to reduce the possibility of undesired stimulation being delivered during automated closed-loop control. These one or more thresholds and / or limits may be adjusted based on one or more periodicity metrics.

[0023] The system 100 can monitor one or more signals of the patient for selecting one or more parameters defining stimulation and / or adjusting stimulation in a closed-loop manner. In one example, the signal is a bioelectrical signal of a patient, such as a brain signal (e.g., local field potential or LFP) with a frequency within a Beta frequency band and / or a Gamma frequency band of the brain of the patient. For example, the monitored signal may be a power of the respective Beta frequency band and / or Gamma frequency band (determined based on which frequency varies during stimulation delivery and / or under the influence of medication). In yet a further example, the signal can be a signal indicative of a physiological parameter of the patient, such as a severity of a symptom of the patient, a movement of the patient, a posture of the patient, a respiratory function of the patient, a heart rate, or an activity level of the patient. The system 100 may use a single signal, a combination of different signals, or a composite measure generated from multiple signals for initially selecting and / or adjusting one or more parameters that define subsequent stimulation therapy. The system 100, via the IMD 106, can be configured to deliver electrical stimulation to the patient, wherein one or more parameters defining the electrical stimulation are proportional to the magnitude of the monitored signal or adjusted in response to a magnitude of the monitored signal exceeding one or more thresholds.

[0024] The system 100 may be configured to treat one or more patient conditions, such as a movement disorder, neurodegenerative impairment, a mood disorder, or a seizure disorder of the patient 112. The patient 112 ordinarily is a human patient. In some cases, however, the therapy system 100 may be applied to other mammalian or non-mammalian, non -human patients. The therapy system 100 may provide therapy to manage symptoms of movement disorders and / or other patient conditions, such as, but not limited to, seizure disorders (e.g., epilepsy) or mood (or psychological) disorders (e.g., major depressive disorder (MDD), bipolar disorder, anxiety disorders, post-traumatic stress disorder, dysthymic disorder, and obsessive-compulsive disorder (OCD)). At least some of these disorders may be manifested in one or more patient movement behaviors. As described herein, a movement disorder or other neurodegenerative impairment may include symptoms such as, for example, muscle control impairment, gait dysfunction, motion impairment or other movement problems, such as rigidity, spasticity, bradykinesia, rhythmicMDT Ref. No. A0013188W001 hyperkinesia, nonrhythmic hyperkinesia, and akinesia. In some cases, the movement disorder may be a symptom of Parkinson’ s disease. However, the movement disorder may be attributable to other patient conditions.

[0025] In some examples, the bioelectrical signals sensed within the brain 120 may reflect changes in electrical current produced by the sum of electrical potential differences across brain tissue. Examples of bioelectrical brain signals include, but are not limited to, electrical signals generated from LFPs sensed within one or more regions of the brain 120, such as an electroencephalogram (EEG) signal, or an electrocorti cogram (ECoG) signal. Local field potentials, however, may include a broader genus of electrical signals within the brain 120.

[0026] In some examples, the bioelectrical brain signals that are used to select a stimulation electrode combination may be sensed within the same region of the brain 120 as the target tissue site for the electrical stimulation. As previously indicated, these tissue sites may include tissue sites within anatomical structures such as the thalamus, subcortical nucleus or globus pallidus of the brain 120, as well as other target tissue sites. The specific target tissue sites and / or regions within the brain 120 may be selected based on the patient condition. Thus, in some examples, the electrodes used for delivering electrical stimulation may be different than the electrodes used for sensing bioelectrical brain signals. In other examples, the same electrodes may be used to deliver electrical stimulation and sense brain signals

[0027] Electrical stimulation generated by the IMD 106 may be configured to manage a variety of disorders and conditions. In some examples, the stimulation generator of the IMD 106 is configured to generate and deliver electrical stimulation pulses to the patient 112 via electrodes of a selected stimulation electrode combination. However, in other examples, the stimulation generator of the IMD 106 may be configured to generate and deliver a continuous wave signal, e.g., a sine wave or triangle wave. In either case, a stimulation generator within the IMD 106 may generate the electrical stimulation therapy for DBS according to a therapy program that is selected at that given time in therapy. In examples in which the IMD 106 delivers electrical stimulation in the form of stimulation pulses, a therapy program may include a set of therapy parameter values (e.g., stimulation parameters), such as a stimulation electrode combination for delivering stimulation to the patient 112, pulse frequency, pulse width, and a current or voltage amplitude of the pulses. As previously indicated, the electrode combination may indicate specific electrodes 116, 118 that are selected to deliver stimulation signals to tissue of the patient 112 and the respective polarities of the selected electrodes.

[0028] The IMD 106 may be implanted within a subcutaneous pocket above the clavicle, or, alternatively, on or within the cranium 122 or at any other suitable site within the patient 112.MDT Ref. No. A0013188W001 Generally, the IMD 106 is constructed of a biocompatible material that resists corrosion and degradation from bodily fluids. The IMD 106 may comprise a hermetic housing to substantially enclose components, such as a processor, therapy module, and memory.

[0029] As shown in FIG. 1, the implanted lead extension 110 is coupled to the IMD 106 via a connector 108 (also referred to as a connector block or a header of the IMD 106). In the example of FIG. 1, the lead extension traverses from the implant site of the IMD 106 and along the neck of the patient 112 to the cranium 122 of the patient 112 to access the brain 120. In the example shown in FIG. 1, the leads 114A and 114B (referred to collectively as “the leads 114”) are implanted within the right and left hemispheres, respectively, of the patient 112 in order deliver electrical stimulation to one or more regions of brain 120, which may be selected based on the patient condition or disorder controlled by the therapy system 100. The specific target tissue site and the stimulation electrodes used to deliver stimulation to the target tissue site, however, may be selected, e.g., according to the identified patient behaviors and / or other sensed patient parameters. Other implant sites for the leads 114 and the IMD 106 are contemplated. For example, the IMD 106 may be implanted on or within the cranium 122, in some examples. Alternatively or in combination, the leads 114 may be implanted within the same hemisphere, or the IMD 106 may be coupled to a single lead implanted in a single hemisphere.

[0030] Existing lead sets include axial leads carrying ring electrodes disposed at different axial positions and so-called “paddle” leads carrying planar arrays of electrodes. Selection of electrode combinations within an axial lead, a paddle lead, or among two or more different leads presents a challenge to the clinician. In some examples, more complex lead array geometries may be used.

[0031] Although the leads 114 are shown in FIG. 1 as being coupled to a common lead extension, in other examples, the leads 114 may be coupled to the IMD 106 via separate lead extensions or directly to the connector 108. The leads 114 may be positioned to deliver electrical stimulation to one or more target tissue sites within the brain 120 to manage patient symptoms associated with a neurological disease or condition of the patient 112. The leads 114 may be implanted to position the electrodes 116, 118 at desired locations of the brain 120 through respective holes in the cranium 122. The leads 114 may be placed at any location within the brain 120 such that the electrodes 116, 118 are capable of providing electrical stimulation to target tissue sites within the brain 120 during treatment. For example, the electrodes 116, 118 may be surgically implanted under the dura mater of the brain 120 or within the cerebral cortex of the brain 120 via a burr hole in the cranium 122 of the patient 112, and electrically coupled to the IMD 106 via one or more the leads 114.MDT Ref. No. A0013188W001

[0032] In the example shown in FIG. 1, the electrodes 116, 118 of the leads 114 are shown as ring electrodes. Ring electrodes may be used in aDBS applications because they are relatively simple to program and are capable of delivering an electrical field to any tissue adjacent to the electrodes 116, 118. In other examples, the electrodes 116, 118 may have different configurations. For example, in some examples, at least some of the electrodes 116, 118 of the leads 114 may have a complex electrode array geometry that is configured to produce shaped electrical fields. The complex electrode array geometry may include multiple electrodes (e.g., partial ring or segmented electrodes) around the outer perimeter of each lead 114, rather than one ring electrode. In this manner, electrical stimulation may be directed in a specific direction from the leads 114 to enhance therapy efficacy and reduce possible adverse side effects from stimulating a large volume of tissue. In some examples, a housing of the IMD 106 may include one or more stimulation and / or sensing electrodes. In alternative examples, the leads 114 may have shapes other than elongated cylinders as shown in FIG. 1. For example, the leads 114 may be paddle leads, spherical leads, bendable leads, or any other type of shape effective in treating the patient 112 and / or minimizing invasiveness of the leads 114.

[0033] In the example shown in FIG. 1, the IMD 106 includes a memory to store a plurality of therapy programs that each define a set of therapy parameter values. In some examples, the IMD 106 may select a therapy program from the memory based on various parameters, such as sensed patient parameters and the identified patient behaviors. The IMD 106 may generate electrical stimulation based on the selected therapy program to manage the patient symptoms associated with a neurological disease or condition.

[0034] The external programmer 104 can wirelessly communicate with the IMD 106 as needed to provide or retrieve therapy information. The programmer 104 may be an external computing device that the user, e.g., a clinician and / or the patient 112, may use to communicate with the IMD 106. For example, the programmer 104 may be a clinician programmer that the clinician uses to communicate with the IMD 106 and program one or more therapy programs for the IMD 106. Alternatively, the programmer 104 may be a patient programmer that allows the patient 112 to select programs and / or view and modify therapy parameters. The clinician programmer may include more programming features than the patient programmer. In other words, more complex or sensitive tasks may only be allowed by the clinician programmer to prevent an untrained patient from making undesirable changes to the IMD 106.

[0035] When the programmer 104 is configured for use by the clinician, the programmer 104 may be used to transmit initial programming information to the IMD 106. This initial information may include hardware information, such as the type of the leads 114 and theMDT Ref. No. A0013188W001 electrode arrangement, the position of the leads 114 within the brain 120, the configuration of electrode array 116, 118, initial programs defining therapy parameter values, and any other information the clinician desires to program into the IMD 106. The programmer 104 may also be capable of completing functional tests (e.g., measuring the impedance of the electrodes 116, 118 of the leads 114). In addition, or as an alternative, to the programmer 104, a different external computing device may perform any of the functionality of the programmer 104. The external computing device may be a networked device and in communication with the IMD 106 directly or via the programmer 104.

[0036] The clinician may also store therapy programs within the IMD 106 with the aid of the programmer 104. During a programming session, the system 100 may determine one or more therapy programs that may provide efficacious therapy to the patient 112 to address symptoms associated with the patient condition, and, in some cases, specific to one or more different patient states, such as a sleep state, movement state, rest state, or symptom forecast risk parameter. For example, the system 100 may select one or more stimulation electrode combination with which stimulation is delivered to brain 120. During the programming session, the system 100 may evaluate the efficacy of the specific program being evaluated based on feedback provided by the clinician, the patient 112, or based on one or more physiological parameters or forecasted risk parameter of the patient 112 (e.g., muscle activity, muscle tone, rigidity, tremor, etc.). Alternatively, identified patient behavior from video information may be used as feedback during the initial and subsequent programming sessions.

[0037] The programmer 104 may also be configured for use by the patient 112. When configured as a patient programmer, the programmer 104 may have limited functionality (compared to a clinician programmer) in order to prevent the patient 112 from altering critical functions of the IMD 106 or applications that may be detrimental to the patient 112. In this manner, the programmer 104 may only allow the patient 112 to adjust values for certain therapy parameters or set an available range of values for a particular therapy parameter. When the programmer 104 is configured for use by the patient 112 (e.g., a patient programmer), the programmer 104 may have a limited set of adjustments and / or data available to the user compared with a clinician programmer. In this manner, the patient programmer version may prevent the patient from causing detrimental changes to therapy, but allow the patient to make some adjustments to therapy as desired.

[0038] The programmer 104 may also provide an indication to the patient 112 when therapy is being delivered, when patient input has triggered a change in therapy or when the power source within the programmer 104 or the IMD 106 needs to be replaced or recharged. For example,MDT Ref. No. A0013188W001 the programmer 104 may include an alert LED, may flash a message to the patient 112 via a programmer display, generate an audible sound or somatosensory cue to confirm patient input was received, e.g., to indicate a patient state or to manually modify a therapy parameter.

[0039] The therapy system 100 may be implemented to provide chronic stimulation therapy to the patient 112 over the course of several months or years. However, the system 100 may also be employed on a trial basis to evaluate therapy before committing to full implantation. If implemented temporarily, some components of the system 100 may not be implanted within the patient 112. For example, the patient 112 may be fitted with an external medical device, such as a trial stimulator, rather than the IMD 106. The external medical device may be coupled to percutaneous leads or to implanted leads via a percutaneous extension. If the trial stimulator indicates DBS, the system 100 provides effective treatment to the patient 112, and the clinician may implant a chronic stimulator within the patient 112 for relatively long-term treatment.

[0040] Although the IMD 106 is described as delivering electrical stimulation therapy to the brain 120, the IMD 106 may be configured to direct electrical stimulation to other anatomical regions of the patient 112 in other examples. For instance, the system 100 may include an implantable drug pump in addition to, or in place of, the IMD 106. Further, an IMD may provide other electrical stimulation such as spinal cord stimulation to treat a neurological disease or condition.

[0041] According to the techniques of the disclosure, the system 100 can define a homeostatic window (e.g., one or more thresholds of an adaptive stimulation mode) and / or a therapeutic window for delivering aDBS to the patient 112. The system 100 may adaptively deliver electrical stimulation and adjust one or more parameters defining the electrical stimulation within a parameter range defined by upper and lower limits of therapeutic window based on the activity of the sensed bioelectrical signal, e.g., LFP signal, evoked resonant neural activity (ERNA), and EEG, within the homeostatic window. For example, the system 100 may adjust the one or more parameters defining the electrical stimulation in response to the sensed signal falling below the lower threshold or exceeding the upper threshold of the homeostatic window but may not adjust the one or more parameters defining the electrical stimulation such that they fall below the lower limit or exceed the upper limit of therapeutic window.

