Method and system for detecting stroke

Abstract

A method for assessing whether a subject has had a stroke and treating the subject accordingly is provided. The method includes recording electroencephalography (EEG) data from electrodes placed on the subject's head. Electrical activity from a particular location on the subject's head is then compared to electrical activity at a contralateral location and the electrical activity measured at all electrodes. These relative electrical activities are used to determine the probability that a stroke occurred in a particular part of the brain, and the subject is treated accordingly.

Description

[Technical Field]

[0001] Introduction Stroke is a devastating disease associated with significant neurological damage and medical costs 1、2 Early detection is crucial for timely treatment with thrombolysis or mechanical thrombectomy, but diagnosis is often delayed or misdiagnosed as a stroke mimic. 3~5 The development of technologies to assist frontline providers in detecting stroke is an important direction for patient care. 1、6 .

[0002] Quantitative electroencephalography (QEEG) may be a useful tool for detecting cerebral ischemia 7 Poorly perfused brain tissue will develop electrical inhibition before reaching the metabolic derangement that causes stroke. 8、9 These changes occur within seconds of loss of cerebral perfusion. 10、11 suggests that EEG could be used to create an early warning system for stroke detection and prevention.

[0003] While such neurologists may exist in large or specialized hospitals, they are rarely immediately available when a patient has a stroke. For example, if a citizen at home or work believes he or she is having a stroke, he or she can contact emergency services, which will dispatch paramedics to the citizen. However, paramedics generally lack the training and knowledge to interpret EEG data for stroke detection. Therefore, the paramedics must perform a stroke assessment without the aid of EEG data. In addition, even after transport to a hospital, a neurologist is not always available at the hospital.

[0004] In other cases, patients experience a stroke while already in the hospital. However, if the patient is unconscious (e.g., asleep or comatose), the patient is unable to alert medical staff that they are feeling unwell. Therefore, medical staff may not be able to perform a stroke evaluation or may not realize that a stroke evaluation should be performed. Additionally, having neurologists continuously monitor patients' EEGs in the hospital for the possibility of a stroke occurring consumes significant resources.

[0005] References: 1.Ferriero DM, Fullerton HJ, Bernard TJ, et al.Management of Stroke in Neonates and Children:A Scientific Statement From the American Heart Association / American Stroke Association.Stroke.2019;50(3).doi:10.1161 / STR.0000000000000183

[0006] 2.Gardner MA,Hills NK,Sidney S,Johnston SC,Fullerton HJ.The 5-year direct medical cost of neonatal and childhood stroke in a population-based cohort.Neurology.2010;74(5):372-378.doi:10.1212 / WNL.0b013e3181cbcd48

[0007] 3. Mackay MT, Monagle P, Babl FE. Improving diagnosis of childhood arterial ischaemic stroke. Expert Review of Neurotherapeutics.2017;17(12):1157-1165.doi:10.1080 / 14737175.2017.1395699

[0008] 4.Mackay MT,Churilov L,Donnan GA,Babl FE,Monagle P.Performance of bedside stroke recognition tools in discriminating childhood stroke from mimics.Neurology.2016;86(23):2154-2161.

[0009] 5.Braun KPJ,Kappelle LJ,Kirkham FJ,Frcpch MB.Diagnostic pitfalls in paediatric ischaemic stroke.Developmental Medicine & Child Neurology.48(12):985-990.doi:10.1017 / S0012162206002167.

[0010] 6.Werho DK,Pasquali SK,Yu S,et al.Epidemiology of Stroke in Pediatric Cardiac Surgical Patients Supported With Extracorporeal Membrane Oxygenation.The Annals of Thoracic Surgery.2015;100(5):1751-1757.doi:10.1016 / j.athoracsur.2015.06.020

[0011] 7.Rosenthal ES,Biswal S,Zafar SF,et al.Continuous electroencephalography predicts delayed cerebral ischemia after subarachnoid hemorrhage:A prospective study of diagnostic accuracy:EEG Accurately Predicts DCI.Ann Neurol.2018;83(5):958-969.doi:10.1002 / ana.25232

[0012] 8. Foreman B, Claassen J. Quantitative EEG for the detection of brain ischemia. Critical Care. 2012;2012(16:216). doi:10.1007 / 978-3-642-25716-2

[0013] 9. Appavu BL, Temkit MH, Foldes ST, et al. Quantitative Electroencephalography After Pediatric Anterior Circulation Stroke. Journal of Clinical Neurophysiology. 2020; Publish Ahead of Print. doi:10.1097 / WNP.0000000000000813

[0014] 10. Kamitaki BK, Tu B, Wong S, Mendiratta A, Choi H. Quantitative EEG Changes Correlate With Post-Clamp Ischemia During Carotid Endarterectomy. Journal of Clinical Neurophysiology. 2021;38(3):213-220. doi:10.1097 / WNP.0000000000000686

[0015] 11. Blume WT, Ferguson GG, McNeill DK. Significance of EEG changes at carotid endarterectomy. Stroke. 1986;17(5):891-897. doi:10.1161 / 01.STR.17.5.891 Summary of the Invention

[0016] A method for assessing whether a subject has had a stroke and treating the subject accordingly is provided. The method includes recording electroencephalography (EEG) data from electrodes placed on the subject's head. Electrical activity from a particular location on the subject's head is then compared with electrical activity at a contralateral location and the electrical activity measured at each electrode. These relative electrical activities are used to determine the probability that a stroke has occurred in a particular part of the brain, and the subject is treated accordingly. Such a method can improve the accuracy and speed of stroke detection.

[0017] In particular, the present method has the advantage of assessing stroke without requiring previous EEG data from the same patient. For example, in some cases, a patient has symptoms of stroke, and paramedics are dispatched to the patient's home to assess the probability of stroke. Therefore, no previous EEG data exists, and the paramedics cannot compare the patient's current electrical activity with normal previous EEG activity from the same patient. However, the current method utilizes a comparison between contralateral locations of the patient's brain, thereby enabling the detection of asymmetric activity. Therefore, the current method has the advantage of not requiring previous EEG data.

