Predicting abnormal brain activity based on an unexpected change in measured tissue excitability in response to stimulation

Abstract

Abnormal activity in the brain can be detected by identifying changes in tissue excitability measured in response to stimulation of fiber tracts such as the corpus callosum. A stimulating electrode contact can apply a stimulation to the fiber tract, which can have one or more terminals to a cortex of a patient's brain, and a recording electrode contact positioned in the fiber tract and / or the cortex to record a tissue excitability response to the stimulation. A controller can receive the tissue excitability response from the at least one recording electrode, identify a parameter of the tissue excitability response, and predict the abnormal activity in the patient's brain based on the parameter being different from a baseline for the parameter. When an abnormal activity is predicted the patient can be alerted to start a therapy and / or a therapy can be applied in closed loop manner by a connected therapy device.

Description

NONPROVISIONAL APPLICATIONPREDICTING ABNORMAL BRAIN ACTIVITY BASED ON AN UNEXPECTED CHANGE IN MEASURED TISSUE EXCITABILITY IN RESPONSE TO STIMULATIONGovernment Support

[0001] This invention was made with government support under NS114120 awarded by the National Institutes of Health. The government has certain rights in the invention.Cross-Reference to Related Applications

[0002] This application claims the benefit of U.S. Provisional Application Serial No. 63 / 717,347, filed 7 November 2024, entitled “TISSUE EXCITABILITY MEASUREMENTS FOR OPTIMIZING STIMULATION PROTOCOLS DURING AND AFTER ELECTRODE IMPLANTATION FOR NEUROMODULATIONTECHNIQUES”, the entirety of this provisional applications is incorporated by reference for all purposes.Technical Field

[0003] This disclosure relates generally to treating abnormal brain activity, and more specifically to systems and methods for predicting abnormal brain activity caused by a brain disorder based on an unexpected change in measured tissue excitability in response to stimulation to preemptively treat the abnormal brain activity.Background

[0004] Epilepsy is a chronic brain disorder characterized by recurrent, unprovoked seizures caused by sudden, abnormal electrical activity in the brain. There have been significant improvements in epileptic therapies (e.g., deep brain stimulation, pharmaceutical treatments, and other forms of neuromodulation) to prevent and treat seizures. Some epileptic therapies can be administered at regular intervals to attempt to prevent the occurrence of seizures. Other epileptic therapies can be administered after a seizure has already begun (e.g., to lessen the effects or length of the seizure). However, these epileptic therapies do not work for all patientsand cannot entirely prevent seizures from occurring. In fact, no therapy exists that can preemptively treat a seizure before the seizure happens at least because predicting seizure activity has been prohibitively difficult.Summary

[0005] Described herein is a system and method for predicting abnormal brain activity. The predicted abnormal brain activity can be used to provide a therapy that can preemptively treat the abnormal brain activity. Abnormal activity can be predicted based on one or more unexpected changes in parameters (e.g., amplitude, ratio of interspike intervals, or the like) in real-time related to tissue excitability of the brain can be detected in response to stimulation of a fiber tract in the brain to enable clinicians to predict the abnormal brain activity, tailor therapies, and optimize stimulation parameters in closed-loop systems for proactive control.

[0006] In an aspect, the present disclosure can include a system that can predict abnormal brain activity caused by a brain disorder based on an unexpected change in measured tissue excitability in response to stimulation to preemptively treat the abnormal brain activity. The system can include at least one stimulating electrode contact configured to be positioned in a fiber tract to apply a stimulation to the fiber tract. The fiber tract has one or more terminals to a cortex of a patient's brain. The system can also include at least one recording electrode contact configured to be positioned in the fiber tract and / or the cortex to record a tissue excitability response to the stimulation. The system also includes a signal generator in communication with the at least one stimulating electrode contact and configured to provide a stimulation waveform to the at least one stimulating electrode contact for the stimulation; and a controller in communication with the signal generator and the at least one recording electrode contact. The controller can include: a non-transitory memory storing instructions; and a processor configured to access the memory to execute the instructions to at least: receive the tissue excitability response from the at least one recording electrode; identify a parameter of the tissue excitability response; and predict an abnormal activity in the patient’s brain based on the parameter being different from a baseline for the parameter.

[0007] In another aspect, the present disclosure can include a method for predicting abnormal brain activity caused by a brain disorder. Steps of the method can be performed by a system comprising a processor. The method includesconfiguring a stimulation having a single pulse of a constant frequency; applying the stimulation via a stimulating electrode within a fiber tract having one or more terminals in a cortex of a patient’s brain; receiving a tissue excitability response to the stimulation from a recording electrode within the fiber tract and / or the cortex; and predicting whether an abnormal activity is likely by comparing an amplitude of the tissue excitability response to a baseline of the amplitude, wherein the abnormal activity is predicted when the amplitude is different from the baseline of the amplitude.

[0008] In still another aspect, the present disclosure can include another method for predicting abnormal brain activity caused by a brain disorder. Steps of the method can be performed by a system comprising a processor. The method includes configuring a stimulation having paired pulses; applying the stimulation to a fiber tract having one or more terminals in a cortex of a patient’s brain; receiving a first cortical evoked potential and a second cortical evoked potential in response to the stimulation; and predicting if an abnormal activity is likely based on an amount of change of the ratio of the second evoked potential over the first evoked potential compared to a baseline of the ratio of the second evoked potential over the first evoked potential.Brief Description of the Drawings

[0009] The foregoing and other features of the present disclosure will become apparent to those skilled in the art to which the present disclosure relates upon reading the following description with reference to the accompanying drawings, in which:

[0010] FIG. 1 is a diagram showing an example system that can predict abnormal brain activity caused by a brain disorder based on an unexpected change in measured tissue excitability in response to stimulation to preemptively treat the abnormal brain activity;

[0011] FIG. 2 is a diagram showing example placements of the stimulating electrode contact(s) and recording electrode contact(s) of the system of FIG. 1 ;

[0012] FIG. 3 is a diagram showing an example controller of the system of FIG. 1 ;

[0013] FIGS. 4-6 are process flow diagrams illustrating methods for predicting abnormal brain activity caused by a brain disorder;

[0014] FIGS. 7 - 9 each show images of the experimental design and graphical representations of experimental results of experiment 1 ;

[0015] FIGS. 10 - 12 each show graphical representations of experimental results of experiment 1 ;

[0016] FIGS. 13 - 15 each show images of the experimental design and graphical representations of experimental results of experiment 2; and

[0017] FIG. 16 and 17 show images of the experimental design and graphical representations of experimental results of experiment 3.Detailed DescriptionI. Definitions

[0018] Unless otherwise defined, all technical terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present disclosure pertains.

[0019] As used herein, the singular forms “a,” “an,” and “the” can also include the plural forms, unless the context clearly indicates otherwise.

[0020] As used herein, the terms “comprises” and / or “comprising,” can specify the presence of stated features, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, steps, operations, elements, components, and / or groups.

[0021] As used herein, the term “and / or” can include any and all combinations of one or more of the associated listed items.

[0022] As used herein, the terms “first,” “second,” etc. should not limit the elements being described by these terms. These terms are only used to distinguish one element from another. Thus, a “first” element discussed below could also be termed a “second” element without departing from the teachings of the present disclosure. The sequence of operations (or acts / steps) is not limited to the order presented in the claims or figures unless specifically indicated otherwise.

[0023] As used herein, the term “abnormal brain activity”, also referred to as “abnormal tissue activity”, can refer to irregular electrical activity (e.g., signals and / or patterns) in a patient’s brain that deviate from normal electrical functioning. The irregular electrical signals can be indicative of a brain disorder. Although labeled abnormal brain activity, in some instances, the abnormal brain activity deviating fromnormal electrical functioning can cause an abnormal reaction in muscle cells, nerve cells, and / or other organs of the human body that conduct electricity.

[0024] As used herein, the term “brain disorder” can refer to one or more conditions and / or disorders affecting the structure, function, chemistry, and the like, of a patient’s brain causing a variety of symptoms, including changes in mood, behavior, movement, cognition, and sensation. Examples of brain disorders can include epilepsy (causing seizures), seizure-like states, motor-disorders (including Parkinson’s disease and dystonia), strokes, brain tumors, neurodegenerative diseases, or the like.

[0025] As used herein, the term “seizure” can refer to unusual and / or uncontrolled electrical activity in the patient’s brain that can cause changes in behavior, movements, feelings, and / or levels of consciousness.

[0026] As used herein, the term “seizure-like state” can refer to one or more episodes that resemble epileptic seizures but can have other causes, such as psychological conditions, sleep conditions, mental health conditions, or the like.

[0027] As used herein, the term “corpus callosum” can refer to a wide, thick nerve fiber tract, including a flat bundle of commissural fibers (including thousands of axons), located beneath the cerebral cortex in the brain. The corpus callosum connects the left and right sides of the brain, allowing for communication between both hemispheres. The corpus callosum includes white matter tracts that can terminate in one or more cell bodies in the cortex.

[0028] As used here, the term “fiber tracts”, also referred to as “white matter tracts”, can refer to bundles of myelinated nerve fibers that connect different regions of the brain and spinal cord.

[0029] As is used herein, the terms “cerebral cortex” and “cortex” can refer to the outer layer of neural tissue of the cerebrum of the brain that is separated into two cortices by the longitudinal fissure that divides the cerebrum into the left and right hemispheres. The two hemispheres are joined beneath the cortex by the corpus callosum. The cortex is composed of folded gray matter, is the largest site of neural integration in the central nervous system (brain and spinal cord), and plays a key role in attention, perception, awareness, thought, memory, language, and consciousness.