[0042] In one example, the external programmer 104 issues commands to the IMD 106, via instructions transmitted from the external programmer 104 to the IMD 106, causing the IMD 106 to deliver electrical stimulation therapy via the electrodes 116, 118 via the leads 114. As described above, in one example, therapeutic window can define an upper bound and / or a lower bound for one or more parameters defining the delivery of electrical stimulation therapy to theMDT Ref. No. A0013188W001 patient 112. In other words, the one or more bounds for therapeutic window may refer to the limits of values that the parameter defining stimulation can be adjusted. For example, the one or more parameters include a current amplitude (for a current-controlled system) or a voltage amplitude (for a voltage-controlled system), a pulse rate or frequency, and a pulse width. In examples where the electrical stimulation is delivered according to a “burst” of pulses, or a series of electrical pulses defined by an “on-time” and an “off-time,” the one or more parameters may further define one or more of a number of pulses per burst, an on-time, and an off-time. In one example, therapeutic window defines an upper bound and a lower bound for one or more parameters, such as upper and lower threshold for a current amplitude of the electrical stimulation therapy (in current-controlled systems) or upper and lower threshold of a voltage amplitude of the electrical stimulation therapy (in voltage-controlled systems). While the examples herein are typically given with respect to adjusting a voltage amplitude or a current amplitude, the techniques herein may equally be applied to a homeostatic window and a therapeutic window using other parameters, such as, e.g., pulse rate or pulse width. Example implementations of therapeutic windows are provided in further detail below.

[0043] Typically, a patient programmer 104 may not have access to adjustments to any thresholds or limits for sensing or stimulation related to aDBS. For example, the patient programmer 104 may only enable a patient to adjust a stimulation parameter value between limits set by the clinician programmer. However, in other examples, the system 100 may provide aDBS by permitting the patient 112, e.g., via the patient programmer 104, to indirectly adjust the activation, deactivation, and magnitude of the electrical stimulation by adjusting the lower and upper threshold of the homeostatic window. In one example, the patient programmer 104 may only be enabled to adjust an upper or lower threshold of the homeostatic window a small magnitude or percentage of the clinician-set value. In another example, by adjusting one or both thresholds of the homeostatic window, the patient 112 may adjust the point at which the sensed signal deviates from the homeostatic window, triggering the system 100 to adjust one or more parameters of the electrical stimulation within a parameter range defined by the lower and upper threshold of therapeutic window.

[0044] In some examples, a patient may provide feedback, e.g., via the programmer 104, to adjust one or both thresholds of the homeostatic window. For example, the programmer 104 may provide an input mechanism where the patient can provide an input indicating when therapy is no longer effective or a side effect is felt. The programmer 104 may then automatically adjust a threshold of the homeostatic window and / or a bound of therapeutic window in order to reduce the issue associated with the patient feedback. For example, the programmer 104 may re-run theMDT Ref. No. A0013188W001 threshold determination process described herein or analyze stored sensed bioelectric signals (e.g., LFP signals) and associated stimulation amplitudes to adjust one or more of the thresholds of the adaptive mode. In this manner, the programmer 104 and / or the IMD 106 may automatically adjust one or more thresholds of the homeostatic window based on one or more physiological or bioelectrical signals of the patient 112 sensed by the IMD 106. For example, in response to deviations in the signal of the patient outside of the homeostatic window, the system 100 (e.g., the IMD 106 or the programmer 104) may automatically adjust one or more parameters defining the electrical stimulation therapy delivered to the patient in a manner that is proportional to the magnitude of the sensed signal and within therapeutic window defining lower and upper thresholds for the one or more parameters. The adjustment to the one or more stimulation therapy parameters based on the deviation of the sensed signal may be proportional or inversely proportional to the magnitude of the signal.

[0045] Hence, in some examples, the system 100, via the programmer 104 or the IMD 106, may adjust one or more parameters of the electrical stimulation, such as voltage or current amplitude, within therapeutic window based on patient input that adjusts the homeostatic window, or based on one or more signals, such as sensed physiological parameters or sensed bioelectrical signals, or a combination of two or more of the above. In particular, the system 100 may adjust a parameter of the electrical stimulation, automatically in response to the sensed signal satisfying the one or more thresholds of the homeostatic window and / or in response to patient input that adjusts the homeostatic window, provided the value of the electrical stimulation parameter is constrained to remain within a range specified by the upper and lower bound of therapeutic window. This range may be considered to include the upper and lower bound themselves.

[0046] In some examples where the system 100 adjusts multiple parameters of the electrical stimulation, the system 100 may adjust at least one of a voltage amplitude or current amplitude, a stimulation frequency, a pulse width, or a selection of electrodes, and the like. In such an example, the system 100 may set an order or sequence for adjustment of the parameters (e.g., adjust voltage amplitude or current amplitude, then adjust stimulation frequency, and then adjust the selection of electrodes). Additionally or alternatively, the system 100 may randomly select a sequence of adjustments to the multiple parameters. The system 100 may adjust a value of a first parameter of the parameters of the electrical stimulation. If the signal does not exhibit a response to the adjustment of the first parameter, the system 100 may adjust a value of a second parameter of the parameters of the electrical stimulation, and so on until the signal returns to within the homeostatic window.MDT Ref. No. A0013188W001

[0047] As discussed herein, in some embodiments the patient and / or clinician may adjust one or more of the stimulation parameters based on a determined risk parameter indicative of a likelihood of an occurrence of a symptom of a neurological condition. In these and other embodiments, the system 100 may automatically adjust the parameters in response to the risk parameter determination.

[0048] To adaptively adjust a parameter that defines DBS based on a bioelectrical signal, for example, two or more the electrodes 116, 118 of the IMD 106 may be configured to monitor a bioelectrical signal (e.g., an LFP signal) of the patient 112. In some examples, at least one of the electrodes 116, 118 may be provided on a housing of the IMD 106, providing a unipolar stimulation and / or sensing configuration. In one example, the bioelectrical signal may be selected to be a signal within a Beta frequency band of the brain 120. For example, bioelectrical signals within the Beta frequency band of the patient 112 may correlate to one or more symptoms of Parkinson’s disease in the patient 112. Generally, bioelectrical signals within the Beta frequency of the patient 112 may be approximately proportional to the severity of the symptoms of the patient 112. For example, as rigidity induced by Parkinson’s disease increases, bioelectrical signals within the Beta frequency of the patient 112 increase (e.g., magnitude of the signal and / or spectral power). Moreover, bioelectrical signals within the Beta frequency are considered proportional because the system 100 may be configured such that an increase in signal magnitude may trigger the system 100 to increase delivered stimulation therapy magnitude according to disclosed techniques. Similarly, as rigidity induced by Parkinson’s disease decreases, bioelectrical signals within the Beta frequency of the patient 112 decrease (e.g., magnitude of the signal and / or spectral power), and the decrease may trigger the system 100 to decrease the magnitude of delivered stimulation. However, in some examples, these relationships between signal changes and symptom changes may be inversed for some patients which require the system to react in an inverse manner. In some examples, one or more frequencies in the Gamma band may be responsive to stimulation for some patients. This Gamma band responsiveness may come with or without Beta suppression from stimulation therapy. Therefore, the system may determine the Gamma band frequencies and the corresponding therapy programs or adaptive mode may be used for aDBS based on the initial parameter determination process that can be automated by the system.

[0049] In some examples, the IMD 106 may include one or more other (non-electrode) sensors, such as one or more of an accelerometer, a bonded piezoelectric crystal, a mercury switch, or a gyroscope. In some examples, these sensors may provide a signal that indicates a physiological parameter of the patient, which in turn varies as a function of patient activity. For example, the device may monitor a signal that indicates the heart rate, electrocardiogram (ECG) morphology,MDT Ref. No. A0013188W001 electroencephalogram (EEG) morphology, respiration rate, respiratory volume, core temperature, subcutaneous temperature, or muscular activity of the patient.

[0050] In some examples, the one or more sensors generate a signal both as a function of patient activity and patient posture. For example, accelerometers, gyroscopes, or magnetometers may generate signals that indicate both the activity and the posture of the patient 112. The external programmer 104 may use such information regarding posture to determine whether the external programmer 104 should perform adjustments to the therapeutic window.

[0051] For example, in order to identify posture, the sensors, such as accelerometers, may be oriented substantially orthogonally with respect to each other. In addition to being oriented orthogonally with respect to each other, each of the sensors used to detect the posture of the patient 112 may be substantially aligned with an axis of the body of the patient 112. When accelerometers, for example, are aligned in this manner, the magnitude and polarity of DC components of the signals generated by the accelerometers indicate the orientation of the patient relative to the Earth’s gravity, e.g., the posture of the patient 112. Further information regarding use of orthogonally aligned accelerometers to determine patient posture may be found in a commonly assigned U.S. Patent No. 5,593,431, which issued to Todd J. Sheldon, the entire content of which is incorporated by reference herein. In some embodiments, the sensor(s) may comprise one or more accelerometers in different arrangements and orientations.

[0052] Other sensors that may generate a signal that indicates the posture of the patient 112 include electrodes that generate a signal as a function of electrical activity within muscles of the patient 112, e.g., an electromyogram (EMG) signal, or a bonded piezoelectric crystal that generates a signal as a function of contraction of muscles. Electrodes or bonded piezoelectric crystals may be implanted in the legs, buttocks, chest, abdomen, or back of the patient 112, and coupled to one or more of the external programmer 104 and the IMD 106 wirelessly or via one or more leads. Alternatively, electrodes may be integrated in a housing of the IMD 106, or piezoelectric crystals may be bonded to the housing when the IMD 106 is implanted in the buttocks, chest, abdomen, or back of the patient 112. The signals generated by such sensors when implanted in these locations may vary based on the posture of the patient 112, e.g., may vary based on whether the patient is standing, sitting, or lying down.

[0053] Further, the posture of the patient 112 may affect the thoracic impedance of the patient. Consequently, sensors may include an electrode pair, including one electrode integrated with the housing of HMDs 106 and one of the electrodes 116, 118, that generate a signal as a function of the thoracic impedance of the patient 112, and the IMD 106 may detect the posture or posture changes of the patient 112 based on the signal. In one example (not depicted), the electrodesMDT Ref. No. A0013188W001 of the pair may be located on opposite sides of the patient’s thorax. For example, the electrode pair may include electrodes located proximate to the spine of a patient for delivery of SCS therapy, and the IMD 106 with an electrode integrated in its housing may be implanted in the abdomen or chest of the patient 112. As another example, the IMD 106 may include electrodes implanted to detect thoracic impedance in addition to the leads 114 implanted within the brain of the patient 112. The posture or posture changes may affect the delivery of DBS or SCS therapy to the patient 112 for the treatment of any type of bioelectrical disorder, and may also be used to detect patient sleep, as described herein.

[0054] Additionally, changes of the posture of the patient 112 may cause pressure changes with the cerebrospinal fluid (CSF) of the patient. Consequently, sensors may include pressure sensors coupled to one or more intrathecal or intracerebroventricular catheters, or pressure sensors coupled to IMDs 106 wirelessly or via one of the leads 114. CSF pressure changes associated with posture changes may be particularly evident within the brain of the patient, e.g., may be particularly apparent in an intracranial pressure (ICP) waveform.

[0055] The architecture of the system 100 illustrated in FIG. 1 is shown as an example. The techniques as set forth in this disclosure may be implemented in the example the system 100 of FIG. 1, as well as other types of systems not described specifically herein. Nothing in this disclosure should be construed so as to limit the techniques of this disclosure to the example architecture illustrated by FIG. 1.

[0056] FIG. 2 is a block diagram of the example IMD 106 of FIG. 1 configured for delivering continuous and / or adaptive deep brain stimulation therapy. In the example shown in FIG. 2, the IMD 106 includes processing circuitry 210, memory 211, stimulation generator 202, sensing module 204, switch module 206, telemetry module 208, sensor 212, and power source 220. Each of these modules may be or include electrical circuitry configured to perform the functions attributed to each respective module. For example, the processing circuitry 210 may include one or more processors part of the processing circuitry, the switch module 206 may include switch circuitry, the sensing module 204 may include sensing circuitry, the stimulation generator 202 may include stimulation generation circuitry, and the telemetry module 208 may include telemetry circuitry. The switch module 206 may not be necessary for multiple current source and sink configurations in which each current source and sink are directly connected to each electrode, but may be connected or disconnected via a respective switch. The memory 211 may include any volatile or non-volatile media, such as a random-access memory (RAM), read only memory (ROM), non-volatile RAM (NVRAM), electrically erasable programmable ROM (EEPROM), flash memory, and the like. The memory 211 may store computer-readable instructions that, whenMDT Ref. No. A0013188W001 executed by the processing circuitry 210, cause the IMD 106 to perform various functions. The memory 211 may be a storage device or other non-transitory medium.

[0057] In the example shown in FIG. 2, the memory 211 stores therapy programs 214 and sense electrode combinations and associated stimulation electrode combinations 218 in separate memories within the memory 211 or separate areas within the memory 211. Each stored therapy program 214 defines a particular set of electrical stimulation parameters (e.g., a therapy parameter set), such as a stimulation electrode combination, electrode polarity, current or voltage amplitude, pulse width, and pulse rate. In some examples, individual therapy programs may be stored as a therapy group, which defines a set of therapy programs with which stimulation may be generated. The stimulation signals defined by therapy programs of therapy group may be delivered together on an overlapping or non-overlapping (e.g., time-interleaved) basis. The therapy programs 214 may also store adaptive stimulation parameters that define adaptive stimulation, such as one or more thresholds for a homeostatic window and / or one or more limits for a therapeutic window (e.g., parameter limits). The processing circuitry 210 may directly change and / or update any of these parameter values based on commands from the programmer 104, for example.

[0058] Sense and stimulation electrode combinations 218 stores sense electrode combinations and associated stimulation electrode combinations. As described above, in some examples, the sense and stimulation electrode combinations may include the same subset of the electrodes 116, 118, a housing of the IMD 106 functioning as an electrode, or may include different subsets or combinations of such electrodes. Thus, the memory 211 can store a plurality of sense electrode combinations and, for each sense electrode combination, store information identifying the stimulation electrode combination that is associated with the respective sense electrode combination. The associations between sense and stimulation electrode combinations can be determined, e.g., automatically by the processing circuitry 210. In some examples, corresponding sense and stimulation electrode combinations may comprise some or all of the same electrodes. In other examples, however, some or all of the electrodes in corresponding sense and stimulation electrode combinations may be different. For example, a stimulation electrode combination may include more electrodes than the corresponding sense electrode combination in order to increase the efficacy of the stimulation therapy. In some examples, as discussed above, stimulation may be delivered via a stimulation electrode combination to a tissue site that is different than the tissue site closest to the corresponding sense electrode combination but is within the same region, e.g., the thalamus, of brain 120 in order to mitigate any irregular oscillations or other irregular brain activity within the tissue site associated with the sense electrode combination.MDT Ref. No. A0013188W001

[0059] The stimulation generator 202, under the control of the processing circuitry 210, generates stimulation signals for delivery to the patient 112 via selected combinations of the electrodes 116, 118. An example range of electrical stimulation parameters believed to be effective in DBS to manage a movement disorder and / or other neurological disease or condition of a patient include:

[0060] 1. Pulse Rate, i.e., Frequency: between approximately 40 Hertz and approximately 500 Hertz, such as between approximately 40 to 185 Hertz or such as approximately 140 Hertz.