[0018] Comparison between contralateral brain locations also offers other advantages. For example, a stroke-free individual who undergoes anesthesia may have suppressed EEG activity that is "abnormal" compared to a healthy, awake individual. Despite having suppressed electrical brain activity, an anesthetized individual must still be evaluated for the presence of a stroke. In this way, comparing one side of the brain to the other provides the ability to detect stroke without relying on comparing the subject's EEG to a "baseline," i.e., a collection of EEG data collected from healthy volunteers. Comparing the left anterior brain to the right anterior brain will reveal asymmetry without the need for a baseline, thereby detecting a stroke. [Brief explanation of the drawings]

[0019] [Figure 1A]Figure 1 shows the raw EEG traces used to calculate COIN for the power band between 4 and 16 Hz on all channels. Pre and post refer to before and after stroke. [Figure 1B] The "After" section shows values ​​from each channel averaged and mapped to the topographic visualizer. The "After" section shows dark blue (i.e., COIN values ​​of -10) in columns Fp1-F7 and Fp1-F3, and light red (i.e., COIN values ​​of +5) in columns Fb-T8 and C4-P4. [Figure 1C] Negative values ​​from the topographic visualizer are added together to produce a summary value. The L side has a COIN value of dark blue (i.e., -4) and the R side has a COIN value of light red (i.e., +2). [Figure 1D] Summary values ​​are calculated every 4 seconds and smoothed using a 5-minute moving average (orange line). [Figure 2-1] Topographic visualization of COINs is shown in order of increasing relative infarct volume (RIV). Row 1 shows neuroimages, row 2 shows topographic visualizations using a full montage, and row 3 shows topographic visualizations using a restricted circumferential montage. Circles in row 1 indicate neuroimages obtained before EEG, and stars in row 1 indicate seizures shown on EEG. The darkest areas are dark blue and correspond to very negative COIN values. [Figure 2-2] Topographic visualization of COINs is shown in order of increasing relative infarct volume (RIV). Row 1 shows neuroimages, row 2 shows topographic visualizations using a full montage, and row 3 shows topographic visualizations using a restricted circumferential montage. Circles in row 1 indicate neuroimages obtained before EEG, and stars in row 1 indicate seizures shown on EEG. The darkest areas are dark blue and correspond to very negative COIN values. [Figure 2-3]Topographic visualization of COINs is shown in order of increasing relative infarct volume (RIV). Row 1 shows neuroimages, row 2 shows topographic visualizations using a full montage, and row 3 shows topographic visualizations using a restricted circumferential montage. Circles in row 1 indicate neuroimages obtained before EEG, and stars in row 1 indicate seizures shown on EEG. The darkest areas are dark blue and correspond to very negative COIN values. [Figure 2-4] Topographic visualization of COINs is shown in order of increasing relative infarct volume (RIV). Row 1 shows neuroimages, row 2 shows topographic visualizations using a full montage, and row 3 shows topographic visualizations using a restricted circumferential montage. Circles in row 1 indicate neuroimages obtained before EEG, and stars in row 1 indicate seizures shown on EEG. The darkest areas are dark blue and correspond to very negative COIN values. [Figure 2-5] Topographic visualization of COINs is shown in order of increasing relative infarct volume (RIV). Row 1 shows neuroimages, row 2 shows topographic visualizations using a full montage, and row 3 shows topographic visualizations using a restricted circumferential montage. Circles in row 1 indicate neuroimages obtained before EEG, and stars in row 1 indicate seizures shown on EEG. The darkest areas are dark blue and correspond to very negative COIN values. [Figure 2-6] Topographic visualization of COINs is shown in order of increasing relative infarct volume (RIV). Row 1 shows neuroimages, row 2 shows topographic visualizations using a full montage, and row 3 shows topographic visualizations using a restricted circumferential montage. Circles in row 1 indicate neuroimages obtained before EEG, and stars in row 1 indicate seizures shown on EEG. The darkest areas are dark blue and correspond to very negative COIN values. [Figure 3A] A summary of COIN values ​​for all control subjects is shown. [Figure 3B] A summary of COIN values ​​for all stroke subjects is shown. [Figure 3C] Summary COIN values ​​are shown along with box plots showing median and interquartile range for control subjects over the first 6 hours of recording. [Figure 3D] Summary COIN values ​​are shown along with box plots showing median and interquartile range for stroke subjects versus relative infarct volume in the first 6 hours of recording. [Figure 4A] Shown are the means and standard errors of the medians from each study comparing controls with all stroke patients, anterior circulation, posterior circulation, patients with strokes greater than 5% relative infarct volume, and patients with strokes greater than 10% relative infarct volume. [Figure 4B] Receiver operating characteristic curves from logistic regression are shown. [Figure 4C] Cut-off values ​​determined using the Youden J statistic (sensitivity + specificity - 1) are shown, and the optimal cut-off value was calculated as the COIN value with the maximum Youden J statistic. [Figure 5A] The time-independent distribution of COIN values ​​is shown. [Figure 5B] All stroke vs. controls for accuracy and AUROC are shown. [Figure 5C] Specificity and sensitivity for all strokes versus controls are shown. [Figure 5D] Shown are RIV vs. control >5% for accuracy and AUROC. [Figure 5E] Specificity and sensitivity of >5% RIV vs. controls are shown. [Figure 6A] COIN values ​​as a function of stroke volume are shown. [Figure 6B] 6B shows sensitivity as a function of specificity of 1 for the data of FIG. 6A. [Figure 6C] Sensitivity, specificity, and sensitivity + specificity - 1 for the data in Figure 6A are shown. [Figure 7A] Shows a 3D graph where qij is a function of rij and sij. The upper region is red, the lower region is blue, and the middle region is gray. [Figure 7B]A three-dimensional graph is shown in which qij is a function of rij and sij, and qij is set to zero when rij multiplied by sij is less than zero. [Figure 8] A flow diagram of EEG consultations for 694 hospitalized patients classified into different populations is shown. [Figure 9A] A topographical visualization of the COIN in a patient during the first period is shown, along with the raw EEG data. The lower axis ranges from 0 to 4 seconds. The vertical axis shows data from 16 different channels. The topographical visualization shows a very faint blue color in the patient's anterior right brain. In particular, the patient's right side is indicated by an "R" and is shown on the left of the figure. [Figure 9B] Data corresponding to the patient in Figure 9A is shown for the second time period. The patient developed a deep blue area in the posterior right side of his brain. Dark areas are deep blue and have very negative COIN values. [Figure 9C] Data corresponding to the patient in Figures 9A and 9B are shown for the third period. A very deep blue area developed in the posterior right portion of the brain, and a very red area developed on the left side of the brain. The R-side area is dark blue and has a very negative COIN value. [Figure 10-1] COIN neuroimaging of a patient's brain, showing topographical COIN visualization in the order of stroke volume, C-value, and ascending stroke volume. L: left, R: right, ACA: anterior cerebral artery, MCA: middle cerebral artery, PCA: posterior cerebral artery, NS: not specified, Vol: stroke volume. The largest and darkest areas are dark blue and have very negative COIN values. Contralateral locations are light red and have slightly positive COIN values. [Figure 10-2] COIN neuroimaging of a patient's brain, showing topographical COIN visualization in the order of stroke volume, C-value, and ascending stroke volume. L: left, R: right, ACA: anterior cerebral artery, MCA: middle cerebral artery, PCA: posterior cerebral artery, NS: not specified, Vol: stroke volume. The largest and darkest areas are dark blue and have very negative COIN values. Contralateral locations are light red and have slightly positive COIN values. [Figure 10-3]COIN neuroimaging of a patient's brain, showing topographical COIN visualization in the order of stroke volume, C-value, and ascending stroke volume. L: left, R: right, ACA: anterior cerebral artery, MCA: middle cerebral artery, PCA: posterior cerebral artery, NS: not specified, Vol: stroke volume. The largest and darkest areas are dark blue and have very negative COIN values. Contralateral locations are light red and have slightly positive COIN values. [Figure 10-4] COIN neuroimaging of a patient's brain, showing topographical COIN visualization in the order of stroke volume, C-value, and ascending stroke volume. L: left, R: right, ACA: anterior cerebral artery, MCA: middle cerebral artery, PCA: posterior cerebral artery, NS: not specified, Vol: stroke volume. The largest and darkest areas are dark blue and have very negative COIN values. Contralateral locations are light red and have slightly positive COIN values. [Figure 10-5] COIN neuroimaging of a patient's brain, showing topographical COIN visualization in the order of stroke volume, C-value, and ascending stroke volume. L: left, R: right, ACA: anterior cerebral artery, MCA: middle cerebral artery, PCA: posterior cerebral artery, NS: not specified, Vol: stroke volume. The largest and darkest areas are dark blue and have very negative COIN values. Contralateral locations are light red and have slightly positive COIN values. [Figure 10-6] COIN neuroimaging of a patient's brain, showing topographical COIN visualization in the order of stroke volume, C-value, and ascending stroke volume. L: left, R: right, ACA: anterior cerebral artery, MCA: middle cerebral artery, PCA: posterior cerebral artery, NS: not specified, Vol: stroke volume. The largest and darkest areas are dark blue and have very negative COIN values. Contralateral locations are light red and have slightly positive COIN values. DETAILED DESCRIPTION OF THE INVENTION