[0030] As used herein, the term “stimulation” can refer to delivery of one or more signals (e.g., electrical signals) to at least a portion of neural tissue, such as one or more fiber tracts, to activate conduction within a nerve or group of nerves in the one or more fiber tracts or associated cell bodies. As an example, the signal can be delivered by one or more stimulating contacts of one or more electrodes, which can be placed within, adjacent, and / or near the fiber tract(s). Stimulation can be electrical, where an electrical signal having a monopolar, a bipolar, or the like, current waveform and each electrical signal can have parameters such as waveform shape, frequency, amplitude, or the like.

[0031] As used herein, the term “tissue excitability response” can refer to a change in membrane conductance in a neural tissue or muscle in response to external stimulation. For example, neural tissue can generate electrical signals in response to application of an electrical signal (e.g., stimulation). For example, a tissue excitability response can include Evoked potentials, Cortico-callosal Evoked Potentials (CcEPs), Corpus Callosum Evoked Echoes (CcEEs), Compound Action Potentials (CAPs), or the like. Normally, tissue excitability in the brain is tightly regulated and fairly regular and tissue excitability response to an unchanged stimulation is fairly similar (e.g., can form a baseline). However, the onset of abnormal activity (e.g., because of one or more brain disorders) can cause one or more changes in a tissue excitability response, some of which can indicate the one or more brain disorders.

[0032] As used herein, the term “recording” refers to the act of measuring tissue excitability in response to a stimulation by one or more recording electrode contacts and keeping chronical of the measurements.

[0033] As used herein, the terms “patient” and “subject” can be used interchangeably and refer to any warm-blooded organism including, but not limited to, a human being, a pig, a rat, a mouse, a dog, a cat, a goat, a sheep, a horse, a monkey, an ape, a rabbit, a cow, etc.II. Overview

[0034] Epilepsy, as well as other conditions of the brain and / or body that involve unexpected, abnormal tissue excitability, can have episodes (e.g., seizures, jerking motions, etc.) that are extremely difficult to predict, and therefore difficult to treateffectively. Current therapeutic interventions aim to normalize and modulate neural activity and / or calm tissue excitability (including surgical resection (depending on the portion(s) of the brain affected), pharmacological intervention, or neuromodulation (e.g., vagus nerve stimulation (VNS), transcranial electrical stimulation (TES), deep brain stimulation (DBS) (High-Frequency Stimulation and Low-Frequency Stimulation), or the like). Using epilepsy as a non-limiting example, surgical resection and pharmacological interventions are not effective and / or available for a significant number of patients with epilepsy. Neuromodulation therapies are promising alternatives for these patients (and for patients who currently use pharmacological interventions). Currently, however, there is no mechanism to simply and accurately predict episode onset and tailor stimulation parameters in response to a predicted episode for an individual patient. Stimulation with sub-optimal parameters can have unwanted side effects, provide no therapeutic benefit, and, in extreme cases, may put patients at risk of death or permanent brain damage. Accurate assessment of cortical excitability can help to define personalized parameters for the patient and remains a critical unmet need in epilepsy care. Neurologists currently rely on indirect or qualitative measures — such as interictal spikes, seizure frequency, or patient- reported auras — to infer seizure risk, none of which reliably capture the brain’s dynamic susceptibility to seizures.

[0035] The systems and methods described herein can predict onset of abnormal activity of the brain, such as seizures, in real time to enable clinicians to predict seizure likelihood, tailor therapies, and optimize stimulation parameters in closed-loop systems. This can transform epilepsy management from reactive intervention to proactive control, tailor stimulation parameters to individual needs, minimize side effects, and maximize therapeutic efficacy. The systems and methods assess tissue excitability (e.g., detect changes in tissue excitability response to stimulation) allowing for real-time adjustment recording and prediction of seizure onset based on changes in parameters of the measure of tissue excitability. Additionally, the systems and methods can employ closed-loop systems to continuously monitor at least one tissue excitability response, detect a change in the at least one tissue excitability response, and take corrective action based on the change in the tissue excitability response.III. Systems

[0036] An aspect of the present disclosure relates to predicting abnormal brain activity (e.g., seizures, seizure like states, muscle spasms, or the like), which can be caused by a brain disorder (e.g., epilepsy, Parkinson’s, dystonia, or the like). After detection, the abnormal brain activity can be stopped, prevented, and / or lessened by taking corrective action. Although epilepsy and seizures are described herein for simplicity of explanation, it will be understood that any type of “abnormal brain activity” can be predicted and corrected in similar ways.

[0037] As shown in FIG. 1 , a system 10 can be used to detect an unexpected change in measured tissue excitability response to stimulation. The system 10 can predict abnormal brain activity based on changes in the measured tissue excitability response to the stimulation and can alert and / or preemptively treat the abnormal brain activity to stop, prevent, and / or lessen the length and / or symptoms of the abnormal brain activity. The system 10 can include a signal generator 12 that can be in communication (wired and / or wireless) with at least one stimulating electrode contact (e.g., stimulating electrode contact(s) 14) to apply an electrical stimulation to neural tissue (e.g., at least one fiber tract). The stimulating electrode contact(s) 14 can be positioned within, on, near, and / or adjacent to the at least one fiber tract and can apply the stimulation to the at least one fiber tract. The stimulating electrode contact(s) can provide monopolar or bipolar stimulation with single pulses and / or paired pulses. The system 10 can further include at least one recording electrode contact (e.g., recording electrode contact(s) 16) that can measure a tissue excitability response from the neural tissue (e.g., from the at least one fiber tract and / or the cortex where the at least one fiber tract has at least one terminal). The recording electrode contact(s) 16 can be positioned on, within, near, and / or adjacent to the neural tissue (e.g., the at least one fiber tract and / or the cortex where the at least one fiber tract has at least one terminal). The tissue excitability response can be, for example at least one of an evoked potential, a cortico-callosal evoked potential, a corpus callosum evoked echo, a cortical evoked potential, or a compound action potential.

[0038] The system 10 can also include a controller 18 that can include a non- transitory memory (e.g., memory 20) storing instructions, and previous data and a processor 22 to access the memory and execute the instructions for predictingabnormal brain activity. It should be noted that the memory 20 and processor 22 can be embodied as separate devices or can be embodied together as one device (e.g., a microprocessor). The controller 18 can be in communication (wired and / or wireless) with the signal generator 12 to configure and apply the stimulation and the recording electrode contact(s) 16 to receive the measured tissue excitability responses. The controller 18 can also, optionally, be in communication (wired and / or wireless) with one or more alert devices 24 and / or therapy devices 26. The controller 18 can, for instance, receive the tissue excitability response from the at least one recording electrode; identify a parameter of the tissue excitability response; and predict an abnormal activity in the patient’s brain based on the parameter being different from a baseline for the parameter of the tissue excitability response.

[0039] When the abnormal activity is predicted the alert device 24 can, for instance, provide an alert (e.g., visual, audible, haptic, or the like) to a patient, a caregiver, and / or a medical professional that the abnormal activity has been predicted. The alert can include the amount of change of the amplitude compared to the baseline, a graphical representation of the amplitudes for a predetermined amount of time (e.g., the last minute, the last five minutes, the last 10 minutes, or the like), a suggested course of treatment based on the amount and / or direction of change of the amplitude, a predicted time before seizure onset, or the like. When the abnormal activity is predicted the controller 20 can additionally and / or alternatively instruct the therapy device 26 and / or the signal generator 12 and stimulating electrode contact(s) 14 to apply a therapy to the patient to preemptively stop, prevent, and / or lessen the abnormal activity’s severity and / or length. The therapy can be an electrical therapy that can be provided by the signal generator 12 and the stimulating electrode contact(s) 14 or by another stimulating electrode contact (not shown) positioned elsewhere in and / or on the brain. The therapy can also be a pharmaceutical therapy (e.g., an injection, providing a pill, or the like) or another form of neuromodulation (e.g., heat, light, magnetic, cold, or the like) that can be provided by therapy device 26 associated with the patient.

[0040] In some instances, the controller 18 can also configure the stimulation (e.g., a stimulation waveform) and then the signal generator 12 can generate the stimulation. In other instances, the signal generator 12 can configure and generate the stimulation (e.g., the stimulation waveform). The stimulation can be electricalsignal (e.g., that can be described as a waveform) having a monopolar, bipolar, or the like, current waveform with a single pulse (monopolar or bipolar) and / or a dual pulse (monopolar or bipolar). Each waveform can have one or more parameters such as waveform shape, frequency, amplitude, or the like, which can be constant or variable. For example, the stimulation can include a single pulse having a constant frequency (but may have variable or constant other parameters). In another example, the stimulation can include paired pulses (otherwise referred to as dual pulses) that can have a constant or variable interpulse frequency and may have variable or constant other parameters.

[0041] In response to the application of the stimulation to the neural tissue (e.g., the fiber tract(s) and / or the cortex) by the stimulating electrode contact(s) 14 the neural tissue can have a tissue excitability response that can be recorded by recording electrode(s) 16. FIG. 2 shows illustrations of example configurations of the stimulating and recording electrode contact(s) 15 and 16 relative to the fiber tract(s) with one or more terminals in the cortex. It should be understood that the fiber tract(s) and cortex are shown as boxes for the ease of illustration and description and are not anatomically correct or to scale. The fiber tract(s) can be at least one of a corpus callosum, a dorsal hippocampal commissure, a ventral hippocampal commissure, or the like. The fiber tract(s) can each have one or more terminals to cell layers of the cortex (not shown for ease of illustration).