[0061] 2. In the case of a voltage-controlled system, Voltage Amplitude: between approximately 0.1 volts and approximately 50 volts, such as between approximately 2 volts and approximately 3 volts.

[0062] 3. In the alternative case of a current-controlled system, Current Amplitude: between approximately 0.2 milliamps to approximately 100 milliamps, such as between approximately 1.3 milliamps and approximately 2.0 milliamps.

[0063] 4. Pulse Width: between approximately 10 microseconds and approximately 5000 microseconds, such as between approximately 100 microseconds and approximately 1000 microseconds, or between approximately 180 microseconds and approximately 450 microseconds.

[0064] Accordingly, in some examples, the stimulation generator 202 generates electrical stimulation signals in accordance with the electrical stimulation parameters noted above, subject to application of the upper and lower threshold of a therapeutic window to one or more of the parameters, such that an applicable parameter resides within the range prescribed by the window. Other ranges of therapy parameter values may also be useful and may depend on the target stimulation site within the patient 112. While stimulation pulses are described, stimulation signals may be of any form, such as continuous-time signals (e.g., sine waves) or the like.

[0065] The processing circuitry 210 may include fixed function processing circuitry and / or programmable processing circuitry, and may comprise, for example, any one or more of a microprocessor, a controller, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), discrete logic circuitry, or any other processing circuitry configured to provide the functions attributed to the processing circuitry 210 herein may be embodied as firmware, hardware, software or any combination thereof. The processing circuitry 210 may control the stimulation generator 202 according to therapy programs 214 stored in the memory 211 to apply particular stimulation parameter values specified by one or more of programs, such as voltage amplitude or current amplitude, pulse width, or pulse rate.MDT Ref. No. A0013188W001

[0066] In the example shown in FIG. 2, the set of electrodes 116 includes electrodes 116A, 116B, 116C, and 116D, and the set of electrodes 118 includes electrodes 118A, 118B, 118C, and 118D. The processing circuitry 210 also controls the switch module 206 to apply the stimulation signals generated by the stimulation generator 202 to selected combinations of the electrodes 116, 118. In particular, the switch module 206 may couple stimulation signals to selected conductors within the leads 114, which, in turn, deliver the stimulation signals across selected the electrodes 116, 118. The switch module 206 may be a switch array, switch matrix, multiplexer, or any other type of switching module configured to selectively couple stimulation energy to selected the electrodes 116, 118 and to selectively sense bioelectrical brain signals with selected the electrodes 116, 118. Hence, the stimulation generator 202 is coupled to the electrodes 116, 118 via the switch module 206 and conductors within the leads 114. In some examples, however, the IMD 106 does not include the switch module 206.

[0067] The stimulation generator 202 may be a single channel or multi-channel stimulation generator. In particular, the stimulation generator 202 may be capable of delivering a single stimulation pulse, multiple stimulation pulses, or a continuous signal at a given time via a single electrode combination or multiple stimulation pulses at a given time via multiple electrode combinations. In some examples, however, the stimulation generator 202 and the switch module 206 may be configured to deliver multiple channels on a time-interleaved basis (e.g., pulses from one channel are at least partially alternating with at least some pulses from another channel). For example, the switch module 206 may serve to time divide the output of the stimulation generator 202 across different electrode combinations at different times to deliver multiple programs or channels of stimulation energy to the patient 112. Alternatively, the stimulation generator 202 may comprise multiple voltage or current sources and sinks that are coupled to respective electrodes to drive the electrodes as cathodes or anodes. In this example, the IMD 106 may not require the functionality of the switch module 206 for time-interleaved multiplexing of stimulation via different electrodes.

[0068] The electrodes 116, 118 on respective leads 114 may be constructed of a variety of different designs. For example, one or both of the leads 114 may include two or more electrodes at each longitudinal location along the length of the lead, such as multiple electrodes at different perimeter locations around the perimeter of the lead at each of the locations A, B, C, and D. In one example, the electrodes may be electrically coupled to the switch module 206 via respective wires that are straight or coiled within the housing the lead and run to a connector at the proximal end of the lead. In another example, each of the electrodes of the lead may be electrodes deposited on a thin film. The thin film may include an electrically conductive trace for each electrode that runsMDT Ref. No. A0013188W001 the length of the thin film to a proximal end connector. The thin film may then be wrapped (e.g., a helical wrap) around an internal member to form the lead 114. These and other constructions may be used to create a lead with a complex electrode geometry.

[0069] Although the sensing module 204 is incorporated into a common housing with the stimulation generator 202 and the processing circuitry 210 in FIG. 2, in other examples, the sensing module 204 may be in a separate housing from the IMD 106 and may communicate with the processing circuitry 210 via wired or wireless communication techniques. Example bioelectrical brain signals include, but are not limited to, a signal generated from local field potentials (LFPs) within one or more regions of the brain 20. EEG, ECoG, and evoked resident neural activity (ERNA) signals are other examples of electrical signals that may be measured within the brain 120 or by electrodes placed in other locations with respect to brain the 120.

[0070] The sensor 212 may include one or more sensing elements that sense values of a respective patient parameter. For example, the sensor 212 may include one or more accelerometers, optical sensors, chemical sensors, temperature sensors, pressure sensors, or any other types of sensors. The sensor 212 may output patient parameter values that may be used as feedback to control delivery of therapy. The IMD 106 may include additional sensors within the housing of the IMD 106 and / or coupled via one of the leads 114 or other leads. In addition, the IMD 106 may receive sensor signals wirelessly from remote sensors via the telemetry module 208, for example. In some examples, one or more of these remote sensors may be external to patient (e.g., carried on the external surface of the skin, attached to clothing, or otherwise positioned external to the patient).

[0071] The telemetry module 208 supports wireless communication between the IMD 106 and the external programmer 104 or another computing device under the control of the processing circuitry 210. The processing circuitry 210 of the IMD 106 may receive, as updates to programs, values for various stimulation parameters such as magnitude and electrode combination, from the programmer 104 via the telemetry module 208. The updates to therapy programs may be stored within the therapy programs 214 portion of the memory 211. The telemetry module 208 in the IMD 106, as well as telemetry modules in other devices and systems described herein, such as the programmer 104, may accomplish communication by radiofrequency (RF) communication techniques. In addition, the telemetry module 208 may communicate with the external programmer 104 via proximal inductive interaction of the IMD 106 with the programmer 104. Accordingly, the telemetry module 208 may send information to the external programmer 104 on a continuous basis, at periodic intervals, or upon request from the IMD 106 or the programmer 104.MDT Ref. No. A0013188W001

[0072] The power source 220 delivers operating power to various components of the IMD 106. The power source 220 may include a small rechargeable or non-rechargeable battery and a power generation circuit to produce the operating power. Recharging may be accomplished through proximal inductive interaction between an external charger and an inductive charging coil within the IMD 106. In some examples, power requirements may be small enough to allow the IMD 106 to utilize patient motion and implement a kinetic energy-scavenging device to trickle charge a rechargeable battery. In other examples, traditional batteries may be used for a limited period of time.

[0073] According to the techniques of the disclosure, the processing circuitry 210 of the IMD 106 delivers, via the electrodes 116, 118 interposed along the leads 114 (and optionally the switch module 206), electrical stimulation therapy to the patient 112. The DBS therapy can be defined by one or more therapy programs 214 having one or more parameters stored within the memory 211. The therapy programs 214 may specify the continuous and / or adaptive mode(s) of the DBS and optionally, corresponding threshold(s). For example, the one or more parameters may include a current amplitude (for a current-controlled system) or a voltage amplitude (for a voltage-controlled system), a pulse rate or frequency, and a pulse width, or quantity of pulses per cycle. The collection of one or more of these parameter values may define a parameter set that defines each therapy program. In examples where the electrical stimulation is delivered according to a “burst” of pulses, or a series of electrical pulses defined by an “on-time” and an “off-time,” the one or more parameters may further define one or more of a number of pulses per burst, an on-time, and an off-time. In one example, therapeutic window defines an upper limit and / or a lower limit for a voltage amplitude of the electrical stimulation therapy. In another example, therapeutic window defines an upper limit and / or a lower limit for a current amplitude of the electrical stimulation therapy. In particular, a parameter of the electrical stimulation therapy, such as voltage or current amplitude, is constrained to a therapeutic window having an upper limit and a lower limit, such that the voltage or current amplitude may be adjusted provided the amplitude remains greater than or equal to the lower limit and less than or equal to the upper limit. It is noted that a single limit may be used in some examples.

[0074] In one example, the processing circuitry 210, via the electrodes 116, 118 of the IMD 106, monitors the behavior of a signal of the patient 112 that correlates to one or more symptoms of a disease of the patient 112 within a homeostatic window. The processing circuitry 210, via the electrodes 116, 118, delivers DBS to the patient 112, and may adjust one or more parameters defining the electrical stimulation within a parameter range defined by lower and upper thresholds of a therapeutic window based on the activity of the sensed signal within the homeostatic window.MDT Ref. No. A0013188W001

[0075] In one example, the signal is a bioelectrical signal (e.g., a LFP signal) within the Beta frequency band of the brain 120. The signal within the Beta frequency band of the patient 112 may correlate to one or more symptoms of Parkinson’s disease in the patient 112. Generally speaking, bioelectrical signals within the Beta frequency band of the patient 112 may be approximately proportional to the severity of the symptoms of the patient 112. For example, as tremor induced by Parkinson’s disease increases, one or more of the electrodes 116, 118 detect an increase in the magnitude of bioelectrical signals within the Beta frequency band of the patient 112.

[0076] Similarly, as tremor induced by Parkinson’s disease decreases, the processing circuitry 210, via the one or more of the electrodes 116, 118, detects a decrease in the magnitude of the bioelectrical signals within the Beta frequency band of the patient 112. In another example, the signal is a bioelectrical signal within the Gamma frequency band of the brain 120. The signal within the Gamma frequency band of the patient 112 may also correlate to one or more side effects of the electrical stimulation therapy. However, in contrast to bioelectrical signals within the Beta frequency band, generally speaking, bioelectrical signals within the Gamma frequency band of the patient 112 may be approximately inversely proportional to the severity of the side effects of the electrical stimulation therapy. For example, as side effects due to electrical stimulation therapy increase, the processing circuitry 210, via one or more of the electrodes 116, 118, detects a decrease in the magnitude of the signal within the Gamma frequency band of the patient 112. Similarly, as side effects due to electrical stimulation therapy decrease, the processing circuitry 210, via the one or more of the electrodes 116, 118, detects an increase in the magnitude of the signal within the Gamma frequency band of the patient 112.

[0077] In response to detecting that the signal of the patient, e.g., a sensed bioelectrical signal, has deviated from the homeostatic window, the processing circuitry 210 dynamically adjusts the magnitude of the one or more parameters of the electrical stimulation therapy such as, e.g., pulse current amplitude or pulse voltage amplitude, to drive the signal of the patient back into the homeostatic window. For example, wherein the signal is a bioelectrical signal within the Beta frequency band of the brain 120, the processing circuitry 210, via the one or more of the electrodes 116, 118, monitors the Beta magnitude of the patient 112. Upon detecting that the Beta magnitude of the patient 112 exceeds the upper bound of the homeostatic window, the processing circuitry 210 increases a magnitude of the electrical stimulation delivered via the electrodes 116, 118 at a maximum ramp rate, e.g., determined automatically or by the clinician until the magnitude of the bioelectrical signal within the Beta band falls back to within the homeostatic window, or until the magnitude of the electrical stimulation reaches an upper limit of a therapeutic window determinedMDT Ref. No. A0013188W001 by the system 100 (FIG. 1). Similarly, upon detecting that the Beta magnitude of the patient 112 falls below the lower bound of the homeostatic window, the processing circuitry 210 decreases stimulation magnitude at a maximum ramp rate determined by the system 100 until the Beta magnitude rises back to within the homeostatic window, or until the magnitude of the electrical stimulation reaches a lower limit of a therapeutic window determined by the system 100. Upon detecting that the Beta magnitude is presently within the threshold of the homeostatic window or has returned to within the threshold of the homeostatic window, the processing circuitry 210 holds the magnitude of the electrical stimulation constant. In other examples, the processing circuitry 210 may automatically determine the ramp rate at which stimulation parameters are adjusted to cause the brain signal to fall back within the target range. The ramp rate may be selected based on prior data indicating general patient comfort or comfort or preferences of the specific patient.

[0078] In some examples, the processing circuitry 210 continuously measures the signal in real time. In other examples, the processing circuitry 210 periodically samples the signal according to a predetermined frequency or after a predetermined amount of time. In some examples, the processing circuitry 210 periodically samples the signal at a frequency of approximately 150 Hertz.

[0079] Furthermore, the processing circuitry 210 can deliver electrical stimulation therapy that is constrained by an upper limit and a lower limit of a therapeutic window. In some examples, values defining therapeutic window are stored within the memory 211 of the IMD 106. For example, in response to detecting that the brain signal has deviated from the homeostatic window, the processing circuitry 210 of the IMD 106 may adjust one or more parameters of the electrical stimulation therapy to provide responsive treatment to the patient 112. For example, in response to detecting that the signal has exceeded an upper threshold of the homeostatic window and prior to delivering the electrical stimulation therapy, the processing circuitry 210 increases an amplitude of stimulation (e.g., but not above the upper limit) in order to bring the signal back down below the upper threshold. For example, in a voltage-controlled system wherein the clinician has set the upper limit of therapeutic window to be 3 Volts, the processing circuitry 210 can increase the voltage amplitude to values no greater than 3 Volts in an attempt to decrease the brain signal below the upper threshold.

[0080] In another example, in response to detecting that the signal has fallen below a lower threshold of the homeostatic window and prior to delivering the electrical stimulation therapy, the processing circuitry 210 decreases the voltage amplitude, for example, but not lower than the magnitude of the lower limit. For example, in the above voltage-controlled system wherein the clinician has set the lower bound of therapeutic window to be 1.2 Volts, the processing circuitry 210 can decrease the voltage amplitude down to no lower than 1.2 Volts in an attempt to raise theMDT Ref. No. A0013188W001 brain signal back above the lower threshold and into the homeostatic window. Thus, the processing circuitry 210 of the IMD 106 may deliver aDBS to the patient 112 wherein the one or more parameters defining the aDBS is within therapeutic window defined by a lower and upper limit for the parameter.