[0020] A method for assessing whether a subject has had a stroke and treating the subject accordingly is provided. The method includes recording electroencephalography (EEG) data from electrodes placed on the subject's head. Electrical activity from a particular location on the subject's head is then compared with electrical activity at a contralateral location and the electrical activity measured at each electrode. These relative electrical activities are used to determine the probability that a stroke occurred in a particular part of the brain, and the subject is treated accordingly.

[0021] Before describing the present invention in more detail, it is to be understood that the invention is not limited to particular embodiments described, as such may vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting, since the scope of the present invention will be limited only by the appended claims.

[0022] Where a range of values ​​is provided, unless the context clearly dictates otherwise, it is understood that each intervening value, to the tenth of the unit of the lower limit, between the upper and lower limit of that range is also specifically disclosed. Each smaller range between any stated value or intervening value in a stated range and any other stated value or intervening value in that stated range is encompassed within the invention. The upper and lower limits of these smaller ranges may independently be included or excluded in the range, and each range where either limit is included in the smaller range, neither limit is included in the smaller range, or both limits are included in the smaller range is also encompassed within the invention, subject to any specifically excluded limit in the stated range. Where a stated range includes one or both of the limits, ranges excluding either or both of those included limits are also included in the invention.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Although any methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, some possible exemplary methods and materials may be described herein. Any and all publications mentioned herein are incorporated by reference to disclose and describe the methods and / or materials in connection with which the publications are cited. It is understood that the present disclosure supersedes any disclosure of an incorporated publication to the extent there is a conflict.

[0024] It should be noted that, as used herein and in the appended claims, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, a reference to "a droplet" includes a plurality of such droplets; a reference to a "separate entity" includes a reference to one or more separate entities, etc. It should be further noted that the claims may be drafted to exclude any element, e.g., any optional element. Accordingly, this statement is intended to serve as a predicate for use of exclusive terminology such as "solely," "solely," and the like, or the use of a "negative" limitation in connection with the recitation of claim elements.

[0025] The publications discussed herein are provided solely for their disclosure prior to the filing date of the present application. Further, the dates of publication provided may be different from the actual publication dates, which may need to be independently confirmed. To the extent that the definition or usage of any term herein conflicts with the definition or usage of that term in an application or reference incorporated herein by reference, the present application shall control.

[0026] As will be apparent to those skilled in the art upon reading this disclosure, each of the individual embodiments described and illustrated herein has individual components and features which may be readily separated from or combined with the features of any of the other several embodiments without departing from the scope or spirit of the invention. Any recited method can be carried out in the order of events recited or in any other order which is logically possible.

[0027] definition The terms "subject" and "patient" are used interchangeably herein to refer to an animal, such as a human.