[0042] FIG. 2, element A shows that the at least one stimulating electrode contact 14(1 ) can be positioned on, in, near, and / or adjacent to the at least one fiber tract and at least one recording electrode contact 16(1) can be positioned in, on, near, and / or adjacent the cortex. In this configuration the stimulation can be a single pulse stimulation at a constant frequency, and the tissue excitability response can be a cortico-callosal evoked potential and / or an evoked potential. The parameter identified from the tissue excitability response can be an amplitude of the cortico- callosal evoked potential and / or the evoked potential, and the abnormal activity can be predicted when the amplitude of the cortico-callosal evoked potential and / or an evoked potential is different from (e.g., above and / or below) a baseline for the amplitude of the cortico-callosal evoked potential and / or an evoked potential during normal brain activity for the patient.

[0043] FIG. 2, element B shows that the at least one stimulating electrode contact 14(2) can be positioned on, in, near, and / or adjacent to the at least one fiber tract and at least one recording electrode contact 16(2) can be positioned in, on, near, and / or adjacent the same portion or another portion of the at least one fiber tract. In this configuration when the stimulation can be a single pulse stimulation at a constant frequency, then the tissue excitability response can be a corpus callosum evoked echo. The parameter identified from the tissue excitability response can be an amplitude of the corpus callosum evoked echo, and the abnormal activity can be predicted when the amplitude of the corpus callosum evoked echo is different from (e.g., above and / or below) a baseline amplitude of the corpus callosum evoked echo during normal brain activity for the patient.

[0044] FIG. 2, element C shows that the at least one stimulating electrode contact 14(3) can be positioned on, in, near, and / or adjacent to the at least one fiber tract and at least one recording electrode contact 16(3) can be positioned in, on, near, and / or adjacent the cortex. In this configuration when the stimulation can be a paired pulses stimulation then the tissue excitability response can be a first cortical evoked potential and a second cortical evoked potential. The parameter identified from the tissue excitability response can be the ratio of the second evoked potential over the first evoked potential, and the abnormal activity can be predicted when the ratio of the second evoked potential over the first evoked potential changes relative to a baseline ratio at different interspike intervals.

[0045] FIG. 3 shows an example of the controller 18 in greater detail. While not shown in FIG. 3, the controller 18 can be in communication (wired and / or wireless) with the signal generator, the recording electrode contact(s), and optionally an alert device and / or therapy device as shown in FIG. 1 . The controller 18 can include the non-transitory memory (e.g., memory 20) that can store instructions, and can store previously identified parameters, and the processor 22 that can access the memory to execute the instructions.

[0046] The instructions can optionally include an instruction to stimulate 30 to apply the stimulation from the signal generator through the at least one stimulating electrode contact to the at least one fiber tract (e.g., the corpus callosum, dorsal hippocampal commissure, ventral hippocampal commissure, or the like). Stimulating the at least one fiber tract can have an effect felt in the fiber tract(s) and / or in thecortex where the fiber tract(s) have one or more terminals to cell layers of the cortex. In one instance, the stimulation can have a single pulse with a constant frequency. The single pulse with a constant frequency can have one or more other parameters that can be varied, including pulse shape, pulse duration, amplitude, duty cycle, or the like depending on the patient and / or the brain disorder. In another instance, the simulation can be paired pulses (e.g., dual pulses). The paired pulses can have a constant offset frequency and / or a variable offset frequency and can have one or more other parameters that can be varied, including pulse shape, pulse duration, amplitude, duty cycle, or the like depending on the patient and / or the brain disorder. The simulation signal(s) can include a monopolar or bipolar waveform including a regularly spaced repeating pulse (each monopolar or bipolar) or a dual pulse with an intrapulse interval (monopolar or bipolar). The stimulation signal(s) can be configured by a medical professional associated with the patient, the processor 22, and / or a signal generator, and / or can be stored in the memory 20.

[0047] After the simulation, the processor 18 and / or the controller 20 can receive 32 a tissue excitability response to the stimulation from the at least one recording electrode contact (e.g., recording electrode contact(s) 16). The time after the stimulation can be based on the tissue excitability response being measured and / or the distance between the stimulating electrode contact(s) and the recording electrode contact(s). The tissue excitability response can depend on the location of the at least one recording electrode (as discussed in greater detail in FIG. 2) and can be at least one of an evoked potential, a cortico-callosal evoked potential, a corpus callosum evoked echo, a compound action potential, or the like.

[0048] From the received tissue excitability response, the processor 20 can identify 34 at least one parameter (also referred to as a biomarker) of the tissue excitability response. The at least one parameter of the tissue excitability response can be an amplitude when the tissue excitability response is a cortico-callosal evoked potential, an evoked potential, a corpus callosum evoked echo, or the like in response to a single pulse stimulation. In another instance, the at least one parameter of the tissue excitability response can be a ration of a second cortical evoked potential over a first cortical evoked potential when the tissue excitability response is the first and second cortical evoked potentials in response to paired pulse stimulation.

[0049] The processor 20 can then predict 36 abnormal brain activity (e.g., onset, current abnormal activity, building abnormal activity, etc.) based on whether the identified parameter of the tissue excitability response is different from a baseline for the parameter. The baseline can be a baseline range determined for the patient based on previously recorded tissue excitability responses during normal brain activity. Abnormal brain activity can alter the tissue excitability responses. Abnormal brain activity can be predicted, for instance, when the parameter (e.g., amplitude, ratio, or the like) is greater than the baseline range. In another instance, abnormal brain activity can be predicted when the parameter is below the baseline range.

[0050] For instance, a prediction can be made about whether or not an abnormal activity is likely (e.g., to start / be starting) by comparing the amplitude of the cortico-callosal evoked potential and / or the evoked potential to a baseline. In another instance, the prediction can be made about whether or not an abnormal activity is likely by comparing the amplitude of the corpus callosum evoked echo to a baseline. In a further instance, the prediction can be made about whether or not an abnormal activity is likely by comparing the ratio of a second cortical evoked potential over a first cortical evoked potential is different for different interspike intervals.

[0051] When abnormal activity is not predicted, then the cycle of stimulation, receiving recorded tissue excitability responses, and identifying the amplitude of received tissue excitability can continue for a next time (e.g., can repeat every second, 5 seconds, 30 seconds, minute, 10 minutes, 30 minutes, hour, or the like). When abnormal activity is predicted then the processor 20 can further execute instructions to alert 38 and / or treat 40. The instruction to alert 38 can include sending an alert to an alert device (e.g., alert device 24) that can alert a patient and / or a medical professional about the predicted abnormal brain activity. The alert can be visual, audible, and / or haptic. The alert can include the amount of change of the amplitude compared to the baseline, a graphical representation of the amplitudes for a predetermined amount of time (e.g., the last minute, the last five minutes, the last 10 minutes, or the like), a suggested course of treatment based on the amount and / or direction of change of the amplitude, a predicted time before seizure onset, or the like.

[0052] The instruction to treat 40 can include automatically providing a therapy to the patient to preemptively stop, prevent, and / or lessen the abnormal activity’sseverity and / or length. The therapy can be an electrical therapy that can be provided by the same system (e.g., through the at least one stimulating electrode contact 14) or by another stimulating electrode contact positioned elsewhere in and / or on the brain. The therapy can be a pharmaceutical therapy (e.g., an injection, providing a pill, or the like), or another form of neuromodulation (e.g., heat, light, magnetic, cold, or the like) that can be provided by another therapy device associated with the patient.

[0053] It should be noted that predictions of abnormal activity can also be utilized for diagnostic reasons (not shown), including pre-surgical evaluations, determination of best therapeutic interventions for a specific patient, determination of mechanistic foundations of abnormal activity onset, or the like.IV. Methods

[0054] Another aspect of the present disclosure can include methods 100, 200, and 300 (FIGS. 4-6) for predicting abnormal brain activity, which may be caused by a brain disorder, and stopping, preventing, or lessening the abnormal brain activity by taking corrective action. Although epilepsy and seizures are described herein for simplicity of explanation, it will be understood that any type of “abnormal brain activity” can be predicted and corrected in similar ways. The methods 100, 200, and 300 can utilize a system (shown in FIGS. 1 -3) to predict abnormal brain activity caused by a brain disorder. At least one step of the methods can be executed by at least one component that includes at least a processor.

[0055] For purposes of simplicity, the methods are shown and described as being executed serially; however, it is to be understood and appreciated that the present disclosure is not limited by the illustrated order as some steps could occur in different orders and / or concurrently with other steps shown and described herein. Moreover, not all illustrated aspects may be required to implement the method, nor is the method necessarily limited to the illustrated aspects.

[0056] Predicting abnormal activity caused by a brain disorder can transform brain disorder management from reactive intervention to proactive control. For example, predicting seizure onset or seizure like states of an epileptic patient can transform epilepsy management from reactive intervention to proactive control. This can facilitate better tailored therapeutic interventions, including tailoring stimulationparameters to individual needs, minimizing side effects, and maximizing therapeutic efficacy. Moreover, proactive treatment can prevent and / or significantly reduce the negative effects caused by uncontrolled abnormal activity (e.g., loss of consciousness, loss of muscle control, sensory disturbances, loss of memory, or the like). While not wishing to be bound by theory, it is important to note that time is of the essence when predicting abnormal brain activity and current methods can be expensive, cumbersome, and / or take unnecessary and even detrimental time to make predictions. The methods 100, 200, and 300 described herein utilize simple and real time methods for detecting increasing seizure propensity and excitability in the cortex.