[0081] In the foregoing example, the limit of therapeutic window is inclusive (i.e., the upper and lower limit are valid values for the one or more parameters). However, in other examples, the limit of therapeutic window is exclusive (i.e., the upper and lower limits are not valid values for the one or more parameters). In such an example of an exclusive therapeutic window, the processing circuitry 210 instead sets the adjustment to the one or more parameters to be the next highest valid value (in the case of an adjustment potentially exceeding the upper limit) or the next lowest valid value (in the case of an adjustment potentially exceeding the lower limit).

[0082] In another example, values defining therapeutic window are stored within a memory 311 of the external programmer 104. In this example, in response to detecting that the signal has deviated from the homeostatic window, the processing circuitry 210 of the IMD 106 transmits, via telemetry module 208, data representing the measurement of the signal to the external programmer 104. In one example, in response to detecting that the signal has exceeded an upper threshold of the homeostatic window, the processing circuitry 210 of the IMD 106 transmits, via the telemetry module 208, data representing the measurement of the signal to the external programmer 104. The external programmer 104 may determine to adjust a parameter value to reduce the signal below the upper threshold as long as the parameter value remains within the one or more limits to the parameter.

[0083] In another example, the processing circuitry 210, via the telemetry module 208 and from the external programmer 104, receives instructions to adjust one or more limits of therapeutic window. For example, such instructions may be in response to patient feedback on the efficacy of the electrical stimulation therapy, or in response to one or more sensors that have detected a signal of the patient. Such signals from sensors may include bioelectrical signals, such as a signal within the Beta frequency band or signal within the Gamma frequency band of the brain 120, or physiological parameters and measurements, such as a signal indicating one or more of a patient activity level, posture, and respiratory function. Further, such signals from sensors may indicate a lack of reduction of one or more symptoms of the patient 112, such as tremor or rigidity or the presence of side effects due to electrical stimulation therapy, such as paresthesia. In response to these instructions, the processing circuitry 210 may adjust one or more thresholds of the homeostatic window. For example, the processing circuitry 210 may adjust the magnitude of the upper threshold, the lower threshold, or shift the overall position of the homeostatic window suchMDT Ref. No. A0013188W001 that the threshold, defined by the homeostatic window, for adjustment of the one or more parameters of electrical stimulation, is itself adjusted. Thereafter, the processing circuitry 210, via electrodes 116 and 118, delivers the adjusted electrical stimulation to the patient 112.

[0084] FIG. 3 is a block diagram of the external programmer 104 of FIG. 1. Although the programmer 104 may generally be described as a hand-held device, the programmer 104 may be a larger portable device or a more stationary device. In some examples, the programmer 104 may be referred to as a tablet computing device. In addition, in other examples, the programmer 104 may be included as part of an external charging device or include the functionality of an external charging device. As illustrated in FIG. 3, the programmer 104 may include a processing circuitry 310, memory 311, user interface 302, telemetry module 308, and power source 320. The memory 311 may store instructions that, when executed by the processing circuitry 310, cause the processing circuitry 310 and the external programmer 104 to provide the functionality ascribed to the external programmer 104 throughout this disclosure. Each of these components, or modules, may include electrical circuitry that is configured to perform some or all of the functionality described herein. For example, the processing circuitry 310 may include processing circuitry configured to perform the processes discussed with respect to the processing circuitry 310.

[0085] In general, the programmer 104 comprises any suitable arrangement of hardware, alone or in combination with software and / or firmware, to perform the techniques attributed to the programmer 104, and the processing circuitry 310, the user interface 302, and the telemetry module 308 of the programmer 104. In various examples, the programmer 104 may include one or more processors, which may include fixed function processing circuitry and / or programmable processing circuitry, as formed by, for example, one or more microprocessors, DSPs, ASICs, FPGAs, or any other equivalent integrated or discrete logic circuitry, as well as any combinations of such components. The programmer 104 also, in various examples, may include a memory 311, such as RAM, ROM, PROM, EPROM, EEPROM, flash memory, a hard disk, a CD-ROM, comprising executable instructions for causing the one or more processors to perform the actions attributed to them. Moreover, although the processing circuitry 310 and the telemetry module 308 are described as separate modules, in some examples, the processing circuitry 310 and the telemetry module 308 may be functionally integrated with one another. In some examples, the processing circuitry 310 and the telemetry module 308 correspond to individual hardware units, such as ASICs, DSPs, FPGAs, or other hardware units.

[0086] The memory 311 (e.g., a storage device) may store instructions that, when executed by the processing circuitry 310, cause the processing circuitry 310 and the programmer 104 to provide the functionality ascribed to the programmer 104 throughout this disclosure. For example,MDT Ref. No. A0013188W001 the memory 311 may include instructions that cause the processing circuitry 310 to obtain a parameter set from memory, select one or more parameters for electrical stimulation or adaptive stimulation according to sensed signals, or receive user input and send a corresponding command to IMD 106, or instructions for any other functionality. In addition, the memory 311 may include a plurality of programs, where each program includes a parameter set that defines stimulation therapy.

[0087] The user interface 302 may include a button or keypad, lights, a speaker for voice commands, a display, such as a liquid crystal (LCD), light-emitting diode (LED), or organic lightemitting diode (OLED). In some examples the display may be a touch screen. The user interface 302 may be configured to display any information related to the delivery of stimulation therapy, identified patient behaviors, sensed patient parameter values, automatically selected parameters, prompts for user input regarding stimulation parameters or adaptive stimulation parameters, patient behavior criteria, or any other such information. The user interface 302 may also receive user input via the user interface 302. The input may be, for example, in the form of pressing a button on a keypad or selecting an icon from a touch screen. The user interface 302 may refer to hardware configured to present information to the user and / or receive input from the user. In some examples, the processing circuitry 310 directly controls this hardware. In other examples, the processing circuitry 310 may communicate with drive hardware that controls hardware of the user interface 302. In some examples, the user interface 302 may include display and / or interactive display configurations as described herein.

[0088] The telemetry module 308 may support wireless communication between the IMD 106 and the programmer 104 under the control of the processing circuitry 310. The telemetry module 308 may also be configured to communicate with another computing device via wireless communication techniques, or direct communication through a wired connection. In some examples, the telemetry module 308 provides wireless communication via an RF or proximal inductive medium. In some examples, the telemetry module 308 includes an antenna, which may take on a variety of forms, such as an internal or external antenna. In some examples, telemetry modules 308 may support communications with intermediate devices between the programmer 104 and the IMD 106 or other external devices.

[0089] Examples of local wireless communication techniques that may be employed to facilitate communication between the programmer 104 and the IMD 106 include RF communication according to the 802.11 or Bluetooth specification sets or other standard, inductive telemetry, or any proprietary telemetry protocols. In this manner, other external devices may be capable of communicating with the programmer 104 without needing to establish a secure wirelessMDT Ref. No. A0013188W001 connection. As described herein, the telemetry module 308 may be configured to transmit a spatial electrode movement pattern or other stimulation parameter values to the IMD 106 for delivery of stimulation therapy.

[0090] According to the techniques of the disclosure, in some examples, the processing circuitry 310 of the external programmer 104 defines the parameters of a homeostatic therapeutic window, stored in the memory 311, for delivering aDBS to the patient 112. In one example, the processing circuitry 310 of the external programmer 104, via the telemetry module 308, issues commands to the IMD 106 causing the IMD 106 to deliver electrical stimulation therapy via the electrodes 116, 118 via the leads 114.

[0091] The present disclosure illustrates various user interfaces and techniques for managing the sensing of physiological signals, such as brain signals, programming adaptive stimulation therapy, and managing electrical stimulation as described herein. The programmer 104, or another external computing device, may output the user interfaces and screens described herein. The example user interface screens may be separately presented or selectable in any order, or the programmer 104 (for example) may present each screen in order as part of one or more automated programming processes to assist the user through the programming process for setting up or adjusting continuous and / or adaptive stimulation therapy. User input may be prompted at various times, either to select parameter values or to confirm automatically selected parameter values. In some examples, the programmer 104 may perform each step automatically and present the user with fully automated and selected parameters at the end of the process. The user may confirm the parameter values or review one or more of the parameter values using each respective screen of the user interface as needed to customize the stimulation therapy, which may include adaptive stimulation therapy such as aDBS.II. Symptom Forecasting of Neurological Diseases and Conditions

[0092] For patients with neurological diseases such as epilepsy and Parkinson’s disease, motor symptoms (e.g., seizures or tremors) can occur at seemingly irregular intervals across multiple timescales (e.g., hours, days, weeks). Unfortunately, these symptoms may interfere with a patient’s daily activities and can lead to injury or even death. Further, patient anxieties and / or fears associated with the risk of such symptoms can affect the patient’s psychological state and mental health. Accordingly, it may be desirable to provide a patient and / or the patient’s clinician with an indication of the likelihood of a symptom occurring, for example via the patient programmer, a clinician portal, etc. Further, the symptom risk may be leveraged to improve treatment outcomes for the patient, such as adjustment of one or more of the stimulation parametersMDT Ref. No. A0013188W001 or the patient’s medication protocol to prevent symptom occurrence and / or reduce symptom severity of the neurological disease or condition.

[0093] FIG. 4A is a block diagram illustrating a representative example of a workflow 400 including a model generation workflow 400a and model execution workflow 400b for symptom forecasting of a neurological disease. In some embodiments, some or all of the processes described with respect to the workflow 400 are implemented as computer-readable instructions (e.g., program code) that are configured to be executed by one or more processors of a computing device (e.g., an implantable device, mobile device, personal computer, workstation, remote server, etc.). For instance, the workflow 400 can be performed using any of the systems and devices described herein, such as the system 100 and IMD 106 of FIGS. 1-3. As an example, the workflow 400 can be performed on a microprocessor and / or neural processing unit of the implantable device. Further, the processes of the workflow 400 may be distributed across multiple devices. For instance, the processes of the workflow 400 can be performed on one, some, or all of: the implanted device, a patient programmer, a remote server, and a mobile device of the patient. Further, the workflow 400 can be combined with any of the processes of the workflows and methods described herein.

[0094] The model generation workflow 400a can include receiving first bioelectrical data 402a indicative of bioelectrical activity. In some embodiments, the bioelectrical activity corresponds to sensed brain activity, such as LFP, from one or more regions of the patient’s brain, and the first bioelectrical data 402a may comprise LFP data. In some cases, the bioelectrical activity may correspond to activity with smaller and / or greater spatial and / or temporal resolutions than LFPs, and in such cases the first bioelectrical data 402a may additionally or alternatively comprise other brain activity. For instance, the first bioelectrical data 402a may additionally or alternatively comprise evoked resonant neural activity (ERNA) signals, evoked compound action potential (ECAP) signals, compound action potentials, electroencephalogram (EEG) signals, electromyograph (EMG) signals, cardiac signals, nerve signals, temperature signals, etc.

[0095] The first bioelectrical data 402a may be measured using one or more electrodes implanted in the patient’s brain. The one or more electrodes can be surface electrodes, depth electrodes, subdural grid electrodes, implanted multi -el ectrode arrays (MEAs), microwire electrodes, etc. of an IMD (e.g., the IMD 106 of the system 100 of FIGS. 1-3). In some embodiments, the one or more electrodes are part of a system configured to deliver continuous and / or adaptive DBS (e.g., the one or more electrodes can be or be similar to the electrodes 116, 118 of the system 100 of FIGS. 1-3). The first bioelectrical data 402a can be measured from one or more brain structures, such as the subcortical nucleus, a subthalamic region, a subcortical region, globus pallidus internus, ventral intermedius nucleus, anterior limb of the internal capsule, anteriorMDT Ref. No. A0013188W001 nucleus of the thalamus, hippocampus, medial thalamus, nucleus accumbens, pedunculopontine nucleus, posterior hypothalamus, subgenual cortex, ventromedial nucleus of the hypothalamus, deep cerebellar nuclei, etc.

[0096] Optionally, the model generation workflow 400a can include receiving first motion data 404a indicative of patient movement. The first motion data 404a may take the form of acceleration data, velocity data, speed data, position data, etc. In some embodiments, the first motion data 404a is indicative of patient movement in one or more directions (e.g., the X-direction, Y-direction, and / or Z-direction). The patient movement may include movement of the patient’s body such as movement of the patient’s head, face, limbs, etc. Alternatively or in combination the patient movement may be indicative of the patient’s gait, posture, etc. In some embodiments, the patient movement may correlate directly to symptom occurrence. For instance, rapid changes in the movement of the patient’s hand can correlate to a tremor, and a rapid change in head position can correlate to a seizure.

[0097] The first motion data 404a can be measured using any suitable motion sensor disclosed herein, such as an accelerometer, a gyroscope, and / or inertial monitoring unit (IMU). In some embodiments, the motion sensor is part of the same system or device as the one or more electrodes used to measure the first bioelectrical data 402a (e.g., part of a system configured to deliver continuous and / or adaptive DBS). An example of such a configuration is provided in connection with the system 100 of FIGS. 1-3. In some embodiments, the motion sensor is disposed within the IMD 106. Additionally or alternatively, the motion sensor may be disposed along one or both leads 114. Additionally or alternatively, the system may be in communication with one or more motions sensors associated and / or carried by one or more wearable devices. Said motion sensors may be disposed on the patient’s head, limbs, torso, legs, or elsewhere on or in the patient’s body.

[0098] The first bioelectrical data 402a may be measured continuously or at predetermined intervals (e.g., every hour for a period of 20 minutes, every four hours for a period of 30 minutes, every day for an hour, etc.) while the patient is outside of the clinical setting and performing the activities of normal living. Likewise, the first motion data 404a may be measured continuously or at predetermined intervals (e.g., every hour for a period of 20 minutes, every four hours for a period of 30 minutes, every day for an hour, etc.) while the patient is outside of the clinical setting and performing the activities of normal living. The length of the measurement period may be adjusted for the neurological disease or condition. For example, every four hours for a period of 30 minutes in a patient with epilepsy or every hour for a period of 20 minutes in a patient with Parkinson’s disease. In some examples, such as patients with Parkinson’s disease, the patient may take aMDT Ref. No. A0013188W001 medication and / or receive electrical stimulation therapy during the period. Alternatively, the patient may be off medication and / or electrical stimulation therapy during the period, e.g., during a first period, and the patient may be on medication and / or electrical stimulation therapy during a second period different than the first period.