[0028] The terms "determining," "measuring," and "assessing" are used interchangeably herein.

[0029] method Methods are provided for assessing whether a subject has had a stroke and treating the subject accordingly. In some examples, the methods include: (a) recording electroencephalography (EEG) data from electrodes placed at scalp locations on the subject's head; (b) determining relative electrical activity from the EEG data, wherein determining determining the relative electrical activity at each scalp location compared to the electrical activity at all scalp locations; determining relative electrical activity at each scalp location compared to electrical activity at its corresponding contralateral scalp location; (c) determining a high probability of stroke based on the determined relative electrical activity; and (d) treating a subject for a high probability of stroke.

[0030] Recording EEG data As described above, an exemplary method includes recording electroencephalography (EEG) data by measuring electrical activity with electrodes placed at scalp locations on a subject's head. The term "electrogram" (EGM) is used herein to refer to the recording of electrical activity. While the electrodes are placed on the subject's scalp, the recorded EEG data corresponds to brain activity beneath such scalp locations.

[0031] In some cases, the step a) of recording EEG data includes the substeps of i) recording electrical activity from electrodes and ii) generating EEG data from the recorded electrical activity.

[0032] Thus, electrical activity is recorded by actual electrodes on the subject's head. However, EEG data can be categorized by its "channels," with one channel per scalp location.

[0033] In some cases, a channel of EEG data is generated from electrical activity recorded by a single electrode. For example, there may be four electrodes placed at four scalp locations. These four electrodes record four sets of electrical activity, and the four sets of electrical activity are four channels of EEG data corresponding to the four scalp locations. For example, there may be an "Fp1" electrode and an "F7" electrode.

[0034] In other cases, a channel of EEG data is generated from electrical activity recorded by a first electrode, electrical activity recorded by a second electrode, and optionally electrical activity recorded by additional electrodes. Typically, the first and second electrodes are placed close to each other, e.g., less than 5 cm apart, such as less than 4 cm, less than 3 cm, less than 2 cm, less than 1 cm, or less than 0.5 cm. By using two (or more) electrodes to create a specific channel, signals from adjacent electrodes can be used to cancel any distant physiological electrical activity, for example, from the heart. For example, electrical activity recorded by a first electrode can be subtracted from electrical activity recorded by a second electrode. Thus, brain electrical activity near a specific scalp location can be estimated by recording electrical activity with two electrodes near or at the scalp location and then combining the two recordings to generate a specific channel of EEG data corresponding to the specific scalp location. If such electrodes are designated "Fp1" and "F7", the corresponding channels may be designated "Fp1-F7".

[0035] At least one pair of scalp locations are "contralateral" to one another. A patient's head may be referred to as being bisected by a median plane that separates the right and left sides of the subject's head. The term "contralateral" is used herein to refer to a scalp location that is on the opposite side of the median plane from another scalp location. Stated another way, if a first scalp location is contralateral to a second scalp location, reflecting the first scalp location through the median plane will arrive at the second scalp location. The term "median plane" is used interchangeably with "median plane" and "midsagittal plane."

[0036] In some cases, there are four or more scalp locations, for example, including two pairs of contralateral scalp locations. In some embodiments, there are six or more scalp locations, for example, including three pairs of contralateral scalp locations. In some cases, there are eight or more scalp locations and four or more pairs, or ten or more scalp locations and five or more pairs. In some cases, an EEG data channel is generated from a single electrode at a scalp location. In some cases, an EEG data channel is generated from two or more electrodes at or near a scalp location.

[0037] For simplicity, the following embodiments may describe cases where a single electrode recording is used to generate an EEG data channel for a particular scalp location, however, it is understood that each embodiment may be modified so that the EEG data channel for a particular scalp location originates from multiple electrodes.

[0038] In some cases, EEG data recording is performed for 1 to 25 seconds, such as 2 to 10 seconds. In some cases, EEG data is recorded for 4 seconds. These periods of recording are then used to determine reference and control electrical activity.

[0039] Determination of reference electrical activity An exemplary method includes determining the relative electrical activity at each scalp location compared to the electrical activity at all scalp locations, also referred to herein as determining "reference electrical activity" or "reference electrical activity." For example, if there are four total electrodes, the electrogram from the first electrode can be compared to the electrogram from the second, third, and fourth electrodes. Additionally, the electrogram from electrode 2 can be compared to the electrogram from electrodes 1, 3, and 4; the electrogram from electrode 3 can be compared to the electrogram from electrodes 1, 2, and 4; and the electrogram from electrode 4 can be compared to the electrogram from electrodes 1, 2, and 3.

[0040] In some cases, such determination of the reference electrical activity involves determining the average (i.e., mean) of the electrical activity from all of the electrodes. Thus, the reference electrical activity of the first electrode involves comparing the raw electrical activity of the first electrode to the average electrical activity of all of the electrodes.

[0041] Determination of symmetric electrical activity Additionally, exemplary methods include determining the relative electrical activity at each scalp location compared to the electrical activity at its corresponding contralateral scalp location, also referred to herein as determining "symmetrical electrical activity." For example, in an exemplary embodiment, electrode 1 is contralateral to electrode 2, and electrode 3 is contralateral to electrode 4. Determining the symmetrical electrical activity at electrode 1 involves comparing the raw electrical activity at electrode 1 to the raw electrical activity at its contralateral electrode, electrode 2. Similarly, the symmetrical electrical activity at electrode 2 is determined by the activity at electrode 2 compared to the activity at electrode 1. Additionally, the symmetrical electrical activity at electrode 3 is determined by comparing electrode 3 to electrode 4, and the symmetrical electrical activity at electrode 4 is determined by comparing the activity at electrode 3 to electrode 4.

[0042] In some embodiments, determining the symmetric electrical activity at an electrode comprises dividing the electrical activity at an electrode by the electrical activity of its contralateral electrode.

[0043] Determining the probability of stroke An exemplary method includes determining a probability that the subject has experienced a stroke based on the determined relative electrical activity. Stated another way, the method includes determining a probability that the subject has experienced a stroke based on the determined reference electrical activity and the determined symmetric electrical activity.