[0057] FIG. 4 shows a method 100 for predicting abnormal activity from a brain disorder utilizing recorded cortex responses to stimulation of at least one fiber tract (e.g., the corpus callosum, the dorsal hippocampal commissure, or the ventral hippocampal commissure, or the like). At 102, a stimulation (e.g., an electrical stimulation in the form of a current waveform) having a single pulse with a constant frequency can be configured by a system (e.g., system 10). The configuring can be performed by a medical professional associated with the patient, the processor (e.g., processor 22), and / or a signal generator (e.g., signal generator 12). The single pulse with a constant frequency can have one or more other parameters that can be varied, including pulse shape, pulse duration, amplitude, duty cycle, or the like depending on the patient and / or the brain disorder.

[0058] At 104, the stimulation can be applied to at least one fiber tract having one or more terminals in a cortex of a patient’s brain (e.g., corpus callosum, dorsal or ventral hippocampal commissure, or the like). The stimulation can be applied by at least one stimulating electrode contact (e.g., stimulating electrode contact(s) 14) positioned within, on, near, and / or adjacent the at least one fiber tract and in communication (e.g., wired and / or wireless) with a signal generator. The stimulation can be monopolar or bipolar. At 106, the tissue excitability response to the stimulation can be received from at least one recording electrode contact (e.g., recording electrode contact(s) 16) positioned within, on, near, and / or adjacent to the portion of the cortex with the one or more terminals of the fiber tract. The tissue excitability response can be a cortico-callosal evoked potential and / or an evoked potential.

[0059] At 108, an amplitude of the cortico-callosal evoked potential and / or the evoked potential received from the at least one recording electrode contact can be identified. Then, at 110, a prediction can be made about whether or not an abnormal activity is likely (e.g., to start / be starting) by comparing the amplitude of the cortico- callosal evoked potential and / or the evoked potential to a baseline. The baseline can be a baseline range determined for the patient based on previously recorded cortico- callosal evoked potentials and / or evoked potentials in response to the stimulation during normal brain activity. Abnormal brain activity can alter the tissue excitability responses (e.g., the cortico-callosal evoked potential and / or the evoked potential). Abnormal brain activity can be predicted, for instance, when the amplitude is greater than the baseline range. In another instance, abnormal brain activity can be predicted when the amplitude is below the baseline range. When abnormal activity is not predicted, then the cycle of stimulation, receiving recorded tissue excitability responses, and identifying the amplitude of received tissue excitability can continue for a next time (e.g., can repeat every second, 5 seconds, 30 seconds, minute, 10 minutes, 30 minutes, hour, or the like).

[0060] When abnormal activity of the brain is predicted then, optionally, at 112, the patient and / or a medical professional can be alerted (e.g., by an alert device 24) about the predicted abnormal brain activity and / or to provide the patient with a therapy to pre-emptively stop and / or at least partially prevent the abnormal activity. The alert can be visual, audible, and / or haptic. The alert can include the amount of change of the amplitude compared to the baseline, a graphical representation of the amplitudes for a predetermined amount of time (e.g., the last minute, the last five minutes, the last 10 minutes, or the like), a suggested course of treatment based on the amount and / or direction of change of the amplitude, a predicted time before seizure onset, or the like. Also optionally, at 114, when the abnormal activity is predicted, then a therapy can be provided to the patient to preemptively stop, prevent, and / or lessen the abnormal activity’s severity and / or length. The therapy can be an electrical therapy that can be provided by the same system (e.g., through the at least one stimulating electrode contact) or by another stimulating electrode contact positioned elsewhere in and / or on the brain. The therapy can be a pharmaceutical therapy (e.g., an injection, providing a pill, or the like), or another form of neuromodulation (e.g., heat, light, magnetic, cold, or the like) that can beprovided by another therapy device associated with the patient. Predictions of abnormal activity can also be utilized for diagnostic reasons (not shown), including pre-surgical evaluations, determination of best therapeutic interventions for a specific patient, determination of mechanistic foundations of abnormal activity onset, or the like.

[0061] FIG. 5 shows a method 200 for predicting abnormal activity from a brain disorder utilizing tissue excitability responses recorded from a same fiber tract being stimulated (e.g., the corpus callosum, the dorsal hippocampal commissure, or the ventral hippocampal commissure, or the like). At 202, a stimulation (e.g., an electrical stimulation in the form of a current waveform) having a single pulse with a constant frequency can be configured by a system (e.g., system 10). The configuring can be performed by a medical professional associated with the patient, the processor (e.g., processor 22), and / or a signal generator (e.g., signal generator 12). The single pulse with a constant frequency can have one or more other parameters that can be varied, including pulse shape, pulse duration, amplitude, duty cycle, or the like depending on the patient and / or the brain disorder.

[0062] At 204, the stimulation can be applied to at least one fiber tract having one or more terminals in a cortex of a patient’s brain (e.g., corpus callosum, dorsal or ventral hippocampal commissure, or the like). The stimulation can be applied by at least one stimulating electrode contact (e.g., stimulating electrode contact(s) 14) positioned within, on, near, and / or adjacent the at least one fiber tract and in communication (e.g., wired and / or wireless) with a signal generator. The stimulation can be monopolar or bipolar. At 206, the tissue excitability response to the stimulation can be received from at least one recording electrode contact (e.g., recording electrode contact(s) 16) positioned within, on, near, and / or adjacent to the fiber tract. The tissue excitability response can be a corpus callosum evoked echo (e.g., a signal that echoes back to the corpus callosum from the cortex after stimulation).

[0063] Without wishing to be bound by theory axonal responses appear at a fixed time after stimulation based on the distance between stimulation and recording electrode contacts (Speed = distance / time). If distance decreases, time decreases because axonal response speed is constant. However, cellular responses always take longer (e.g., because the distance to cell bodies is much greater than thedistance to axons) so the time at which cellular responses take to appear back at a recording electrode is greater as well. So, at a given distance between stimulation and recording electrode contacts the corpus callosum evoked echo can be observed. The distance and time can depend on whether the patient is human or animal, size differences, or the like.

[0064] At 208, an amplitude of the corpus callosum evoked echo can be received from the at least one recording electrode contact can be identified. Then, at 210, a prediction can be made about whether or not an abnormal activity is likely (e.g., to start / be starting) by comparing the amplitude of the corpus callosum evoked echo to a baseline. The baseline can be a baseline range determined for the patient based on previously recorded corpus callosum evoked echoes in response to the stimulation during normal brain activity. Abnormal brain activity can alter the tissue excitability responses (e.g., the corpus callosum evoked echoes). Abnormal brain activity can be predicted, for instance, when the amplitude is greater than the baseline range. In another instance, abnormal brain activity can be predicted when the amplitude is below the baseline range. When abnormal activity is not predicted, then the cycle of stimulation, receiving recorded tissue excitability responses, and identifying the amplitude of received tissue excitability can continue for a next time (e.g., can repeat every second, 5 seconds, 30 seconds, minute, 10 minutes, 30 minutes, hour, or the like).

[0065] When abnormal activity of the brain is predicted then, optionally, at 212, the patient and / or a medical professional can be alerted (e.g., by an alert device 24) about the predicted abnormal brain activity and / or to provide the patient with a therapy to pre-emptively stop and / or at least partially prevent the abnormal activity. The alert can be visual, audible, and / or haptic. The alert can include the amount of change of the amplitude compared to the baseline, a graphical representation of the amplitudes for a predetermined amount of time (e.g., the last minute, the last five minutes, the last 10 minutes, or the like), a suggested course of treatment based on the amount and / or direction of change of the amplitude, a predicted time before seizure onset, or the like. Also optionally, at 214, when the abnormal activity is predicted, then a therapy can be provided to the patient to preemptively stop, prevent, and / or lessen the abnormal activity’s severity and / or length. The therapy can be an electrical therapy that can be provided by the same system (e.g., throughthe at least one stimulating electrode contact) or by another stimulating electrode contact positioned elsewhere in and / or on the brain. The therapy can be a pharmaceutical therapy (e.g., an injection, providing a pill, or the like), or another form of neuromodulation (e.g., heat, light, magnetic, cold, or the like) that can be provided by another therapy device associated with the patient. Predictions of abnormal activity can also be utilized for diagnostic reasons (not shown), including pre-surgical evaluations, determination of best therapeutic interventions for a specific patient, determination of mechanistic foundations of abnormal activity onset, or the like.

[0066] FIG. 6 shows a method 300 for predicting abnormal activity from a brain disorder utilizing recorded cortex responses to stimulation of at least one fiber tract (e.g., the corpus callosum, the dorsal hippocampal commissure, or the ventral hippocampal commissure, or the like). At 302, a stimulation (e.g., an electrical stimulation in the form of a current waveform) having paired pulses (e.g., dual pulses) can be configured by a system (e.g., system 10). The configuring can be performed by a medical professional associated with the patient, the processor (e.g., processor 22), and / or a signal generator (e.g., signal generator 12). The paired pulses can have a constant offset frequency and / or a variable offset frequency and can have one or more other parameters that can be varied, including pulse shape, pulse duration, amplitude, duty cycle, or the like depending on the patient and / or the brain disorder.