[0099] The model generation workflow 400a can also include receiving symptom data 405. In some embodiments, the symptom data 405 includes a historical record of symptom occurrence (e.g. when and where symptoms occurred). The historical record of symptom occurrence can be derived from patient diaries, clinical assessments, seizure logs, etc. The symptom data 405 may be measured concurrently with the first bioelectrical data 402a and the first motion data 404a. Alternatively, the symptom data 405 may be reconstructed based on patient data after the first bioelectrical data 402a and the first motion data 404a were recorded. Further, in some embodiments, the symptom data 405 may be derived from the first bioelectrical data 402a and / or the first motion data 404a, such as via an event detection algorithm. For instance, an event detection algorithm can identify symptoms from various indicators in the first bioelectrical data 402a and / or the first motion data 404a, such as sudden changes, artifacts, spiking, asymmetry, synchrony and / or desynchrony across signals, etc. in the data.

[0100] The model generation workflow 400a can further include inputting one or more of the first bioelectrical data 402a, the first motion data 404a, and the symptom data 405 into a model generator 407. The model generator 407 can be configured to identify associations and / or trends in the first bioelectrical data 402a and / or the first motion data 404a that lead up to and / or are concurrent with symptom (e.g., seizure, tremor, dyskinesia, etc.) occurrence in the symptom data 405. These associations and / or trends may be periodic, occurring at particular phases in the first bioelectrical data 402 and / or the first motion data 404a.

[0101] For example, FIG. 4B illustrates an example phase-locked relationship between patient data and symptom occurrence, in accordance with embodiments of the present technology. As shown, a graph 401 illustrates how patient data (e.g., bioelectrical data and / or motion data) fluctuates in period (x-axis), and amplitude (y-axis) over time. Further, the patient data may vary in phase, as indicated by the shading of the patient data, where similar shades correspond to similar phases. As illustrated in the graph 401, symptoms can occur at common phases throughout the patient data. For instance, a first subset of symptom events 410 (shown as boxes) may occur regularly at a first phase of the patient data (dark shading). In contrast, a second subset of symptom events 412 (shown as circles) may occur irregularly at a plurality of phases of the patient data (variable shading). This phase-locked behavior can also be represented by circular phase graph 414, where symptom occurrence can be represented by both phase (e.g., angle with respectMDT Ref. No. A0013188W001 to origin) and frequency (magnitude with respect to origin). This can illustrate which phases of the patient data symptoms frequently occur, and to what degree. For instance, symptoms appear to frequently occur at phases between “increasing” and “peak” phases of the patient data. In some embodiments, magnitude with respect to origin may represent the circular effect or phase-locking value (PLV) of symptom occurrence in the patient data. PL Vs range from 0 to 1, and strong-phase clustering may be indicated by a PLV greater than 0.4, moderate phase clustering may be indicated by a PLV between 0.2 and 0.4, and weak phase clustering may be indicated by a PLV less than 0.2.

[0102] Returning to FIG. 4A, in some embodiments, the model generator 407 is configured to generate a probabilistic model 406. Alternatively or in combination, the model generator 407 can be configured to train and / or retrain an existing probabilistic model 406, such as an untrained machine learning model. The probabilistic model 406 can be generated, trained, and / or retrained based on associations and / or trends in first bioelectrical data 402a and / or the first motion data 404a with respect to the symptom data 405.

[0103] In some embodiments, the probabilistic model 406 may be or include a statistical model. For instance, the probabilistic model 406 may include a Bayesian model (e.g., Bayesian networks), Markov models (e.g., Hidden Markov Models, Markov chains), Gaussian models (e.g., Gaussian mixture models), generalized linear models (e.g., logistic regression), latent variable models, a combination thereof, etc. Alternatively or in combination, the probabilistic model 406 may be or include a machine learning model. The machine learning model can include supervised learning models, unsupervised learning models, semi-supervised learning models, and / or reinforcement learning models. Examples of machine learning models suitable for use with the present technology include, but are not limited to: regression algorithms, instance-based algorithms, regularization algorithms, decision tree algorithms, Bayesian algorithms, clustering algorithms, association rule learning algorithms, artificial neural networks, deep learning algorithms (e.g., convolutional neural networks, recurrent neural networks, long short-term memory networks, stacked autoencoders, generative adversarial networks, capsule networks, graph neural networks, encoder-decoder networks, deep Boltzmann machines, deep belief networks, transformers, vision transformers, language models), principal component analysis (PCA), singular value decomposition (SVD), other dimensionality reduction algorithms, time series forecasting algorithms, and ensemble algorithms.

[0104] In some variations, the probabilistic model 406 can be implemented using a traintest split, k-fold cross-validation, and / or other forms of data partitioning. For example, the probabilistic model 406 may be trained using a subset of the patient data (“training set”), such thatMDT Ref. No. A0013188W001 the probabilistic model 406 learns associations between bioelectrical data and motion data with symptom occurrence. Thereafter, the probabilistic model 406 can be tested on a second subset of the patient data (“test set”). For instance, predicted symptom risk can be compared to empirical symptom occurrences (e.g., to evaluate whether a symptom actually occurred during a predicted high symptom risk), and parameters of the probabilistic model 406 can be tuned based on calculated loss (e.g., errors) or overall prediction accuracy. These parameters can be tuned until satisfactory. Optionally, the probabilistic model 406 can be validated using a third subset of the patient data (“validation set”), with similar implementations as the test set.

[0105] Further, while the model generation workflow 400a has been described primarily with respect to receiving the patient’s first bioelectrical data 402a, first motion data 404a, and symptom data 405, other variations are possible. For instance, the model generation workflow 400a can include receiving bioelectrical data, motion data, and / or symptom data of a plurality of patients (e.g., including patients other than the patient). The model generator 407 may be configured to learn associations between the bioelectrical data, motion data, and symptom data of the plurality of patients to generate, train, and / or retrain a probabilistic model 406. In some embodiments, the probabilistic model 406 may be generalizable across multiple patients. In some embodiments, the bioelectrical data, motion data, and symptom data includes data selected based on one or more patient attributes including, but not limited to, age, sex, height, weight, BMI, lifestyle, or current or previous medical conditions.

[0106] Once the probabilistic model 406 is generated (e.g., constructed and / or trained), the probabilistic model 406 can be used to provide predictions of the patient’s symptom risk. For instance, the model execution workflow 400b can include receiving one or more of the second bioelectrical data 402b and the second motion data 404b (e.g., from the IMD), measured from a time period later than that corresponding to the first bioelectrical data 402a and / or the first motion data 404a. The model execution workflow 400b can include inputting one or more of the second bioelectrical data 402b and the second motion data 404b into the probabilistic model 406. As noted previously, the probabilistic model 406 can be generated, trained, and / or retrained based on the first bioelectrical data 402a, the first motion data 404a, and / or the symptom data 405. Based on the second bioelectrical data 402b and / or the second motion data 404b, the probabilistic model 406 can output a risk parameter 408 indicative of a likelihood of an occurrence of a symptom of the neurological disease or condition. The risk parameter 408 may be provided in any suitable format, including a quantitative metric (e.g., a numerical score, ordinal score, percentage, probability, range, etc.), a qualitative metric (e.g., a rating, categorization, class, description, etc.), or a combination thereof.MDT Ref. No. A0013188W001

[0107] According to some examples, the risk parameter 408 is selected from multiple predetermined risk classifications. For instance, in some cases the risk parameter 408 may be selected from one of three risk classifications, such as “high risk” (indicating a high likelihood that a symptom will occur), “moderate risk” (indicating a moderate risk that a symptom will occur), and “low risk” (indicating a low risk that s symptom may occur). Other numbers or types of classifications are possible. For example, some systems may have only two risk classifications, such as “high risk” and “low risk,” and some systems may have three or more risk classifications. As but one example, a system comprising five risk classifications may include “very high risk,” “high risk,” “moderate risk,” “low risk,” and “very low risk.” In some embodiments, the risk parameter 408 may be a numeric value within a range from 0 to 1, where 0 is low or no risk of symptom occurrence and l is a high or highest risk of symptom occurrence. Many suitable ranges may be used to capture the variation in the risk parameter 408. For example, the risk parameter 408 may be within a range from 1 to 5, 1 to 10, 0 to 10, 0 to 50, 0 to 100, etc. Further, the risk parameter 408 may be provided representationally, such as via one or more symbols (e.g., “!” as a low risk and “!!!” as high risk), colors (e.g., light green as low risk and dark red as high risk), gradients, shapes, etc. In some embodiments, the risk parameter 408 is provided graphically (e.g., a histogram, bar plot, legend). According to certain examples, the probabilistic model 406 determines a symptom risk on a probabilistic scale (e.g., numerical likelihoods) which the system converts into a risk parameter 408 on an ordinal categorical scale (e.g., low, moderate, high).

[0108] Once the risk parameter 408 is determined, the risk parameter 408 can be output on a display. For instance, the risk parameter 408 can be output on a display of a patient programmer, a clinician programmer, a clinician portal, the patient’s mobile device, a display connected to a clinician’s workstation, etc. In some cases, the risk parameter 408 may be displayed alongside historical records of symptom occurrence, as will be further described in connection with FIGS. 8 and 9. Alternatively or in combination, the risk parameter 408 can be outputted audibly and / or haptically such that the patient and / or the patient’s clinician is made aware of the patient’s current risk parameter 408.

[0109] According to some embodiments, the system may not display the risk parameter 408 until the risk parameter 408 has remain unchanged for a prescribed period of time. For instance, should the system determine the risk parameter to be “high risk,” the system may require the determination to remain “high risk” for at least an hour before outputting the risk parameter 408 to the patient and / or clinician. The amount of time a predicted risk parameter 408 spends within a new classification (e.g., “high risk”) prior to triggering a patient and / or clinician alert may be adjustable by the clinician. For example, a “high risk” for seizure state in a patientMDT Ref. No. A0013188W001 with epilepsy may require a duration of at least 3 hours within the classification zone to trigger an alert. In a second example, a “high risk” for dyskinesia state in a patient with Parkinson’s disease may require a duration of at least 30 minutes within the classification zone to trigger an alert. This may prevent rapid fluctuations that may confuse the patient. Optionally, the probabilistic model 406 may predict the risk parameter 408 at regular intervals, such as every 30 minutes, 1 hour, 2 hours, 3 hours, 12 hours, 24 hours, etc. and / or provide risk parameter 408 status updates at the respective interval. This may allow the probabilistic model 406 to be highly sensitive to rapid fluctuations in patient states (e.g., sudden changes in activity).

[0110] In some examples, the workflow 400 may optionally include modifying the patient’s treatment based on the determined risk parameter. Treatment modifications can include adjusting one or more stimulation parameters and / or adjusting instructions for a patient’s medication protocol. The degree and manner in which the stimulation parameter(s) and / or medication protocol are modified may depend on the patient’s medical history, current state, type of neurological condition or disease, etc. and / or may be determined by the clinician. Moreover, the stimulation parameters may be updated automatically in response to the risk parameter determination, or the patient and / or clinician may manually adjust one or more stimulation parameters based on the risk parameter determination. In some embodiments, the system may recommend certain adjustments when outputting the risk parameter. As but one example of a therapy adjustment, a certain risk parameter determination may affect which DBS therapy group the patient is assigned to. A first therapy group, for example, can have stimulation parameters suitable for a “high risk” of symptom occurrence (e.g., 5 mA current, 60 ps pulse, and 140 Hz frequency), a second therapy group can have stimulation parameters suitable for a “moderate risk” of symptom occurrence (e.g., 3 mA current, 60 ps pulse, and 140 Hz frequency), and a third therapy group can have stimulation parameters suitable for a “low risk” of symptom occurrence (e.g., 2.5 mA current, 60 ps pulse, and 140 Hz frequency). Many suitable variations are possible, and a plurality of parameters may be modified at the same time, e.g., as described elsewhere herein.

[0111] In addition to or instead of adjustments to the stimulation parameters, modifying treatment in response to a risk parameter determination may include modifying a medication protocol for the patient. Modifying the medication protocol can include increasing the medication dosage, decreasing the medication dosage, changing a frequency at which the medication is taken, starting a new medication and / or stopping a current medication. For instance, a dosage of an antiepileptic drug (AED), antidepressant, pain reliever, dopamine precursors, or other medications can be increased or decreased based on the patient’s risk parameter 408.MDT Ref. No. A0013188W001

[0112] According to some examples, the workflow 400 may include alerting one or more of the patient or the patient’s clinician based on a determination that the risk parameter is greater than a predetermined threshold and / or is otherwise considered “high risk.” Alerting one or more of the patient or the patient’s clinician can include transmitting a digital alert (e.g., text, notification, alarm). The alert may be transmitted automatically, informing the patient and / or the patient’s clinician of an elevated symptom risk. The alert may be accompanied by instructions, such as instructions to discontinue exercise or pull over to the side of the road if the patient is driving. Optionally, the alert may be transmitted to nearby emergency services and / or other relevant personnel.

[0113] FIGS. 5 A and 5B are schematic illustration of portions of a workflow for symptom forecasting for a patient having an implanted deep brain stimulator (such as the system 100 of FIGS. 1-3) and suffering from epilepsy. Specifically, FIG. 5A illustrates a model generation workflow 500a, and FIG. 5B illustrates a model execution workflow 500b. Some or all of the processes described with respect to the workflows 500a, 500b are implemented as computer-readable instructions (e.g., program code) that are configured to be executed by one or more processors of a computing device (e.g., an implantable device, patient programmer, mobile device, personal computer, workstation, remote server). The workflows 500a, 500b can be performed using any of the systems and devices described herein, such as the system 100 of FIGS. 1-3. As an example, the workflows 500a, 500b can be performed on a microprocessor and / or neural processing unit of the implantable device. Further, the processes of the workflows 500a, 500b may be distributed across multiple devices. For instance, the processes of the workflows 500a, 500b can be performed on one, some, or all of: the implanted device, a patient programmer, a remote server, and a mobile device of the patient. Further, the workflows 500a, 500b can be combined with any of the processes of the methods and workflows described herein.