[0044] As described above, the method involves determining reference electrical activity for at least one scalp location. Similarly, the method involves determining symmetric electrical activity for at least one scalp location. Thus, the probability of stroke can be determined by considering one scalp location at a time. For example, to determine the probability of stroke at the location of electrode 1, the reference activity at electrode 1 can be compared to the symmetric activity at electrode 1. In some embodiments, the determination of stroke probability is repeated for at least one other scalp location (e.g., electrode 2), for example, for all scalp locations. Thus, determining the probability of stroke can include determining the probability of stroke at each scalp location corresponding to each electrode. Thus, not only can the method determine the probability of stroke, but the method can also advantageously determine which region of the brain is most likely to have experienced a stroke.

[0045] In some cases, the determination is a determination of ischemic stroke, whereby a blockage of a blood vessel in the brain or neck causes a temporary or permanent cessation of blood flow to a portion of the brain. In some cases, the determination is a determination of hemorrhagic stroke, whereby a rupture of a blood vessel in the brain or skull causes bleeding to occur in the brain or skull. Ischemic or hemorrhagic stroke can occur spontaneously or can result from any cause, including, but not limited to, traumatic brain injury, embolism (whereby material such as a blood clot, air, fat, or foreign body travels through the bloodstream to the brain and lodges in a blood vessel), atherosclerosis, cerebral aneurysm, or inflammatory disease (cancer, autoimmune, or infectious disease). In some cases, the determination is a determination of any disease affecting an area of ​​the brain that may cause differences in electrical activity, such as a brain tumor or infection (such as an infected abscess).

[0046] Targeted treatment for high rates of stroke If the determined probability of stroke is high, the exemplary method further includes treating the subject for the high probability of stroke. By determining the risk of stroke, an earlier step in the method provides the advantage of a more accurate diagnosis, thereby assisting in the triage and treatment of potential stroke patients. In some embodiments, a high probability of stroke means that the estimated probability of stroke is 1% or higher, such as 5% or higher, 10% or higher, 25% or higher, or 50% or higher. In some embodiments, a high probability of stroke means that a medical professional determines that the subject should be treated for stroke, for example, by performing additional stroke detection measurements or by medical procedures to minimize or reverse the effects of a potential stroke.

[0047] In some cases, treatment for a high probability of stroke includes performing additional stroke detection measurements, such as a brain magnetic resonance imaging (MRI) or a head computed tomography (CT) scan. For example, such measurements may be performed in a hospital.

[0048] In some cases, such treatment includes medical procedures to minimize or reverse the effects of a potential stroke. In some cases, such treatment includes medical or surgical intervention to reverse or minimize the effects of a potential stroke. For example, brain surgery can be performed to control bleeding during a hemorrhagic stroke or to correct a lack of blood flow in an ischemic stroke.

[0049] Before recording EEG data In some cases, before recording the EEG data, the subject had an elevated risk of stroke. In other words, the method is performed because the subject had an elevated risk of stroke. For example, in some cases, the elevated risk of stroke is selected from the group consisting of altered mental status, loss of motor function, loss of tactile sensation in body parts, dizziness, headache, and difficulty speaking. In some embodiments, the elevated risk of stroke is selected from the group consisting of head trauma, hematoma (e.g., epidural, subdural), and subarachnoid hemorrhage. In some cases, the elevated risk of stroke is selected from the group consisting of currently undergoing surgery (e.g., on the brain, heart, or blood vessels) and having undergone surgery within the past 30 days.

[0050] In some embodiments, the subject is in a medical facility (e.g., a hospital) at the start of recording the EEG data. For example, the subject may be undergoing surgery, or the subject may be recovering after surgery. In some embodiments, the subject has recently been admitted to a medical facility, e.g., within the past 24 hours, due to an elevated risk of stroke, e.g., head injury or dizziness.

[0051] In some cases, the subject is outside a medical facility at the start of recording the EEG data. For example, a medical professional (e.g., emergency medical personnel) can be dispatched to contact and evaluate a patient with symptoms consistent with a stroke, and the method can help improve the accuracy of the diagnosis. In such cases, treating the subject for a high probability of a stroke can include transporting the subject to a medical facility, notifying the medical facility of a high probability that the subject has experienced a stroke, or a combination thereof.

[0052] Iterating if initial evaluation does not detect a high probability In some cases, the method includes recording EEG data, determining relative electrical activity from the EEG data, and determining a probability of stroke. If the determined probability of stroke is high, the method includes treating the subject for the high probability of stroke.

[0053] However, if the determined probability of stroke is not high (e.g., low), the method can include repeating the recording step and the two determining steps. Stated another way, EEG data can be continuously recorded, and the two determining steps (labeled b and c above) can be continuously repeated accordingly. Thus, the method can also be referred to as a method of continuously monitoring a subject for stroke. Continuous monitoring can be performed for 1 minute or more, such as 10 minutes or more, 30 minutes or more, 1 hour or more, 4 hours or more. Additionally, the recording step and the two determining steps can be repeated at least once per hour, such as at least once every 10 minutes, at least once per minute, at least once per 10 seconds, or at least once per second.

[0054] Determining electrical activity at different periods In some cases, the probability of stroke is assessed independently at each different time point. For example, data can be recorded for a single period (e.g., 4 seconds), and this single period can be used to determine the probability of stroke. This situation can be referred to as a first period and a second period.

[0055] In other cases, the probability of stroke is assessed by considering readings from multiple consecutive time periods. For example, data may be recorded over a first time period (e.g., 4 seconds), then data may be recorded over a second time period (e.g., another 4 seconds). In such cases, the stroke determination is based on a combination of data from the first and second time periods. Such a combination can help to "smooth" the data and reduce the effects of random error, for example, by using a "moving average." In some cases, each time period of recorded data ranges from 0.5 seconds to 15 seconds, and each stroke determination is made using data from between two and 200 time periods. In some cases, the stroke determination is made by averaging the r(f,j) and s(f,j) values ​​from different time periods. In some cases, the stroke determination is made by averaging the g(f,j) values ​​from different time periods. In some cases, the stroke determination is made by averaging the m(j) values ​​from different time periods.