[0067] At 304, the stimulation can be applied to at least one fiber tract having one or more terminals in a cortex of a patient’s brain (e.g., corpus callosum, dorsal or ventral hippocampal commissure, or the like). The stimulation can be applied by at least one stimulating electrode contact (e.g., stimulating electrode contact(s) 14) positioned within, on, near, and / or adjacent the at least one fiber tract and in communication (e.g., wired and / or wireless) with a signal generator. The stimulation can be monopolar or bipolar. At 306, the tissue excitability response to the stimulation can be received from at least one recording electrode contact (e.g., recording electrode contact(s) 16) positioned within, on, near, and / or adjacent to the portion of the cortex with the one or more terminals of the fiber tract. The tissue excitability response can be a first cortical evoked potential and a second cortical evoked potential (e.g., in response to a first and second pulse of the paired pulses).

[0068] At 308, a ratio of the second evoked potential over the first evoked potential received from the at least one recording electrode contact can be identified. Then, at 310, a prediction can be made about whether or not an abnormal activity is likely (e.g., to start / be starting) by comparing the ratio to a baseline ratio. The baseline can be a baseline range determined for the patient based on previously recorded first and second evoked potentials and identified ratios in response to the stimulation during normal brain activity. The likelihood of abnormal activity can be based on an amount of change of the ratio compared to the baseline. Abnormal brain activity can alter the tissue excitability responses (e.g., the difference between the first and second evoked potentials). Abnormal brain activity can be predicted, for instance, when the change in ratios is greater than the baseline range. In another instance, abnormal brain activity can be predicted when the change in ratios is below the baseline range. When abnormal activity is not predicted, then the cycle of stimulation, receiving recorded tissue excitability responses, and identifying the ratio of second evoked potential over first evoked potential can continue for a next time (e.g., can repeat every second, 5 seconds, 30 seconds, minute, 10 minutes, 30 minutes, hour, or the like).

[0069] When abnormal activity of the brain is predicted then, optionally, at 312, the patient and / or a medical professional can be alerted (e.g., by an alert device 24) about the predicted abnormal brain activity and / or to provide the patient with a therapy to pre-emptively stop and / or at least partially prevent the abnormal activity. The alert can be visual, audible, and / or haptic. The alert can include the amount of change of the ratio compared to the baseline, a graphical representation of the amplitudes for a predetermined amount of time (e.g., the last minute, the last five minutes, the last 10 minutes, or the like), a suggested course of treatment based on the amount and / or direction of change of the amplitude, a predicted time before seizure onset, or the like. Also optionally, at 314, when the abnormal activity is predicted, then a therapy can be provided to the patient to preemptively stop, prevent, and / or lessen the abnormal activity’s severity and / or length. The therapy can be an electrical therapy that can be provided by the same system (e.g., through the at least one stimulating electrode contact) or by another stimulating electrode contact positioned elsewhere in and / or on the brain. The therapy can be a pharmaceutical therapy (e.g., an injection, providing a pill, or the like), or anotherform of neuromodulation (e.g., heat, light, magnetic, cold, or the like) that can be provided by another therapy device associated with the patient. Predictions of abnormal activity can also be utilized for diagnostic reasons (not shown), including pre-surgical evaluations, determination of best therapeutic interventions for a specific patient, determination of mechanistic foundations of abnormal activity onset, or the like.V. Experimental

[0070] The following experiments investigate simulation of the corpus callosum (CC) using single pulses and paired pulses to detect measures of tissue excitability in response to stimulation and changes in the detected measures in response to abnormal brain activity.

[0071] Experiment 1

[0072] This study showed that LFS of the CC achieves a reduction in tissue excitability and provides an antiepileptic effect in the cortex specifically in areas innervated by the fiber tract by activating GABAB receptors and sAHP mechanisms. The mechanisms used by LFS to suppress epileptic activity in the cortex are similar to the mechanisms used by currently available antiepileptic drugs.

[0073] 1. Materials and Methods

[0074] 1.1 Coronal Slice Preparation

[0075] CD-1 mice (postnatal day 10-30) from Charles River of either sex were used in this study and housed per institutional animal care and use committee guidelines. After anesthetizing with isoflurane, the brains were harvested, the cerebellum was detached, and the brains were placed in an oxygenated (02 95%, CO2 5%), cold (3-4°C), sucrose-rich artificial cerebrospinal fluid (ACSF) consisting of the following (in mrnol L-1): 220 sucrose, 3 KCI, 1 .25 NaH2PO4, 2 MgSO4, 26 NaHCOs, 2 CaCL, and 10 D-glucose (pH 7.45).

[0076] The brain was secured on its caudal surface on a vibrating-blade microtome (VTS1000S, Leica) containing sucrose-rich, cold, oxygenated ACSF. Coronal slices (300-pm thickness) were cut and immediately preserved in oxygenated ACSF containing the following (in mmol-L-1): 124 NaCI, 3.75 KCI, 1.25 KH2PO4, 2 MgSO4, 26 NaHCC , 2 CaCL, and 10 D-glucose for at least 60 minbefore being transferred to an interface-recording chamber (BSC-ZT, Harvard Apparatus).

[0077] 1 .2 Electrical Recording

[0078] Epileptiform activity from cortical cell layers Il-Ill of the M1 region of the cortex was recorded using glass recording electrodes made from fire-polished borosilicate glass micropipettes (0.5-mm inner diameter, 1 .0-mm outer diameter) filled with a 150 mmol L-1NaCI solution.

[0079] The signals obtained were amplified using an Axoclamp-2A microelectrode amplifier (Axon Instruments), low-pass filtered (5 kHz), additionally amplified (FLA-01 , Cygnus Technology), and then digitized at 20 kHz (PowerLab, AD Instruments). Analysis was performed using MATLAB (MathWorks).

[0080] 1 .3 Model Selection and Seizure Generation

[0081] Epileptiform activity was induced in brain slices by exposing them to 4-AP in ACSF, as it enhances neuronal excitability in a nonbiased manner toward any neurotransmitter. Furthermore, neuronal excitability was increased by reducing the concentrations of MgSO and CaCIs in ACSF. Various combinations of 4-AP, Mg2+, and Ca2+(see FIG. 7, element B) were applied to the brain slices. Average seizure durations and the percentage of time spent seizing were calculated for each model to identify a model that showed the least variability in both metrics and produced seizures with an average duration of at least 15 s. Seizures were defined as regions of spontaneous activity with amplitude >2 times baseline measurements, a frequency >5 Hz, and lasting >5 s.

[0082] Percent time spent seizing in 10-min windows was computed for 110 min in the selected model to identify where seizure generation remained stable (<.5% change in percent time spent seizing over time). This window was utilized to study LFS.

[0083] 1 .4 Generating Evoked Action Potentials

[0084] The dorsal CC (d.CC) was selected as the stimulation target as it comprises ascending and descending fibers from contralateral and ipsilateral cortical cell layers II, III, and V of the M1 cortex, 18 ensuring that the stimulation elicited responses from these cell layers regularly (see FIG. 8, element A). Midline stimulation to generate evoked potentials (EVPs) on either hemisphere was often unsuccessful (see FIG. 9, element A). As a result, stimulation and recordingelectrodes were placed on the same hemisphere to maximize EVP generation for all LFS and paired-pulse protocols unless otherwise mentioned. EVPs were utilized to confirm electrode positions and slice health (amplitude > 0.6 mV).

[0085] Two tungsten electrodes of opposite polarities (-800 pm apart) were used to elicit EVPs. A software-controlled stimulus isolator (FE180, ADInstruments) was used to generate monophasic pulses of 100-ps pulse duration, and the current intensity was varied between 0.5 mA and 2.5 mA to identify EVP saturation. For all electrical stimulation experiments, the current intensity was set at 90% of the current required to generate the maximum EVP with a pulse duration of 100 ps.

[0086] 1 .5 Optimizing LFS Frequency for Seizure Suppression

[0087] The LFS protocol involved three periods, each 20 min in duration: before, during, and after LFS. Four different frequencies of LFS (1 , 5, 10, or 20 Hz) were applied in the form of monophasic pulse trains (see FIG. 8, element A). Percent reduction in time spent seizing was computed between before-during LFS and before-after LFS suppression of epileptic activity. The frequency of LFS that resulted in the highest suppression of epileptic activity during stimulation was chosen to study the mechanisms of LFS.

[0088] 1 .6 Bilateral Suppression of Epileptic Activity

[0089] Stimulation electrode positions were changed to the midline d.CC; however, this setup prevented the generation of consistent EVPs bilaterally. To account for this, EVP saturation was computed i psi laterally and was used to set the stimulation current.

[0090] 1 .7 Measuring Tissue Excitability with Paired Pulses

[0091] Paired-pulse tests were conducted in normal ACSF, in the 4-AP seizure model, and under pharmacological blockers, each with and without LFS to study changes to neuronal excitability. Interstimulus intervals (ISIs; 50, 100, 200, 400, 600, 800, and 1000 ms) were used to capture changes in neuronal excitability (P2 / P1 ) using the same stimulation and recording setup described in FIG. 8, element A where P1 and P2 are the amplitudes of the first and second EVP.

[0092] In normal ACSF, pairs of pulses with specific ISIs were used to generate EVPs five times, and the resulting P2 / P1 ratio was averaged. This process was repeated for all ISIs listed above, before LFS was applied.