[0114] Referring first to FIG. 5 A, the workflow 500a includes collecting first patient data 501a over a given time period, where the first patient data 501a comprises first bioelectrical data 502a (e.g., LFP data) indicative of bioelectrical activity of a subcortical region of the brain (as measured by the electrodes of the IMD) and optionally, first motion data 504a (e.g., accelerometer data) indicative of patient movement. The first patient data 501a can further include patient- and / or clinician-recorded symptom data comprising symptom events 505 (indicated by “X” marks superimposed on the first bioelectrical data 502a) during the given time period. Each symptom event 505 may include symptom data such as a date and time of the onset of the symptom, a duration of the symptom, a location of the patient when the symptom occurred, a degree of theMDT Ref. No. A0013188W001 symptom (e.g., mild, moderate, severe, etc.), and / or other clinical information pertinent to the time period during which the patient experienced the symptom (e.g., “accompanied by confusion”).

[0115] One or more of the first LFP data 502a, first accelerometer data 504a, and symptom data (e.g., symptom events 505) can be inputted into a model generator 507. The model generator 507 can be configured to generate, train, and / or retrain a probabilistic model 506 for determining a likelihood of a symptom occurring (e.g., a risk parameter). The probabilistic model 506 can be the same or generally similar to the probabilistic models discussed with reference to the workflow 400 of FIG. 4A. For instance, the probabilistic model 506 can be a statistical model, a machine learning model, or a combination thereof.

[0116] The probabilistic model 506 may be based on and / or learn phase-locked associations between the first patient data 501a and the symptom data. For instance, the probabilistic model 506 may process and analyze the first patient data 501a for temporal associations between the first LFP data 502a, first accelerometer data 504a, and the symptom events 505. These associations may be periodic in nature, where at least some of the symptom events 505 have occurred at particular phases in the first LFP data 502a and / or the first accelerometer data 504a, e.g., as described above in connection with FIG. 4B.

[0117] Once the probabilistic model 506 is generated (e.g., constructed and / or trained), the probabilistic model 506 can be used to provide predictions of the patient’s symptom risk. For instance, the model execution workflow 500b can include receiving second bioelectrical data 502b, second motion data 504b (e.g., from the IMD), and / or other sensed data, measured from a time period later than that corresponding to the first bioelectrical data 502a and the first motion data 504a. The model execution workflow 500b can include inputting one or more of the second bioelectrical data 502a and the second motion data 504b into the probabilistic model 506. As noted previously, the probabilistic model 506 can be generated, trained, and / or retrained based on the first bioelectrical data 502a, the first motion data 504b, and / or the symptom data. Based on the second bioelectrical data 502a, the second motion data 504b, and / or other sensed data, the probabilistic model 506 can output a risk parameter 508 indicative of a likelihood of an occurrence of a symptom of the neurological disease or condition. In some embodiments, the system does not sense motion data 504b and / or the probabilistic model 506 does not receive motion data 504b, and the risk parameter 508 is not based on motion data 504b. The risk parameter 508 may be provided in any suitable format, including a quantitative metric (e.g., a numerical score, ordinal score, percentage, probability, range, etc.), a qualitative metric (e.g., a rating, categorization, class, description, etc.), or a combination thereof.MDT Ref. No. A0013188W001

[0118] Based on the risk parameter 508, interventions such as modifying stimulation parameters and / or medication protocol may be initiated, e.g., as described elsewhere herein. In the illustrated example, the patient has a risk parameter 508 that is categorized as “high.” Accordingly, one or more stimulation parameters for the patient may be adjusted, such as by increasing stimulation amplitude. Alternatively or additionally, a medication protocol for the patient can be adjusted, such as by increasing a medication dosage.

[0119] FIGS. 6 A and 6B are schematic illustrations of portions of a workflow for symptom forecasting for a patient having an implanted deep brain stimulator (such as system 100) and suffering from a movement disorder, such as Parkinson’s disease, essential tremor, etc. Specifically, FIG. 6A illustrates a model generation workflow 600a, and FIG. 6B illustrates a model execution workflow 600b. In some embodiments, some or all of the processes described with respect to the workflows 600a, 600b are implemented as computer-readable instructions (e.g., program code) that are configured to be executed by one or more processors of a computing device (e.g., an implantable device, mobile device, personal computer, workstation, remote server). For instance, the workflows 600a, 600b can be performed using any of the systems and devices described herein, such as the system 100 of FIGS. 1-3. As an example, the workflows 600a, 600b can be performed on a microprocessor and / or neural processing unit of the implantable device. Further, the processes of the workflows 600a, 600b may be distributed across multiple devices. For instance, the processes of the workflows 600a, 600b can be performed on one, some, or all of: the implanted device, a patient programmer, a remote server, and a mobile device of the patient. Further, the workflows 600a, 600b can be combined with any of the processes of the methods and workflows described herein.

[0120] Referring first to FIG. 6A, the workflow 600a includes collecting first patient data 601a over a given time period, where the first patient data 601a comprises first bioelectrical data 602a (e.g., LFP data) indicative of bioelectrical activity of a subcortical region of the brain (as measured by the electrodes of the IMD) and optionally, first motion data 604a (e.g., accelerometer data) indicative of patient movement.

[0121] The patient data 601a can further include patient- and / or clinician-recorded symptom data comprising symptom trends of the patient’s motor symptoms. For instance, trends in the severity of the motor symptoms can be recorded over time (“symptom severity 616”), where a low symptom severity may correspond to low or zero motor impairment and a high symptom severity may correspond to substantial or high motor impairment. In some embodiments, the symptom severity 616 is mapped continuously, e.g., the patient, clinician, and / or another party provides a real-time account or estimate of the patient’s motor impairment. Alternatively, theMDT Ref. No. A0013188W001 symptom severity 616 may be based on (e.g., reconstructed and / or interpolated from) a series of symptom events. Each symptom event may include symptom data such as a date and time of the onset of the symptom, a duration of the symptom, a location of the patient when the symptom occurred, a degree of the symptom (e.g., mild, moderate, severe, etc.), and / or other clinical information pertinent to the time period during which the patient experienced the symptom (e.g., “accompanied by sudden pain”). In some examples, symptom events can be connected via projected trendlines.

[0122] In the illustrated example, the symptom severity 616 corresponds to symptom severity of dyskinesia. Dyskinesia can present as involuntary, repetitive, and / or erratic movements. In some cases, symptoms of dyskinesia can be reduced using electrical stimulation. As shown in the patient data 601a, electrical stimulation is turned off or reduced for the patient (shading corresponding to “OFF”) when the symptom severity 708 decreases beyond a predetermined threshold (e.g., as detected by the motion and / or sensor data) and / or when the symptom risk parameter decreases, as will be described further herein. Conversely, electrical stimulation may be turned back on or increased for the patient when the symptom severity 708 increases, and / or when the symptom risk parameter increases.

[0123] One or more of the first LFP data 602a, first accelerometer data 604a, and symptom data (e.g., symptom severity 616) can be inputted into a model generator 607. The model generator 607 can be configured to generate, train, and / or retrain a probabilistic model 606 for determining a likelihood of a symptom occurring (e.g., a risk parameter). The probabilistic model 606 can be the same or generally similar to the probabilistic models discussed with reference to the workflow 400 of FIG. 4A. For instance, the probabilistic model 606 can be a statistical model, a machine learning model, or a combination thereof.

[0124] The probabilistic model 606 may be based on and / or learn phase-locked associations between the first patient data 601a and the symptom data. For instance, the probabilistic model 606 may process and analyze the first patient data 601a for temporal associations between the first LFP data 602a, first accelerometer data 604a, and symptom severity 616. These associations may be periodic in nature, where fluctuations in the symptom severity 616 have occurred at particular phases in the first LFP data 602a and / or the first accelerometer data 604a, e.g., as described above in connection with FIG. 4B.

[0125] Once the probabilistic model 606 is generated (e.g., constructed and / or trained), the probabilistic model 606 can be used to provide predictions of the patient’s symptom risk. For instance, the model execution workflow 600b can include receiving second bioelectrical data 602b, second motion data 604b (e.g., from the IMD), and / or other sensed data, measured from a timeMDT Ref. No. A0013188W001 period later than that corresponding to the first bioelectrical data 602a and the first motion data 604a. The model execution workflow 600b can include inputting the second bioelectrical data 602a, the second motion data 604b, and / or the other sensed data into the probabilistic model 606. As noted previously, the probabilistic model 606 can be generated, trained, and / or retrained based on the first bioelectrical data 602a, the first motion data 604b, and / or the symptom data. Based on the second bioelectrical data 602a, the second motion data 604b, and / or other sensed data, the probabilistic model 606 can output a risk parameter 608 indicative of a likelihood of an occurrence of a symptom of the neurological disease or condition. In some embodiments, the system does not sense motion data 604b and / or the probabilistic model 606 does not receive motion data 604b, and the risk parameter 608 is not based on motion data 604b. The risk parameter 608 may be provided in any suitable format, including a quantitative metric (e.g., a numerical score, ordinal score, percentage, probability, range, etc.), a qualitative metric (e.g., a rating, categorization, class, description, etc.), or a combination thereof.

[0126] Based on the risk parameter 608, interventions such as modifying stimulation parameters and / or medication protocol may be initiated, e.g., as described elsewhere herein. In the illustrated example, the patient has a risk parameter 608 that is categorized as “high.” Accordingly, one or more stimulation parameters for the patient may be adjusted, such as by increasing stimulation amplitude. Alternatively or additionally, a medication protocol for the patient can be adjusted, such as by increasing a medication dosage.

[0127] In certain embodiments, the present technology provides for communication of the patient’s likelihood of experiencing a symptom — such as "high," "moderate," or "low" — via a display or other user interface accessible to the patient (e.g., a mobile phone, a patient programmer, etc.). By having this information, the patient can take proactive measures (such as adjusting their stimulation parameters and / or medication protocol), as well as make informed decisions about their immediate activities and surroundings. For instance, upon receiving a “high” risk classification, an epilepsy patient may choose to stop driving, sit or lie down in a safe location, notify a caregiver, or take other safety precautions such as wearing a protective helmet. Conversely, awareness of a “low” seizure risk may provide the patient with greater confidence to engage in daily activities, such as exercising, working, or socializing, with reduced concern for imminent seizure onset. This real-time insight into seizure likelihood not only enhances safety but also supports improved autonomy, quality of life, and adherence to personalized therapy plans.

[0128] FIG. 7 is a partially schematic diagram of a computing device (in this case, a patient programmer 700 or “programmer 700”) displaying one or more indications of a patient’s risk parameter, in accordance with embodiments of the present technology. The programmer 700 canMDT Ref. No. A0013188W001 be the same or generally similar to the programmer 104 of the system 100 of FIGS. 1-3. For example, the programmer 700 may include a user interface 702 (similar to user interface 302), processing circuitry, a memory, a telemetry module, a power source, and optionally a speaker. The programmer 700 is configured to wirelessly communicate with a medical device implanted in a patient (e.g., the IMD 106 of the system 100 of FIGS. 1-3), for example to provide instructions for electrical stimulation therapy via the IMD and to receive sensed bioelectrical, motion, and / or other data from the IMD. Note that in some examples the programmer 700 may be a smartphone or other personal electronic device operable by the patient.

[0129] Similar to the user interface 302 of programmer 104, the user interface 702 may be configured to accept inputs and commands from a user, such as a patient or caregiver. In some embodiments, for example as shown in FIG. 7, the user interface 702 comprises a graphical user interface (GUI) 703 comprising a touch sensitive display screen with input capability. Depicted on the GUI of FIG. 7 is one example of a home screen with multiple tiles, including a “Current Risk Forecast” tile 704, a “Diary” tile 706, an “Assessment” tile 710, a “Historical Trends” tile 712, a calendar display 714 of recent risk parameter determinations, and a “Review” tile 716. More or fewer tiles are possible. One or more of the tiles can comprise selectable buttons that enable the user to navigate to other screens and / or functionalities supported by the user interface 702 and / or programmer 700. In some embodiments, the one or more selectable buttons may include the “Diary” tile 706, “Assessment” tile 710, “Historical Trends” tile 712, and “Review” tile 716. The programmer 700 may switch to the appropriate screen in response to user selection of the respective selectable button.

[0130] The Current Risk Forecast tile 704 displays the current risk parameter 708. In some embodiments, the current risk forecast tile 704 is user selectable, and selection of the current risk forecast tile 704 opens a new display screen showing additional information related to the risk parameter determination, such as recent bioelectrical activity and motion data. While in FIG. 7 the risk parameter 708 is a classification (“LOW”) and displayed as text, in these and other embodiments the risk parameter 708 may be displayed in any suitable format, such as a quantitative metric (e.g., a numerical score), a qualitative metric (e.g., a categorization, which may be displayed as text, a symbol, and / or a color), or a combination thereof (e.g., higher numerical scores corresponding to higher risk are displayed in red font while lower numerical scores corresponding to lower risk are displayed in green font, etc.).

[0131] In response to user selection of the “Diary” tile 706, the user interface 702 may switch from the home screen to a diary entry screen (not shown) that prompts the user to enter in symptom data. The symptom data may include prompts for symptom time (e.g., a start time and / orMDT Ref. No. A0013188W001 end time), duration, type, location, degree (e.g., severity), and / or other relevant information (e.g., related clinical information). In some embodiments, the diary entry screen is accompanied by a historical log of diary entries. For instance, the user may be able to compare a current or pending diary entry with historical diary entries. Further, the user may be able to duplicate, modify, and / or delete previous entries. For instance, where the patient has experienced two symptom events of the same nature, the patient may duplicate a previous entry and modify the time without having to complete a full new entry. In some embodiments, the diary entries may be stored locally and / or uploaded to a server and / or remote device, and the diary entries may be used as training data for symptom forecasting models for the patient.

[0132] In response to user selection of the “Assessment” tile 710, the user interface 702 may switch from the home screen to an assessment screen (not shown), where the assessment screen prompts the user to take a cognitive assessment. In some embodiments, the cognitive assessment is configured to determine whether the patient’s cognitive abilities are currently impaired with respect to a baseline cognitive assessment of the patient. For instance, the cognitive assessment may be a quick survey evaluating the patient’s ability to complete a cognitive task such as a memorization task. In some embodiments, the cognitive assessment is a neuropsychological battery test. Other assessments that are possible include mood evaluations, pain evaluations, and / or therapy feedback (e.g., whether stimulation has been effective and / or whether side effects have occurred). Results of the cognitive assessment (e.g., results of a patient cognitive evaluation) can be stored locally and / or uploaded to a server and / or remote device, and the determination of a symptom risk parameter for the patient can be based on the results.