[0056] In some cases, the method comprises: repeating the recording of EEG data for one or more additional time periods; determining relative electrical activity from the EEG data from one of the one or more additional time periods; generating a combined relative electrical activity by combining the relative electrical activity from the EEG data from the first time period and the relative electrical activity from the EEG data from one or more additional time periods; and determining a likelihood of stroke based on the combined relative electrical activity.

[0057] Such methods can help improve data quality by reducing the effects of random errors or reading variations.

[0058] For example, in some cases, one or more additional periods comprise a second period. Thus, the combined relative electrical activity is based on a combination of only the first and second periods. As another example, if two additional periods are present, the stroke probability is assessed based on a total of three periods. Thus, in some cases, there are one or more additional periods, such as 2 or more, 3 or more, 4 or more, 5 or more, 10 or more, 20 or more, 30 or more, or 50 or more. In some cases, each period ranges from 1 second to 5 minutes, such as 2 seconds to 20 seconds.

[0059] In some embodiments, combining the relative electrical activity from different time periods includes averaging the relative electrical activity over time.

[0060] Mathematical aspects of determining relative electrical activity In some cases, determining the relative electrical activity from the EEG data includes generating a matrix from the EEG data. As used herein, the term "matrix" refers to an array of numbers.

[0061] In some embodiments, the method includes generating a matrix A from the EEG data, where matrix A includes elements a(t,j), where a(t,j) refers to the amplitude of electrical activity at time t and scalp location j. Stated another way, an electrode at a first scalp location (i.e., j=1) can record electrical activity in either analog or digital form. If initially recorded in analog, the initial analog data is converted to digital form. The EEG data can be recorded for any suitable period of time, e.g., from 1 second to 10 minutes. The EEG data can be recorded at any suitable frequency, e.g., from once every 10 milliseconds to once every 10 seconds.

[0062] Thus, matrix A is a two-dimensional matrix, with the first dimension corresponding to the time over which the data is recorded, the second dimension corresponding to the identity of the channel, e.g., derived from a single electrode or multiple electrodes, and the value at a particular matrix location being the amplitude of the electrical activity.

[0063] In some embodiments, the method further includes generating a matrix B from matrix A, where matrix B includes elements a(f,j), where a(f,j) refers to the power of electrical activity at frequency f and scalp location j. For example, applying a Fourier transform to matrix A (which lists data as a function of time) can be used to generate matrix B (which lists the intensity of electrical activity as a function of frequency). Thus, converting matrix A to matrix B can be used to convert raw data recorded from a channel into processed data indicating the relative intensity of brain waves at different frequencies. For example, in some cases, the frequencies in matrix B can range from 2 Hz to 18 Hz. In some cases, matrix B includes rows for frequencies in 1 Hz intervals, e.g., matrix B includes rows corresponding to 4 Hz, 5 Hz, 6 Hz, 7 Hz, 8 Hz, 9 Hz, 10 Hz, 11 Hz, 12 Hz, 13 Hz, 14 Hz, 15 Hz, and 16 Hz. For example, if EEG data is recorded from four channels and 12 frequency ranges are used, then matrix B will have the size of a 4 x 12 matrix. In some cases, a single frequency range is used. In other cases, two or more frequency ranges are used, such as four or more, eight or more, or twelve or more.

[0064] In some embodiments, the matrix B is used to generate the reference matrix R and the symmetric matrix S.

[0065] The reference matrix R includes elements r(f,j), where r(f,j) refers to the relative power of electrical activity at frequency f and scalp location j compared to the power of electrical activity at frequency f at all scalp locations. As mentioned above, such determination of the reference electrical activity may involve determining the average (i.e., mean) of the electrical activity from all channels. The average electrical activity from all channels can be described by the following equation:

number

[0066] As described by the above equation, if different frequency ranges (eg, 4 Hz, 5 Hz...15 Hz, 16 Hz) are used, the calculations are performed separately for the different frequency ranges.

[0067] In some cases, determining each r(f,j) includes dividing each a(f,j) by the average of all electrical activity. In some embodiments, each r(f,j) is determined according to the following equation:

number

[0068] The symmetric matrix S contains elements s(f,j), where s(f,j) refers to the relative power of electrical activity at frequency f and scalp location j compared to the power of electrical activity at scalp location j* contralateral to frequency f and scalp location j.

[0069] As noted above, in some cases, determining s(f,j) involves dividing the electrical activity at location j by the electrical activity at its contralateral location j*. In some cases, determining s(f,j) involves additional mathematical operations, such as determining s(f,j) according to an equation.

number

[0070] Thus, in some embodiments of the method, determining the reference electrical activity and symmetric electrical activity includes: (i) generating a matrix A from the EEG data, where matrix A describes amplitude over time; (ii) generating a matrix B from matrix A, where matrix B describes power over one or more frequency ranges; (iii) generating a reference matrix R by comparing the electrical activity from a channel with all channels; and (iv) generating a symmetric matrix S by comparing the electrical activity from a channel with its contralateral channel.

[0071] Mathematical aspects of determining the probability of stroke After generating the reference electrical activity and the control electrical activity, the method includes determining the probability that the subject has experienced a stroke.

[0072] Once the matrices R and S have been generated, determining the probability of stroke can be done by: generating a gating matrix G containing elements g(f,j) by combining matrix R with matrix S; generating a map matrix M containing elements m(j), where m(j) is generated by averaging g values ​​for the same scalp location j across different frequencies f; determining a probability of stroke based on the map matrix M.

[0073] In some cases, each g(f,j) is calculated according to the following equation: g(f,j)=r(f,j) 2 s(f,j)

[0074] In some cases, after calculating such g(f,j) values, the G matrix is ​​further modified. For example, in some cases, each g(f,j) value is changed to zero if its corresponding r(f,j) value has the opposite sign to its corresponding s(f,j) value. For example, for a frequency of 4 Hz and channel 1, if r(4,1) is negative but s(4,1) is positive, g(4,1) is changed to zero. Similarly, if r(4,1) is positive but s(4,1) is negative, g(4,1) is changed to zero. In some cases, this operation avoids the generation of spurious signals at scalp locations where the symmetric signal is low but the reference signal is high, and vice versa.