[0093] To study neural excitability during LFS, 5 min of LFS was used ipsilaterally, followed with the paired-pulse stimulation at a specific ISI (repeated five times and averaged). This process was repeated for all ISIs listed above. To ensure that two consecutive pairs of pulses did not interfere with each other, a delay of 10 s was used between each pair of pulses. After the paired-pulse procedure, hypoxia was induced, and EVPs were generated at 0.1 Hz to study whether the observed EVPs were antidromic or orthodromic.

[0094] After a stable baseline was obtained, a similar procedure was adopted in the case of the 4-AP model and under pharmacological blockers.

[0095] 1 .8 Pharmacology

[0096] To investigate the mechanisms of LFS, specific and nonspecific antagonists for sAHP and GABAB receptors were utilized. Suppression of epileptic activity because of LFS was computed under the following scenarios: no blockers, clotrimazole (10 pmol L-1; medium AHP and sAHP nonspecific antagonist), UCL- 2077 (10 pmol L-1; specific sAHP antagonist), 2-OH- saclofen (500 pmol-L-1; specific GABAB receptor antagonist), and 2-OH- saclofen + UCL-2077 (500 pmol-L"1, 10 pmol-L"1, respectively). The concentrations of antagonists listed above were added to the chosen 4-AP seizure model.

[0097] 1 .9 Data and Statistical Analysis

[0098] The data analysis was performed using MATLAB (Mathworks) and Labchart V8 (ADInstruments). We used a template subtraction algorithm to remove stimulation artifacts during LFS. The data are presented as mean ± SD, n, and p- value, with n indicating the sample size in number of brain slices and p-value representing the two-tailed p-value. The statistical significance was determined using either paired f-tests, two sample f-tests or analysis of variance tests.

[0099] 2. Results

[0100] 2.1 Seizure Generation

[0101] Local field potential was recorded from the M1 cortex, cell layers ll / lll (see FIG. 7, element A, under four different models of epilepsy (see FIG. 7, element B). Model 4 generated activity with the least variability in seizure durations (29.4 s ± 15.15 s) and percentage of time spent seizing (30.4% ± 4.5%) between individual brain slices (see FIG. 7, element C).

[0102] Spontaneous activity generated by all models has a distinct high- frequency segment at the beginning of a seizure followed by lower frequency, and often larger amplitude, spiking activity (see FIG. 7, element C. Unlike models 1 and 2, model 4 did not result in status epilepticus. The amplitude and duration of activity generated by model 3 were insufficient. Given these problems with models 1-3, model 4 was chosen to study LFS.

[0103] After approximately 15 min -20 min of exposure to model 4, periodic high-frequency oscillations (>5 Hz), with amplitudes ranging from 0.5 to 3 mV, and trailing low-frequency spikes (<5 Hz) occurred (see FIG. 7, element D, consistent with other studies.

[0104] Model 4 was further examined by measuring the change in time spent seizing over 10-min intervals for 110 min (n = 5; see FIG. 7, element E. A linear regression model was fitted to identify a region with negligible variation in time spent seizing over time. With an increase in time spent seizing of .046% ± .038% (p = .024) every minute, 30 min -100 min was chosen as the experimental window to study LFS.

[0105] 2.2 Five-hertz LFS Provides the Largest Antiepileptic Effect DuringStimulation

[0106] When LFS at 1 , 5, 10, and 20 Hz (n = 8 each) was applied to the d.CC (see FIG. 8, element A, suppression of epileptic activity (%) was observed relative to baseline: 1 Hz, 49.05 ± 28.08, p = .0043 (orange); 5 Hz, 76.47 ± 19.85, p = .00058 (blue); 10 Hz, 47.56 ± 36.02, not significant (NS; red); and 20 Hz, 25.03 ± 33.14, NS (yellow; see FIG. 8, element B). Statistical analysis was performed using a paired t- test between baseline and LFS. LFS of 10 and 20 Hz did not produce significant antiepileptic effects and often increased the percentage of time spent seizing.

[0107] After LFS, suppression of epileptic activity (%) decreased for all LFS frequencies to 17.35 ± 24.92, NS (orange); 23.72 ± 24.09, NS (blue); 12.80 ± 19.01 , NS (red); and 6.33 ± 8.01 , NS (yellow), respectively (see FIG. 8, element B). Seizure suppression following LFS was not significantly different from baseline activity for all frequencies (paired t-test). Five-hertz LFS was chosen as the optimal LFS frequency to perform mechanistic studies.

[0108] 2.3 LFS of the CO can Suppress Epilepsy Activity in the CortexBilaterally

[0109] To examine whether 5-Hz LFS is capable of bilaterally suppressing epileptic activity in vitro as was observed in vivo, simultaneous recordings were made in both the left and right M1 cortices, and LFS was applied to the d.CC at the midline. The stimulation suppressed epileptic activity 49.77% ± 19.01 %, p = .0304 in the left hemisphere and 65.75% ± 26.91 %, p = .0087 in the right hemisphere (n = 5 each, paired t-test).

[0110] 2.4 LFS of the CC Reduces Neuronal Excitability in the Cortex

[0111] Tissue excitability in normal ACSF (n = 5 each) was observed to be a gradual increase from 50 ms to 1000-ms ISIs: 50 ms (.87 ± .06), 100 ms (.95 ± .16), 200 ms (.91 ± .15), 400 ms (.97 ± .16), 600 ms (1 .01 ± .13), 800 ms (1 .06 ± .12), and 1000 ms (1.03 ± .095; see FIG. 10, element A, blue (top)). Tissue excitability in normal ACSF during LFS (n = 5 each) decreased rapidly between 50 and 200 ms and gradually increased between 200 and -1000 ms: 50 ms (.97 ± .18 [NS]), 100 ms (.80 ± .12 [NS]), 200 ms (.61 ± .12 [p = .0026]), 400 ms (.66 ± .08 [p = .011 ]), 600 ms (.68 ± .09 [p = .0054]), 800 ms (.75 ± .07 [p = .0058]), and 1000 ms (.76 ± .03 [p = .0079]; see FIG. 10, element A, orange (bottom)).

[0112] In the 4-AP seizure model (n = 6 each) before stimulation, tissue excitability was elevated at all ISIs: 50 ms (1 .09 ± .26), 100 ms (1 .1 1 ± .24), 200 ms (1 .02 ± .17), 400 ms (.97 ± .03), 600 ms (.98 ± .04), 800 ms (1 .03 ± .09), and 1000 ms (.98 ± .05; see FIG. 10, element A, blue (top)). When LFS was used ( / ? = 6 each), tissue excitability was reduced in a similar range observed in normal ACSF: 50 ms (.93 ± .32 [NS]), 100 ms (.87 ± .36 [NS]), 200 ms (.66 ± .14 [p = .015]), 400 ms (.77 ± .10 [p = .0097]), 600 ms (.79 ± .16 [NS]), 800 ms (.87 ± .03 [p = .032]), and 1000 ms (.87 ± .05 [p - .018]; see FIG. 10, element B, orange (bottom)). The p-values listed above were computed using a paired t-test comparing means before and after stimulation with LFS at each ISI.

[0113] It is important to note that EVPs were interfered with when LFS and 4-AP were introduced. In the case of 4-AP, stimulation often resulted in the generation of spiking superimposed on EVPs. In the case of LFS, across all samples, LFS reduced the amplitude of EVPs. To investigate trends in the reduction in amplitude of EVPs, EVPs resulting from and after 5-Hz LFS were further examined.

[0114] 2.5 Five-hertz LFS Causes a Decrease in Antidromic EVPs

[0115] The reduction in amplitude of EVPs during LFS approximated an exponential decay that reached a steady state amplitude over 120 s (n = 6; see FIG. 11 , element A. To investigate whether these EVPs returned to amplitudes observed before stimulation, EVPs were generated using the same current and pulse width as above at a frequency of .1 Hz post-LFS (n = 5), and the amplitude of EVPs was observed to increase back to what was observed before LFS over 10 min (see FIG. 11 , element B). To test whether an increase in the amplitude of EVPs is correlated to a similar rise in percentage time spent seizing, seizures post-LFS were analyzed, and time spent seizing was computed in 2-min intervals for 20 min (see FIG. 1 1 , element D). The resulting analysis showed a weak trend in time spent seizing over time where, every minute, time spent seizing increased by approximately 1 .03% ± .38%. To further investigate the type of EVPs observed in this setup, hypoxia was induced in slices after the LFS procedure described earlier at .1 Hz (n = 4). Over 1 min, no changes in EVP amplitude were observed (see FIG. 11 , element C).

[0116] 2.6 Five-hertz LFS of the CC Reduces Tissue Excitability in the Cortex using GABAB and sAHP Mechanisms

[0117] Without antagonists, 5-Hz LFS suppressed epileptic activity 76.5% ± 19.85% (see FIG. 12, element A, 4-AP only n = 8) in the cortex. The presence of any individual blocker reduced LFS efficacy significantly. In the presence of sAHP- specific and -nonspecific blockers, suppression of epileptic activity by LFS was decreased to 12.94% ± 15.81% (see FIG. 12, element A, sAHP, n = 8) and 9.82% ± 15.4% (see FIG. 12, element A, mAHP + sAHP, n = 6), respectively. When GABAB receptor antagonists were used, LFS efficacy dropped to 10.61% ± 14.33% (see FIG. 12, element A, GABAB, n = 10). When a combination of specific antagonists for sAHP and GABAB receptors were used together, LFS failed to function, with an efficacy of 2.30% ± 4.89% (see FIG. 12, element A, sAHP + GABAB, n = 6). The reduction in LFS efficacy under all these scenarios was statistically significant relative to when no antagonists were used (p < .00001 ; z-test: two samples for means).