[0133] In response to user selection of the “Historical Trends” tile 712, the user interface 702 may switch from the home screen to an analytics screen, where the analytics screen provides one or more graphical representations of the patient’s symptom history, treatment history, etc. The analytics screen may include past symptom risk parameter predictions with corresponding historical symptom data.

[0134] As shown in FIG. 7, the home screen may further include a calendar 714 displaying the patient’s past risk parameter determinations (e.g., risk scores). The calendar 714 may display the days of the present month or days of the present week. (The calendar 714 depicted in FIG. 7 is an abbreviated monthly calendar for each of illustration.) In any case, each day may include a risk parameter indicator corresponding to the risk parameter category that occurred for the greatest percentage of the day. In some embodiments, for days during which the risk parameter was “high risk” for the greatest percentage of the day, the day may be represented by a red square. For days during which the risk parameter was “moderate risk” for the greatest percentage of the day, the dayMDT Ref. No. A0013188W001 may be represented by a yellow square. For days during which the risk parameter was “low risk” for the greatest percentage of the day, the day may be represented by a green square. Other indications and / or representations of the determined risk parameter are possible. In some embodiments, the current day is indicated on the calendar, for example, via a marking surrounding the present day. In some examples, the individual days comprise user selectable buttons that, upon selection, provide a breakdown of risk scores for a given day. In some cases, this may additionally involve selecting the “Review” tile 716. For instance, selecting the given day and the “Review” tile may cause the user interface 702 to switch from the home screen to a review screen, where the review screen displays the patient’s therapy settings (e.g., stimulation settings), symptom risk parameters, any events logged, and / or any other notes associated with the selected day. According to some embodiments, the calendar 714 is always present on the home screen. In other embodiments, user selection of the “Historical Trends” tile 712 or another tile (not shown) may cause the calendar 714 to be displayed on the home screen, e.g., as shown in the illustrated embodiment.

[0135] Optionally, one or more display screens supported by the user interface 702 (including but not limited to the home screen) can include additional tiles (which may or may not be user selectable) that display information related to the delivery of stimulation therapy, identified patient behaviors, sensed patient parameter values, automatically selected parameters, prompts for user input regarding stimulation parameters or adaptive stimulation parameters, patient behavior criteria, and / or other relevant information.

[0136] FIG. 8 is a partially schematic diagram of a computing device showing a clinician portal 800 for symptom forecasting, in accordance with embodiments of the present technology. The clinical portal 800 can be the same or generally similar to the programmer 104 of the system 100 of FIGS. 1-3 and the patient programmer 700 of FIG. 7. For example, the clinician portal 800 may include a user interface 802 (similar to user interface 302), processing circuitry, a memory, a telemetry module, a power source, and optionally a speaker. The clinician portal 800 is configured to wirelessly communicate with a medical device implanted in a patient (e.g., the IMD 106 of the system 100 of FIGS. 1-3), for example to provide instructions for electrical stimulation therapy via the IMD and to receive sensed bioelectrical, motion, and / or other data from the IMD.

[0137] Similar to the user interface 302 of programmer 104, the user interface 802 may be configured to accept inputs and commands from a clinician. In some embodiments, for example as shown in FIG. 8, the user interface 802 comprises a graphical user interface (GUI) 803 comprising a touch sensitive display screen with input capability. Depicted on the GUI of FIG. 8 is oneMDT Ref. No. A0013188W001 example of a home screen with multiple tiles relevant to a given patient, including a “Current Risk Forecast” tile 804, a “Historical Trends” tile 812, and a calendar display 814 of the patient’s recent risk parameter determinations. More or fewer tiles are possible. The home screen can optionally include one or more data displays that graphically depict sensed data. For example, as shown in FIG. 8, the home screen can include a circular phase graph 818 (as described herein) displaying LFP cycle trends, a graph 820 showing LFP and / or motion data for the current day, and / or a graph 822 showing LFP and / or motion data over several days (e.g., last two days, three days, four days, five days, six days, seven days, etc.). The graphs 820, 822 may optionally incorporate symptom events (denoted by star symbols). More or fewer data displays and different types of data displays are possible.

[0138] One or more of the tiles can comprise selectable buttons that enable the user to navigate to other screens and / or functionalities supported by the user interface 802 and / or clinician portal 800. In some embodiments, the one or more selectable buttons may include the “Historical Trends” tile 812. The clinician portal 800 may switch to the appropriate screen in response to user selection of the respective selectable button.

[0139] The “Current Risk Forecast” tile 804 displays the current risk parameter 808, similar to the “Current Risk Forecast” tile 704. Likewise, the “Historical Trends” tile 812 is similar to the “Historical Trends” tile 712, and the calendar 814 is similar to the calendar 714.

[0140] FIG. 9 is a flow diagram illustrating a method 900 for symptom forecasting of a neurological disease or condition for a patient, in accordance with embodiments of the present technology. In some embodiments, some or all of the processes described with respect to the method 900 are implemented as computer-readable instructions (e.g., program code) that are configured to be executed by one or more processors of a computing device (e.g., an implantable device, mobile device, personal computer, workstation, remote server). For instance, the method 900 can be performed using any of the systems and devices described herein, such as the system 100 of FIGS. 1-3. Further, the method 900 can be combined with any of the processes of the methods and workflows described herein.

[0141] The method 900 can begin at block 902 with receiving bioelectrical data indicative of bioelectrical activity in a subcortical region of a patient’s brain. The bioelectrical data can be generally similar to any of the bioelectrical data described herein, such as the second bioelectrical data 402b of the workflow 400b of FIG. 4B. For instance, the bioelectrical data may correspond to LFPs from one or more subcortical regions of the patient’s brain. In some embodiments, the bioelectrical data is measured using one or more electrodes implanted in the patient’s brain. Suitable sites for measurement include, but are not limited to, the subcortical nucleus, globusMDT Ref. No. A0013188W001 pallidus internus, ventral intermedius nucleus, anterior limb of the internal capsule, anterior nucleus of the thalamus, hippocampus, medial thalamus, nucleus accumbens, pedunculopontine nucleus, posterior hypothalamus, subgenual cortex, ventromedial nucleus of the hypothalamus deep cerebellar nuclei, etc.

[0142] The method 900 can continue at block 904 with receiving motion data indicative of patient movement. The motion data can be generally similar to any of the motion data described herein, such as the second motion data 404b of the workflow 400b of FIG. 4B. For instance, the motion data 404b can be or include accelerometer data corresponding to movement of the patient’s body. In some embodiments, the motion data 404b is measured using a motion sensor that is part of the same device as the one or more implanted electrodes. For instance, the motion sensor and the one or more implanted electrodes may be part of a DBS system or device. In some examples, the system does not measure and / or receive motion data. In such embodiments, the symptom risk parameter determination is not based on motion data.

[0143] Alternatively or additionally to receiving motion data, the method 900 can also include receiving additional data, where the additional data comprises one or more of any of the bioelectrical signals described herein, such as ECoG data, EEG data, ERNA data, temperature data, etc. These signals may be measured by any of the sensors described herein, such as temperature sensors, additional electrodes, pressure sensors, force sensors, etc.

[0144] The method 900 can continue at block 906 with determining a symptom risk parameter indicative of a likelihood of an occurrence of a symptom of the neurological disease or condition. The risk parameter may be based on the bioelectrical data, the motion data, and / or any additional data. In some embodiments, the risk parameter is determined using a probabilistic model, e.g., any of the probabilistic models described herein. For instance, the probabilistic model may be a statistical model and / or a supervised machine learning model. In some embodiments, the bioelectrical data, the motion data, and / or the additional data can be inputted into the probabilistic model, and the probabilistic model can be configured to output the symptom risk parameter. In some embodiments, the probabilistic model is generated using a model generator (e.g., the model generator 407 of the workflow 400a of FIG. 4A). In some embodiments, the probabilistic model may be based on and / or learn associations between one or more of previous bioelectrical data, previous motion data, previous additional data, and previous symptom data for the patient.

[0145] The method 900 can continue at block 908 with outputting an indication of the risk parameter to the patient and / or the patient’s clinician. The risk parameter can be provided in any suitable format, such as a quantitative metric (e.g., a numerical score, percentage, probability, range), a qualitative metric (e.g., a rating, categorization, class, description), or a combinationMDT Ref. No. A0013188W001 thereof. Moreover, the risk parameter can be provided graphically (e.g., a histogram, bar plot, legend). For example, the risk parameter may be a numeric value within a range from 0 to 1, where 0 is low or no risk of symptom occurrence and l is a high or highest risk of symptom occurrence. Alternatively, the risk parameter may be a classification, such as “low risk,” “moderate risk,” or “high risk.” Further, the risk parameter may be provided representationally, such as via one or more symbols (e.g., “!” as a low risk and “!!!” as high risk), colors (e.g., light green as low risk and dark red as high risk), gradients, shapes, etc. Optionally, the probabilistic model may determine a symptom risk on a probabilistic scale (e.g., numerical likelihoods) and the symptom risk may be converted into a risk parameter on an ordinal categorical scale (e.g., low, moderate, high).

[0146] The symptom risk parameter can be output on a display, such as via a user interface of a programmer, as illustrated in FIGS. 7 and 8. The risk parameter may be displayed alongside historical records of symptom occurrence. Alternatively or in combination, the risk parameter can be outputted audibly and / or haptically such that the patient and / or the patient’s clinician is made aware of the patient’s current risk parameter.

[0147] The method 900 can continue at block 910 with modifying stimulation provided to the patient and / or altering a medication dosage for the patient. For instance, the risk parameter can be compared to a symptom risk threshold. When the risk parameter exceeds the symptom risk threshold, intervention may be desirable. In some embodiments, modifying stimulation includes modifying one or more DBS parameters to prevent symptom occurrence or reduce symptom severity for the neurological disease or condition. This may include, for example, increasing an amplitude of stimulation. In some embodiments, altering the medication dosage includes increasing the medication dosage, decreasing the medication dosage, changing a frequency at which the medication is taken, starting a new medication and / or stopping a current medication.

[0148] In some embodiments, the method 900 and / or any of the workflows described herein, are performed on a single device. For instance, the method 900 and / or any of workflows described herein, may be performed using an implantable medical device, a patient programmer, or a remote server. However, in other embodiments, the method 900 and / or any of the workflows described herein, are performed using multiple devices. For example, model generation and / or training may be performed on a remote server, and the model may be improved (e.g., retrained, updated) based on new data collected by an implantable medical device. The implantable medical device can tailor the model to the particular patient, while the training on the remote server can incorporate data from other similar patients such that a general baseline model is provided. In some embodiments, it is advantageous to run the model on the implantable medical device directly so that warnings such as changes in the symptom risk parameter can be directly communicated (e.g.,MDT Ref. No. A0013188W001 via Bluetooth) to a patient’s mobile device, rather than via a patient programmer which may not always be carried by the patient. This may be communicated in the form of a notification on the patient’s phone, for example, or the patient may periodically check the symptom risk parameter. In some embodiments, the patient’s phone is provided with a user application that is configured to show patient data (e.g., numerically, graphically), but does not actively collect or process the data.

[0149] The method 900 illustrated in FIG. 9 can be modified in many different ways. For example, the ordering of the processes shown in FIG. 9 can be varied, some of the processes of the method 900 can be omitted, and / or the method 900 can include additional processes not shown in FIG. 9. For instance, the process of block 910 may be omitted from the method 900, or the process of block 910 may occur simultaneously or substantially simultaneously with the process of block 908. Further, the method 900 may include performing one or more signal processing operations on the received bioelectrical data and / or the received motion data. The signal processing operations may include one or more of denoising, cleaning, normalization, thresholding, filtering, downsampling, equalization, or augmentation techniques.EXAMPLES

[0150] The subject technology is illustrated, for example, according to various aspects described below, including with reference to FIGS. 1-9. Various examples of aspects of the subject technology are described as numbered examples (1, 2, 3, etc.) for convenience. These are provided as examples and do not limit the subject technology.

[0151] Example 1. A system for forecasting symptoms of a neurological disease or condition for a patient, the system comprising:a plurality of electrodes configured to be implanted in and deliver electrical stimulation to a subcortical region of the patient’s brain;a motion sensor;one or more processors operably coupled to the plurality of electrodes and the motion sensor; anda memory operably coupled to the one or more processors and storing instructions that, when executed by the one or more processors, cause the system to perform operations comprising:receiving, from at least one of the electrodes, bioelectrical data indicative of bioelectrical activity in the subcortical region of the patient’s brain, receiving, from the motion sensor, motion data indicative of patient movement,MDT Ref. No. A0013188W001 based on the bioelectrical data and the motion data, determining a risk parameter indicative of a likelihood of an occurrence of a symptom of the neurological disease or condition.

[0152] Example 2. The system of Example 1, wherein the operations further comprise outputting an indication of the risk parameter on a display.

[0153] Example 3. The system of Example 2, wherein the display is part of a mobile computing device, and wherein the display further comprises an indication of historical symptom occurrences of the neurological condition or disease for the patient.

[0154] Example 4. The system of any one of Examples 1 to 3, wherein the operations further comprise adjusting one or more parameters of the electrical stimulation delivered by the plurality of electrodes based on the determined risk parameter.

[0155] Example 5. The system of any one of Examples 1 to 4, wherein the operations further comprise adjusting a medication protocol for the patient based on the determined risk parameter.

[0156] Example 6. The system of any one of Examples 1 to 5, wherein the plurality of electrodes and the motion sensor are physically connected to the same implantable stimulation device.

[0157] Example 7. The system of any one of Examples 1 to 6, wherein determining the risk parameter comprises inputting the bioelectrical data and the motion data into a probabilistic model.

[0158] Example 8. The system of Example 7, wherein the probabilistic model is based on historical bioelectrical data and historical motion data obtained from the patient.

[0159] Example 9. The system of Example 7 or 8, wherein the probabilistic model is based on historical bioelectrical data and historical motion data obtained from a plurality of patients.

[0160] Example 10. The system of any one of Examples 7 to 9, wherein the probabilistic model is based on at least one of (i) the historical bioelectrical data and the historical motion data obtained from the patient and (ii) the historical bioelectrical data and the historical motion data obtained from a plurality of patients.

[0161] Example 11. The system of any one of Examples 7 to 10, wherein the probabilistic model comprises a supervised machine learning model.MDT Ref. No. A0013188W001

[0162] Example 12. The system of any one of Examples 1 to 11, wherein the risk parameter is determined on a probabilistic scale and converted to an ordinal categorical scale.