[0075] As mentioned above, m(j) is generated by averaging the g values ​​for the same scalp location j over different frequencies f. This mathematical operation can also be expressed by the following equation:

number

[0076] In some cases, a high probability of stroke is determined if one or more m(j) values ​​are within a predetermined threshold range, e.g., if one or more m(j) values ​​are less than −10, e.g., less than −15, less than −20, less than −25, less than −30, less than −35, or less than −40.

[0077] Additionally, the value "C" is the sum of all negative m(j) values, as shown in the following equation:

number

[0078] In some cases, a high probability of stroke is determined if the sum of all negative m(j) values ​​(i.e., value C) is within a predetermined threshold range, for example, less than −10, less than −15, less than −20, less than −25, less than −30, less than −35, or less than −40. In some cases, the controller is configured to detect whether the stroke size is greater than or less than 100 ml with a sensitivity of 80% or greater and a specificity of 85% or greater.

[0079] Stroke size In some embodiments in which a stroke is detected, the method further includes determining the size of the stroke. For example, a stroke can be considered large if the value of C is less than -20. In some cases, a large stroke corresponds to a volume of 100 ml or more, and a small stroke corresponds to a volume of less than 100 ml. In some cases, the size of the stroke is determined to be 100 ml or more based on a C value of -20 or less, and the size of the stroke is determined to be less than 100 ml based on a C value of greater than -20.

[0080] controller Aspects of the present disclosure include a controller for assessing whether a subject has had a stroke and communicating the results of the assessment. In some cases, the controller: a) acquiring electroencephalography (EEG) data from electrodes placed at scalp locations on the subject's head; b) determining and calculating relative electrical activity from EEG data; determining the relative electrical activity at each scalp location compared to the electrical activity at all scalp locations; determining relative electrical activity at each scalp location compared to electrical activity at its corresponding contralateral scalp location; c) determining the probability that the subject experienced a stroke near one or more scalp locations based on the relative electrical activity; d) electronically instructing the communication device to communicate the determined probability.

[0081] EEG data is recorded from three or more channels at three or more scalp locations, e.g., four or more channels at four or more scalp locations, e.g., six or more channels at six or more scalp locations, eight or more, ten or more, twelve or more, or fourteen or more. Each channel is located at a single scalp location. As noted above, each channel of EEG data can come from a single electrode or multiple electrodes, but in each case, the particular channel corresponds to a particular scalp location.

[0082] At least one pair of channels is positioned "contralateral" to one another. The patient's head may be said to be bisected by a median plane separating the right and left sides of the subject's head. The term "contralateral" is used herein to refer to a scalp location that is on the opposite side of the median plane from another scalp location. Stated another way, if a first scalp location is contralateral to a second scalp location, then reflecting the first scalp location through the median plane will reach the second scalp location. Thus, in some cases, there are two or more pairs of contralateral channels, for example, three or more pairs, four or more pairs, five or more pairs, six or more pairs, or seven or more pairs.

[0083] Determining the relative electrical activity from the EEG data (i.e., step b) can, in some cases, be performed using the techniques and formulas described above with respect to the methods. Similarly, determining the probability that the subject has experienced a stroke near one or more scalp locations based on the relative electrical activity (i.e., step c) can, in some cases, be performed using the techniques and formulas described above with respect to the methods.

[0084] As noted above, step d) of the procedure performed by the controller involves electronically instructing the communication device to communicate the determined probability.

[0085] In some cases, communicating the determined probabilities includes providing a visual image indicative of one or more of the determined probabilities. For example, the visual image may be displayed by a light emitting diode display, such as a computer monitor or television.

[0086] In some embodiments, the image includes a symbol representing a top view of the subject's head. In some cases, the image uses different colors to indicate different relative probabilities that a stroke has occurred near different scalp locations. In some embodiments, the different colors include a gradient between three or more colors. For example, the gradient can be from blue to white to indicate a gradient of high to low electrical hypoactivity (e.g., stroke). In some cases, the gradient is from red to white to indicate a probability of high to low electrical hyperactivity (e.g., seizure).

[0087] In some cases, the controller is further configured to electronically instruct the alert device to provide a visual alert, an audible alert, or a combination thereof when the determined probability near one or more scalp locations is within a predetermined threshold range. For example, the alert device may be a speaker that generates an audible alert, such as a loud buzzer. In some cases, the alert device provides a visual alert, such as a flashing light. In some cases, the visual alert is provided by a visual notification on a screen, such as a computer monitor, for example, the visual notification flashes repeatedly to attract the attention of a user, such as a nurse. In some cases, the alert (visual, audible, or both) is provided to a wearable device, such as a personal notification device (e.g., a pager) that a nurse can wear while traveling through a healthcare facility.

[0088] In some cases, the controller is configured to repeat steps a), b), c), and d). If the controller provides an alert, the controller may also provide the alert repeatedly. In some embodiments, the controller repeats the steps within 10 minutes, such as within 5 minutes, within 2 minutes, within 1 minute, or within 10 seconds. Thus, the controller may be referred to as repeatedly "monitoring" for possible stroke by repeatedly performing steps a) through d).

[0089] The controller may be configured to detect ischemic stroke, hemorrhagic stroke, or a combination thereof. The controller may be configured to determine stroke size, as discussed above with respect to the method, for example, by setting an alert based on a threshold value for C. Thus, the controller can advantageously detect even relatively small strokes, e.g., less than 100 ml, such as between 25 ml and 100 ml, by setting a threshold value for C of -10. The controller may also be configured to detect only large strokes, e.g., greater than 100 ml, such as between 100 ml and 300 ml, by setting a threshold value for C of -30.