[0118] Tissue excitability was elevated when measured in the presence of GABAB antagonists before LFS (n = 5 each): 50 ms (1.27 ± .14), 100 ms (1.20 ± .19), 200 ms (1 .13 ± .15), 400 ms (1 .04 ± .09), 600 ms (1 .03 ± .11 ), 800 ms (.98 ± .07), and 1000 ms (.97 ± .06; see FIG. 12, element B, GABA Ant. Before LFS).When 5-Hz LFS was utilized, tissue excitability did not drop to any meaningful degree in the 50-400-ms ISI range (n = 5 each): 50 ms (1.18 ± .19 [NS]), 100 ms (1 .07 ± .11 [NS]), 200 ms (1 .03 ± .13 [NS]), 400 ms (.90 ± .07 [NS]), 600 ms (.86 ± .09 [p = .0099]), 800 ms (.80 ± .09 [p = .014]), and 1000 ms (.85 ± .06 [p = .047]; see FIG. 12, element B, GABA Ant. After LFS).

[0119] Experiment 2

[0120] Experiments were run to determine excitability of different tissues in / related to the corpus callosum (CC) based on changes in recording electrode distance from a stimulating electrode. The CC connects the two cerebral hemispheres and is primarily comprised of white matter (e.g., a bundle of nerve fibers (myelinated axons)). Descending tracts of cortical tissue (also referred to as gray matter (e.g., neuron cell bodies, dendrites, and unmyelinated axons)) can descend through and / or near the CC. These Experiments found that a recording electrode positioned significantly closer to the stimulating electrode (0.05 mm or less distance vs. 0.5 mm - 2 mm distance in a mouse CC) can record a cortical response rather than traditional Compound Action Potentials (CAPs) from the bundle of nerve fibers.

[0121] 1. CAP Waveform

[0122] When a portion of the white matter of the CC is stimulated, the resulting electrical activity is known as Compound Action Potentials (CAPs) (offering a window into the behavior of the brain's white matter). A CAP waveform typically has two prominent negative peaks (N1 generated by fast-conducting, myelinated fibers, and N2 generated by slower-conducting, unmyelinated fibers with smaller diameters). Recording electrodes placed at a distance from the stimulating electrode (e.g., from 2 mm to 0.5 mm, or the like) can be used to measure the time it takes for a CAP waveform to appear at the recording electrodes after stimulation of the CC (also referred to as “the delay”). The delay depends on the conduction velocity of the different types of nerve fibers within the CC.

[0123] 2. Distance Experiment

[0124] To isolate other, slower responses (different from N1 and N2), this Experiment manipulated the distance between the stimulation and recording sites in mouse brains. Since the conduction velocity of the fibers remains constant,decreasing the distance reduces the time it takes for the CAPs to appear (Time = Distance / Velocity).

[0125] By decreasing the distance to near 0 mm (e.g., 0.05 mm or less), the N1 and N2 peaks become part of the initial stimulation artifact and eliminates the fast and slow CAPs from the recording, making it possible to observe any other responses that have a longer conduction delay.

[0126] A new response has been identified by placing both stimulating and recording electrodes at a distance approaching 0 mm (e.g., 0.05 mm or less). The experimental hypothesis is that this new response is cortical (Cx) and distinct from the typical white matter responses, known as Compound Action Potentials (CAPs) or N1 and N2 which are obtained at electrode distances greater than 0.5 mm. See FIG. 13 that a new evoked response is observed at D < 0.05 mm that does not follow the S=D / t relationship of N1 and N2.

[0127] 3. Cortical Experiments

[0128] Three Experiments were conducted to demonstrate that the newly observed responses are cortical in nature. The approach was twofold: first, the experiments established that the new responses behaved differently than the typical white matter responses (N1 and N2), and second, the experiments showed that a direct connection to the cortex is essential for the responses to exist.

[0129] 3.1 Calcium Sensitivity

[0130] A key distinction between white matter and cortical tissue is the tissues sensitivity to calcium. As white matter responses (N1 and N2) don't involve synapses, the white matter is calcium-insensitive. In contrast, cortical tissue is sensitive to calcium and cortical responses rely on synaptic transmission and are therefore dependent on calcium. A calcium knockout effectively eliminated the new responses while leaving the N1 and N2 responses unaffected, strongly suggesting a synaptic, and thus cortical, origin for the new signals.

[0131] 3.2 Oxygen Dependence:

[0132] Cortical responses, due to the high metabolic demands of the cortical synapses, require a constant supply of oxygen and are known to cease function within 30 seconds to 2 minutes of oxygen deprivation. When tested for oxygen dependence the new responses exhibited the rapid decay upon oxygen removal similar to known cortical responses. In contrast, the N1 and N2 responses persisted.The difference in oxygen dependence further supports the conclusion that the new responses are cortical rather than white matter-based.

[0133] 3.3 Connectivity with the Cortex

[0134] The anatomical connection between the CC and the cortex was investigated. The corpus callosum is formed by descending fibers from the cortex. It was shown that severing these descending fibers eliminated both the new responses and the previously established cortical responses. However, severing the descending fibers had no effect on the N1 white matter response. This finding provides direct anatomical evidence that a connection to the cortex is necessary for the new responses, solidifying the argument that the new responses are cortical in origin.

[0135] FIG. 14 shows the experimental setup and the experimental results for 3.1 -3.2. FIG. 14, element A shows the experimental design to generate the new response (green) and N1 and N2 (blue, controls). The new response appears at a delay of approximately 3 ms. FIG. 14, element B shows that the new response has a saturation curve similar to cortical responses. FIG. 14, element C shows that the new responses are eliminated by hypoxia before N1 and N2 (white matter responses) suggesting that these aren’t the same. FIG. 14, element D shows that the new responses are completely eliminated by hypocalcemia while N1 and N2 are not, further suggesting that these new responses are distinct from N1 and N2. Lastly, FIG. 14, element E shows that severing the connections between the cortex and the corpus callosum entirely eliminates both cortical and the new responses while having a minimal effect on the white matter responses (N1 and N2; N1 shown).

[0136] 4. Epileptic Connection of New Responses

[0137] The new response can be used to predict epileptic responses using paired pulse stimulation (as discussed in detail above and in the other experiment). The fundamental difference lies in the measured response.

[0138] Instead of directly recording from the cortex, this method utilizes the newly discovered cortical responses that can be recorded from the corpus callosum. The procedure remains consistent: stimulation is applied to the corpus callosum, and the resulting cortical responses are measured. This innovation allows for the assessment of cortical excitability by recording these specific cortical responses directly from the white matter tract itself.

[0139] FIG. 15 shows that the new response changes (increases) in amplitude before seizures appear. The graph depicts time to seizure on the x axis and the amplitude of the new response on the y axis. The graphs show that the amplitude of the new response goes up starting at -20 sec and continues to increase till -5 sec (relative to -25 sec as a control). These differences were significant and suggest that tracking the amplitude of these new responses enables the prediction of tissue excitability in the cortex and thus an incoming seizure.

[0140] Experiment 3

[0141] This experiment can employ a translational framework spanning human intracranial EEG (iEEG) recordings, rodent epilepsy models, and in vitro slice preparations to establish the corpus callosum (CC) as a novel platform for seizure detection, excitability monitoring, and neuromodulation. The design ensures that each Aim is both independent and synergistic, directly addressing the diagnostic, biomarker, and therapeutic gaps that limit current neuromodulation strategies.

[0142] Aim 1 : Corpus callosum-evoked cortical potentials can be validated as excitability biomarkers. Using rodent epilepsy models, the corpus callosum can be stimulated and evoked potential amplitudes in both hemispheres can be measured and then correlated with validated excitability metrics. This can establish a mechanistic, dynamic biomarker of cortical excitability, overcoming the limitations of current noisy and inconsistent measures (e.g., spike rates, high-frequency oscillations).

[0143] Aim 2: A dual-function corpus callosum electrode for pinging the cortex can be developed. Sensing and stimulating capabilities can be integrated into a single corpus callosum electrode capable of detecting rising excitability via corpus callosum evoked echo amplitudes and delivering low-frequency stimulation to suppress seizure initiation and spread. In vitro slice experiments can confirm the cortical origin of CC-evoked responses, while in vivo testing can establish feasibility for closed-loop neuromodulation.

[0144] Together, these studies can provide the conceptual, mechanistic, and technical foundation for a corpus callosum-based interface that unifies seizure detection, excitability monitoring, and neuromodulation in a single site. By bridging human and animal models, the approach ensures clinical relevance, mechanistic clarity, and translational potential.

[0145] 1. Aim 1 - Monitor cortical seizure propensity via CC-evoked potentials in rodent models

[0146] Experiments were conducted to investigate whether the amplitude of corpus callosum (CC) induces evoked potentials (CAEP) can predict the propensity to seizures. Seizures were induced by 4-AP injection, and CAEPs were recorded in response to corpus callosum stimulation before, during, and after seizures. It was found that the amplitude of CAEPs increased approximately 20-30 seconds before the seizure and then returned to baseline level following the seizures. CAEPs increased significantly in both the focus cortex and non-focus cortex (p < 0.05, n = 3 rats) (FIG. 16, element A). When the increase of CAEPs 5s and 55s were compared before the seizure onset, the CAEPs in the focus increased significantly by 224.9 ± 45.2 % (FIG. 16, element B, p<0.01 , n=3 rats). This effect was much larger in the primary focus than the mirror focus of 1 13.8 ± 5.6 % (FIG. 16, element C, p<0.05, n=3 rats). Logistic regression modeling was also conducted on the data, and it showed that the amplitude of evoked potentials is a significant predictor of the propensity to seizure (n = 72 evoked potential events). Logistic regression modeling (n=72 CAEP events) showed CAEP amplitude significantly predicted seizure occurrence (P1 =8.6, p<0.01) (see FIG. 16, element D). These results indicated and the CAEP were indicative of the propensity of seizures and that this propensity is higher on the primary site compared to the mirror site.