[0163] Example 13. The system of Example 12, wherein the risk parameter comprises a risk category.

[0164] Example 14. The system of Example 13, wherein the risk category is selected from the group consisting of: “high risk,” “moderate risk,” and “low risk.”

[0165] Example 15. The system of any one of Examples 1 to 14, wherein the bioelectrical data comprises local field potential (LFP) data.

[0166] Example 16. The system of any one of Examples 1 to 15, wherein the motion sensor is an accelerometer.

[0167] Example 17. The system of any one of Examples 1 to 16, wherein outputting an indication of the risk parameter only occurs if the risk parameter remains unchanged for a predetermined amount of time.

[0168] Example 18. The system of any one of Examples 1 to 17, wherein the neurological condition or disease comprises epilepsy or an epilepsy -related disorder and the symptom comprises a seizure.

[0169] Example 19. The system of any one of Examples 1 to 18, wherein the neurological condition or disease comprises a movement disorder comprising one or more of Parkinson’s disease, essential tremor, and dystonia and the symptom comprises a motor symptom.

[0170] Example 20. The system of any one of Examples 1 to 19, further comprising alerting one or more of the patient or a clinician of a change in the determined risk parameter.

[0171] Example 21. A system for forecasting symptoms of a neurological disease or condition for a patient, the system comprising:one or more processors; anda memory operably coupled to the one or more processors and storing instructions that, when executed by the one or more processors, cause the system to perform operations comprising:receiving, from one or more electrodes implanted at a subcortical region of the patient’s brain, bioelectrical data indicative of bioelectrical activity in the subcortical region,receiving, from a motion sensor disposed on or in the patient’s body, motion data indicative of patient movement,MDT Ref. No. A0013188W001 based on the bioelectrical data and the motion data, determining a risk parameter indicative of a likelihood of an occurrence of a symptom of the neurological disease or condition.

[0172] Example 22. A system for forecasting symptoms of a neurological disease or condition for a patient, the system comprising:a plurality of electrodes configured to be implanted in and deliver electrical stimulation to a subcortical region of the patient’s brain;one or more processors operably coupled to the plurality of electrodes; anda memory operably coupled to the one or more processors and storing instructions that, when executed by the one or more processors, cause the system to perform operations comprising:receiving, from at least one of the electrodes, bioelectrical data indicative of bioelectrical activity in the subcortical region of the patient’s brain, based on the bioelectrical data, determining a risk parameter indicative of a likelihood of an occurrence of a symptom of the neurological disease or condition.

[0173] Example 23. A method for forecasting symptoms of a neurological disease or condition, the method comprising:receiving, via one or more processors, bioelectrical data indicative of bioelectrical activity in a subcortical region of a patient’s brain;receiving, via the one or more processors, motion data indicative of patient movement; determining, via the one or more processors, a risk parameter indicative of a likelihood of an occurrence of a symptom of the neurological disease or condition, based on the bioelectrical data and the motion data; andoutputting, via the one or more processors, an indication of the risk parameter on a display.

[0174] Example 24. The method of Example 23, further comprising adjusting one or more parameters of electrical stimulation delivered to the patient’s brain based on the determined risk parameter.

[0175] Example 25. The method of Example 23 or 24, further comprising modifying a medication protocol for the patient based on the determined risk parameter.

[0176] Example 26. The method of any one of Examples 23 to 25, wherein the risk parameter is a first risk parameter determined at a first time and the method further comprisesMDT Ref. No. A0013188W001 determining, via the one or more processors, a second risk parameter at a second time after the first time, the second risk parameter different than the first risk parameter.

[0177] Example 27. The method of any one of Examples 23 to 26, wherein determining the risk parameter comprises inputting the bioelectrical data and the motion data into a probabilistic model.

[0178] Example 28. The method of Example 27, wherein the probabilistic model is based on historical bioelectrical data and historical motion data obtained from the patient.

[0179] Example 29. The method of Example 27 or 28, wherein the probabilistic model is based on historical bioelectrical data and historical motion data obtained from a plurality of patients.

[0180] Example 30. The method of any one of Examples 27 to 29, wherein the probabilistic model is based on at least one of (i) the historical bioelectrical data and the historical motion data of the patient and (ii) the historical bioelectrical data and the historical motion data obtained from a plurality of patients.

[0181] Example 31. The method of any one of Examples 27 to 30, wherein the probabilistic model comprises a supervised machine learning model.

[0182] Example 32. The method of any one of Examples 23 to 31, wherein outputting an indication of the risk parameter only occurs if the risk parameter remains unchanged for a predetermined amount of time.

[0183] Example 33. The method of any one of Examples 23 to 32, wherein the risk parameter is determined on a probabilistic scale and converted to an ordinal categorical scale.

[0184] Example 34. The method of Example 33, wherein the risk parameter comprises a risk category.

[0185] Example 35. The method of Example 34, wherein the risk category is selected from the group consisting of: “high risk,” “moderate risk,” and “low risk.”

[0186] Example 36. The method of any one of Examples 23 to 35, wherein the bioelectrical data comprises local field potential (LFP) data measured by deep brain stimulation electrodes.

[0187] Example 37. The method of any one of Examples 23 to 36, wherein the neurological condition or disease comprises epilepsy or an epilepsy-related disorder and the symptom comprises a seizure.MDT Ref. No. A0013188W001

[0188] Example 38. The method of any one of Examples 23 to 37, wherein the neurological condition or disease comprises a movement disorder comprising one or more of Parkinson’s disease, essential tremor, and dystonia and the symptom comprises a motor symptom.

[0189] Example 39. The method of any one of Examples 23 to 38, further comprising alerting one or more of the patient or a clinician of a change in the determined risk parameter.

[0190] Example 40. The method of any one of Examples 23 to 39, wherein the display is part of a mobile computing device, and wherein the display further comprises an indication of historical symptom occurrences of the neurological condition or disease for the patient.

[0191] Example 41. One or more computer-readable media storing instructions that, when executed by one or more processors of a medical device, cause the medical device to perform operations comprising:receiving, via one or more processors, bioelectrical data indicative of bioelectrical activity in a subcortical region of a patient’s brain;receiving, via the one or more processors, motion data indicative of patient movement; determining, via the one or more processors, a risk parameter indicative of a likelihood of an occurrence of a symptom of a neurological condition or disease of the patient, based on the bioelectrical data and the motion data; andoutputting, via the one or more processors, an indication of the risk parameter on a display.

[0192] Example 42. The one or more computer-readable media of Example 41, further comprising adjusting one or more parameters of electrical stimulation delivered to the patient’s brain based on the determined risk parameter.

[0193] Example 43. The one or more computer-readable media of Example 41 or 42, further comprising modifying a medication protocol for the patient based on the determined risk parameter.

[0194] Example 44. The one or more computer-readable media of any one of Examples 41 to 43, wherein the risk parameter is a first risk parameter determined at a first time and the operations further comprise determining, via the one or more processors, a second risk parameter at a second time after the first time, the second risk parameter different than the first risk parameter.

[0195] Example 45. The one or more computer-readable media of any one of Examples 41 to 44, wherein determining the risk parameter comprises inputting the bioelectrical data and the motion data into a probabilistic model.MDT Ref. No. A0013188W001

[0196] Example 46. The one or more computer-readable media of Example 45, wherein the probabilistic model is based on historical bioelectrical data and historical motion data obtained from the patient.

[0197] Example 47. The one or more computer-readable media of Example 45 or 46, wherein the probabilistic model is based on historical bioelectrical data and historical motion data obtained from a plurality of patients.

[0198] Example 48. The one or more computer-readable media of any one of Examples 45 to 47, wherein the probabilistic model is based on at least one of (i) the historical bioelectrical data and the historical motion data obtained from the patient and (ii) the historical bioelectrical data and the historical motion data obtained from a plurality of patients.

[0199] Example 49. The one or more computer-readable media of any one of Examples 45 to 48, wherein the probabilistic model comprises a supervised machine learning model.

[0200] Example 50. The one or more computer-readable media of any one of Examples 41 to 49, wherein outputting an indication of the risk parameter only occurs if the risk parameter remains unchanged for a predetermined amount of time.

[0201] Example 51. The one or more computer-readable media of any one of Examples 41 to 50, wherein the risk parameter is determined on a probabilistic scale and converted to an ordinal categorical scale.

[0202] Example 52. The one or more computer-readable media of any one of Examples 41 to 51, wherein the risk parameter comprises a risk category.

[0203] Example 53. The one or more computer-readable media of Example 52, wherein the risk category is selected from the group consisting of: “high risk,” “moderate risk,” and “low risk.”

[0204] Example 54. The one or more computer-readable media of any one of Examples 41 to 53, wherein the bioelectrical data comprises local field potential (LFP) data measured by deep brain stimulation electrodes.

[0205] Example 55. The one or more computer-readable media of any one of Examples 41 to 54, wherein the neurological condition or disease comprises epilepsy or an epilepsy -related disorder and the symptom comprises a seizure.

[0206] Example 56. The one or more computer-readable media of any one of Examples 41 to 55, wherein the neurological condition or disease comprises a movement disorder comprisingMDT Ref. No. A0013188W001 one or more of Parkinson’s disease, essential tremor, and dystonia and the symptom comprises a motor symptom.

[0207] Example 57. The one or more computer-readable media of any one of Examples 41 to 56, further comprising alerting one or more of the patient or a clinician of a change in the determined risk parameter.

[0208] Example 58. The one or more computer-readable media of any one of Examples 41 to 57, wherein the display is part of a mobile computing device, and wherein the display further comprises an indication of historical symptom occurrences of the neurological condition or disease for the patient.CONCLUSION

[0209] Although many of the embodiments are described above with respect to systems, devices, and methods for symptom forecasting of neurological conditions or disorders, the technology is applicable to other applications and / or other approaches, such as symptom forecasting of medical conditions elsewhere in the body. Moreover, other embodiments in addition to those described herein are within the scope of the technology. Additionally, several other embodiments of the technology can have different configurations, components, or procedures than those described herein. A person of ordinary skill in the art, therefore, will accordingly understand that the technology can have other embodiments with additional elements, or the technology can have other embodiments without several of the features shown and described above with reference to FIGS. 1-9.

[0210] The descriptions of embodiments of the technology are not intended to be exhaustive or to limit the technology to the precise form disclosed above. Where the context permits, singular or plural terms may also include the plural or singular term, respectively. Although specific embodiments of, and examples for, the technology are described above for illustrative purposes, various equivalent modifications are possible within the scope of the technology, as those skilled in the relevant art will recognize. For example, while steps are presented in a given order, alternative embodiments may perform steps in a different order. The various embodiments described herein may also be combined to provide further embodiments.

[0211] As used herein, the terms “generally,” “substantially,” “about,” and similar terms are used as terms of approximation and not as terms of degree, and are intended to account for the inherent variations in measured or calculated values that would be recognized by those of ordinary skill in the art.MDT Ref. No. A0013188W001

[0212] Moreover, unless the word “or” is expressly limited to mean only a single item exclusive from the other items in reference to a list of two or more items, then the use of “or” in such a list is to be interpreted as including (a) any single item in the list, (b) all of the items in the list, or (c) any combination of the items in the list. Additionally, the term "comprising" is used throughout to mean including at least the recited feature(s) such that any greater number of the same feature and / or additional types of other features are not precluded. It will also be appreciated that specific embodiments have been described herein for purposes of illustration, but that various modifications may be made without deviating from the technology. Further, while advantages associated with certain embodiments of the technology have been described in the context of those embodiments, other embodiments may also exhibit such advantages, and not all embodiments need necessarily exhibit such advantages to fall within the scope of the technology. Accordingly, the disclosure and associated technology can encompass other embodiments not expressly shown or described herein.

Claims

MDT Ref. No. A0013188W001CLAIMSI / We claim:

1. A system for forecasting symptoms of a neurological disease or condition for a patient, the system comprising:a plurality of electrodes configured to be implanted in and deliver electrical stimulation to a subcortical region of the patient’s brain;a motion sensor;one or more processors operably coupled to the plurality of electrodes and the motion sensor; anda memory operably coupled to the one or more processors and storing instructions that, when executed by the one or more processors, cause the system to perform operations comprising:receiving, from at least one of the electrodes, bioelectrical data indicative of bioelectrical activity in the subcortical region of the patient’s brain, receiving, from the motion sensor, motion data indicative of patient movement, based on the bioelectrical data and the motion data, determining a risk parameter indicative of a likelihood of an occurrence of a symptom of the neurological disease or condition.

2. The system of Claim 1, wherein the operations further comprise outputting an indication of the risk parameter on a display.

3. The system of Claim 2, wherein the display is part of a mobile computing device, and wherein the display further comprises an indication of historical symptom occurrences of the neurological condition or disease for the patient.

4. The system of any one of Claims 1 to 3, wherein the operations further comprise adjusting one or more parameters of the electrical stimulation delivered by the plurality of electrodes based on the determined risk parameter.

5. The system of any one of Claims 1 to 4, wherein the operations further comprise adjusting a medication protocol for the patient based on the determined risk parameter.MDT Ref. No. A0013188W001 6. The system of any one of Claims 1 to 5, wherein the plurality of electrodes and the motion sensor are physically connected to the same implantable stimulation device.

7. The system of any one of Claims 1 to 6, wherein determining the risk parameter comprises inputting the bioelectrical data and the motion data into a probabilistic model.

8. The system of Claim 7, wherein the probabilistic model is based on historical bioelectrical data and historical motion data obtained from the patient.

9. The system of claim 7 or 8, wherein the probabilistic model comprises a supervised machine learning model.

10. The system of Claim 1, wherein the risk parameter is determined on a probabilistic scale and converted to an ordinal categorical scale.

11. The system of any one of Claims 1 to 10, wherein the bioelectrical data comprises local field potential (LFP) data.

12. The system of any one of Claims 1 to 11, wherein outputting an indication of the risk parameter only occurs if the risk parameter remains unchanged for a predetermined amount of time.

13. The system of any one of Claims 1 to 12, wherein the neurological condition or disease comprises epilepsy or an epilepsy-related disorder and the symptom comprises a seizure.

14. The system of any one of Claims 1 to 13, wherein the neurological condition or disease comprises a movement disorder comprising one or more of Parkinson’s disease, essential tremor, and dystonia and the symptom comprises a motor symptom.

15. The system of any one of Claims 1 to 14, further comprising alerting one or more of the patient or a clinician of a change in the determined risk parameter.