[0147] 2. Aim 2 - Corpus Callosum evoked echoes (CCEE) as biomarkers of cortical excitability.

[0148] Electrical stimulation of the corpus callosum (CC) elicits cortical evoked potentials (CAEPs) that, in turn, produce a secondary response detectable within the CC— an “echo” generated by cortical axons projecting through callosal fibers. Data confirms that these CC-evoked echoes (CCEE) are reproducible and reflect bidirectional signaling between the CC and cortex. Because the CCEE’s amplitude and latency vary with cortical excitability and the efficacy of low- frequency stimulation (LFS), CCEEs represent a promising biomarker of both brain state and treatment response.

[0149] An in vitro cortical-callosal slice preparation allows isolation of the cortical contribution to these echoes and differentiation from compound action potentials (CAPs). This preparation will also enable pharmacological manipulation totest how CCEE features track excitability and seizure suppression. Establishing CCEE as a dual biomarker for cortical excitability and LFS efficacy can provide the foundation for a single CC electrode capable of both monitoring and therapy— an essential component of a closed-loop system designed to maintain cortical excitability below the seizure threshold.

[0150] 2.1 Slice Preparation and Recording Setup

[0151] Coronal brain slices can be prepared from CD-1 mice (postnatal day 20- 30, both sexes; Charles River) under approved IACUC protocols. To induce epileptiform activity, 100 pM 4-aminopyridine (4-AP) can be bath-applied in modified ACSF with reduced MgS04(1 .2 mM) and CaCI2(1 .7 mM) to enhance excitability. Recordings can be obtained from cortical layers Il-Ill of M1 and the dorsal corpus callosum (dCC) using glass microelectrodes (0.5 mm ID, 1.0 mm OD, filled with 150 mM NaCI). Signals can be amplified (Axoclamp-2A, Axon Instruments), low-pass filtered (5 kHz), re-amplified (Cygnus FLA-01 ), and digitized at 20 kHz (PowerLab, AD Instruments). A single tungsten stimulating electrode (with a distant return) can evoke cortical evoked potentials (CEPs), compound action potentials (CAPs), and CC-evoked echoes (CCEEs) (see FIG. 17, element A). The dCC is chosen to activate ascending and descending callosal fibers originating from cortical layers II, III, and V (ref).

[0152] As shown with respect to (see FIG. 17, element A) CC-Evoked Cortical Echoes (CCEEs) were recorded close to the stimulation electrode and the CAP at a 2mm distance. B: CAEPs originated from the cortex as recorded from cortical electrode R3. Cutting the slice above the CC, eliminates the CAP but not the CAEP. C: Removing calcium eliminates the CCAP but not the CAP. D: Applying LFS to lower excitability decreases both the CAEP and the CCEE.

[0153] In coronal slices exposed to 100 pM 4-AP, CC stimulation elicited classic CAPs (N1 , N2) in the CC ~2 mm from the stimulating electrode. Near the stimulation site, no CAP was detected due to artifact, yet a distinct delayed potential (“CC echo”) was recorded. Cutting the cortical tissue above the CC abolished both the cortical evoked potential and the CCEE but spared the CAP, confirming cortical origin. Both hypoxia (not shown) and low-Ca2+ACSF eliminated the CCEE but not the CAP (see FIG. 17, elements B and C). Furthermore, 5 Hz LFS of the CC reduced the amplitude of both the CCEE and CAEP (see FIG. 17, element D). These findingsindicate that the cortex can be “pinged” from the CC, and the resulting echoes report cortical excitability.

[0154] From the above description, those skilled in the art will perceive improvements, changes, and modifications. Such improvements, changes and modifications are within the skill of one in the art and are intended to be covered by the appended claims.

Claims

The following is claimed:

1. A system comprising: at least one stimulating electrode contact configured to be positioned in a fiber tract to apply a stimulation to the fiber tract, wherein the fiber tract has one or more terminals to a cortex of a patient’s brain; at least one recording electrode contact configured to be positioned in the fiber tract and / or the cortex to record a tissue excitability response to the stimulation; a signal generator in communication with the at least one stimulating electrode contact and configured to provide a stimulation waveform to the at least one stimulating electrode contact for the stimulation; and a controller in communication with the signal generator and the at least one recording electrode contact, the controller comprising: a non-transitory memory storing instructions; and a processor configured to access the memory to execute the instructions to at least: receive the tissue excitability response from the at least one recording electrode; identify a parameter of the tissue excitability response; and predict an abnormal activity in the patient’s brain based on the parameter being different from a baseline for the parameter.

2. The system of claim 1 , wherein the fiber tract is at least one of a corpus callosum, a fimbria, a dorsal hippocampal commissure, or a ventral hippocampal commissure.

3. The system of claim 1 , wherein the one or more terminals to the cortex are to cell layers of the cortex.

4. The system of claim 1 , wherein the tissue excitability response is at least one of an evoked potential, a cortico-callosal evoked potential, a corpus callosum evoked echo, or a compound action potential.

5. The system of claim 1 , wherein the at least one recording electrode contact is configured to be positioned in the cortex, wherein the stimulation is a single pulse stimulation at a constant frequency and the tissue excitability response is a cortico-callosal evoked potential and / or an evoked potential, wherein the parameter is an amplitude of the cortico-callosal evoked potential and / or the evoked potential, and wherein the abnormal activity is predicted when the amplitude of the cortico- callosal evoked potential and / or an evoked potential is different from a baseline for the amplitude of the cortico-callosal evoked potential and / or an evoked potential.

6. The system of claim 1 , wherein the at least one recording electrode contact is configured to be positioned in the fiber tract, wherein the stimulation is a single pulse stimulation at a constant frequency and the tissue excitability response is a corpus callosum evoked echo, wherein the parameter is an amplitude of the corpus callosum evoked echo, and wherein the abnormal activity is predicted when the amplitude of the corpus callosum evoked echo is above a baseline amplitude of the corpus callosum evoked echo.

7. The system of claim 1 , wherein the at least one recording electrode is configured to be positioned in the cortex, wherein the stimulation is a paired pulses stimulation and the tissue excitability response comprises a first cortical evoked potential and a second cortical evoked potential, wherein the parameter is the ratio of the second evoked potential over the first evoked potential, and wherein the abnormal activity is predicted when the ratio of the second evoked potential over the first evoked potential changes relative to a baseline ratio at different interspike intervals.

8. The system of claim 1 , wherein the processor further executes the instructions to alert the patient and / or a medical professional when the abnormal activity is predicted.

9. The system of claim 1 , wherein the processor further executes the instructions to provide a therapy to the patient to pre-emptively treat the abnormal activity signal.

10. The system of claim 9, wherein the therapy comprises applying another stimulation through the at least one stimulating electrode contact.11 . The system of claim 9, wherein the system further comprises a therapy delivery device configured to provide a pharmaceutical therapy, a light therapy, a heat therapy, magnetic therapy, or an electrical therapy to the patient in response to the abnormal activity is predicted.

12. The system of claim 1 , wherein the abnormal activity is an incoming seizure and / or a seizure-like state.

13. A method comprising configuring, by a system comprising a processor, a stimulation having a single pulse of a constant frequency; applying, by the system, the stimulation via a stimulating electrode within a fiber tract having one or more terminals in a cortex of a patient’s brain; receiving, by the system, a tissue excitability response to the stimulation from a recording electrode within the fiber tract and / or the cortex; and predicting, by the system, whether an abnormal activity is likely by comparing an amplitude of the tissue excitability response to a baseline of the amplitude, wherein the abnormal activity is predicted when the amplitude is different from the baseline of the amplitude.

14. The method of claim 13, wherein the tissue excitability response is a cortico- callosal evoked potential and / or an evoked potential.

15. The method of claim 13, wherein the tissue excitability response is a corpus callosum evoked echo.

16. The method of claim 13, further comprising when the abnormal activity is predicted alerting, by the system, the patient and / or a medical professional to provide a therapy to the patient to pre-emptively stop the abnormal activity.

17. The method of claim 13, further comprising when the abnormal activity is predicted providing, by the system, an electrical therapy to the patient to preemptively stop the abnormal activity.

18. A method comprising configuring, by a system, a stimulation having paired pulses; applying, by the system, the stimulation to a fiber tract having one or more terminals in a cortex of a patient’s brain; receiving, by the system, a first cortical evoked potential and a second cortical evoked potential in response to the stimulation; and predicting, by the system, if an abnormal activity is likely based on an amount of change of the ratio of the second evoked potential over the first evoked potential compared to a baseline of the ratio of the second evoked potential over the first evoked potential.

19. The method of claim 18, further comprising when the abnormal activity is predicted, alerting, by the system, the patient and / or a medical professional to provide a therapy to the patient to pre-emptively stop the abnormal activity.

20. The method of claim 18, further comprising when the abnormal activity is predicted providing, by the system, another therapy to the patient to pre-emptively stop the abnormal activity.

Images