[0090] The controller can achieve stroke detection with high sensitivity and specificity. Such terms are discussed in the art by, for example, Sheffler and Huecker (“Diagnostic Testing Accuracy: Sensitivity, Specificity, Predictive Values ​​and Likelihood Ratios”) and Power et al (“Principles for high-quality, high-value testing”, BMJ Evidence-Based Medicine, doi:10.1136 / eb-2012-100645). The term “accuracy” is also discussed in the art.

[0091] Such terms are described herein based on the following equation:

number

[0092] The controller of the present invention can advantageously detect stroke with both high sensitivity and high selectivity. Therefore, the rates of false positives and false negatives are relatively low. Increasing or decreasing the threshold used can favor either sensitivity or specificity. In some cases, the controller can detect stroke with 50% or greater sensitivity and 70% or greater specificity using EEG data collected for 6 hours or less, e.g., 4 hours or less, 2 hours or less, 30 minutes or less, or 5 minutes or less. In some cases, the controller can detect stroke with 70% or greater sensitivity and 90% or greater specificity using EEG data collected for 6 hours or less, e.g., 4 hours or less, 2 hours or less, 30 minutes or less, or 5 minutes or less. Such detection can be performed using EEG data collected for 6 hours or less, e.g., 4 hours or less, 2 hours or less, 30 minutes or less, or 5 minutes or less. For example, a predetermined threshold of -15 or -20 for C can be used.

[0093] Some strokes are very small and therefore difficult to detect. However, strokes affecting larger areas of the brain are more dangerous but easier to detect. Thus, in some cases, the controller can detect strokes with an infarct volume of 5% or more with a sensitivity of 75% or more and a specificity of 80% or more. "Infarct volume" refers to the percentage of the brain affected by a stroke. In some cases, the controller can detect strokes with an infarct volume of 5% or more with a sensitivity of 90% or more and a specificity of 90% or more. Such detection can be performed using EEG data collected for 6 hours or less, such as 4 hours or less, 2 hours or less, 30 minutes or less, or 5 minutes or less. For example, a predetermined threshold of -15 or -20 for C can be used.

[0094] Such sensitivity and specificity may, in some cases, be higher than that of one skilled in the art who simply considers the raw EEG data itself.

[0095] system Also provided is a system for assessing whether a subject has had a stroke and communicating a result of the assessment, the system comprising: a controller as described above; and one or more of a communication device, an alert device, and an electrode.

[0096] kit Also provided is a kit for assessing whether a subject has had a stroke and communicating the results of the assessment, the system comprising: a controller as described above; and a package including a controller.

[0097] In some cases, the kit further comprises one or more of a communication device, an alert device, and electrodes included in the package. [Example]

[0098] The following examples are presented to provide those of ordinary skill in the art with a complete disclosure and description of how to make and use the present invention, and are not intended to limit the scope of what the inventors regard as their invention, nor are they intended to represent that the following experiments are all or the only experiments performed. Efforts have been made to ensure accuracy with respect to numbers used, but some experimental error and deviation should be accounted for.

[0099] Example 1: Recognizing Ischemic Stroke in Children overview EEG data were recorded for three different pediatric patient populations: patients with confirmed stroke, healthy "control" patients without a diagnosed stroke, and "control" patients without a diagnosed stroke but with certain other conditions. Figure 8 shows a flow diagram of the EEG consultations for 694 hospitalized patients classified into different populations. The data were mathematically analyzed to generate a COIN (Correlate of Injury to the Nervous System) score, which was used to determine the likelihood of ischemic stroke in children. This study demonstrates that a continuous, quantitative measure of cerebral asymmetry derived from scalp EEG data has high sensitivity and specificity for detecting acute stroke with a relative infarct volume greater than 5% in hospitalized children.

[0100] Study population Stroke subjects were children admitted to a pediatric hospital with neuroimaging-confirmed stroke, and the control group had either normal imaging or a normal neurological examination. Both groups included children aged 28 days to 18 years. Stroke in Pediatric Stroke (SIPS II) 13 Database subjects were sampled from multiple institutions, and matched control children were consecutively sampled from the University of California, San Francisco (UCSF) Benioff Children's Hospital (BCH) after applying inclusion criteria.

[0101] Inclusion criteria for both groups were initiation of EEG within 48 hours of stroke or hospitalization and a minimum continuous recording period of 6 hours. Control patients were excluded for a premorbid diagnosis of epilepsy, neurological malignancy, previous stroke, neonatal hypoxic-ischemic encephalopathy, or a concurrent diagnosis of cardiac arrest, acute renal dysfunction, liver, diabetic ketoacidosis, toxin ingestion, syndromic disease, or genetic abnormality. Data on demographic variables, medications, qualitative EEG findings, and neuroimaging findings, including relative infarct volume (RIV, calculated as infarct volume to total brain volume) of stroke patients, were extracted from the SIPS II case report form.

[0102] EEG processing COINs are quantitative features derived from raw EEG tracings. Figures 1A-1D show the EEG preprocessing, artifact identification, and analysis pipeline. Raw EEG files were acquired from the SIPS II study (AIS) or the institution's EEG server. All files were converted to the standard European data format (.EDF) and anonymized to hash patient names, IDs, and offset study start dates. Raw EEG data were sampled at a minimum of 200 Hz and recorded from 19 monopolar channels.

[0103] Data, FieldTrip 14 The data were preprocessed in MATLAB using [SEQ ID NO: 1] and band-pass filtered from 1 to 70 Hz with a 60 Hz notch filter. Montage was applied, dividing the study into 4-second epochs. Artifact detection and rejection were performed using the Fieldtrip algorithm to detect clipping artifacts (amplitude threshold 0.05 uV) and threshold artifacts (>300 uV) with z-score rejection for muscle artifacts (50-99 Hz, z-threshold 4) and movement artifacts (1-5 Hz, z-threshold 4). Epochs containing artifacts longer than 2.5 seconds were completely rejected, and epochs containing artifacts shorter than 2.5 seconds were partially rejected. 14 The input EEG data was passed through a fast Fourier transform to generate power spectral data for each 4-second epoch for each channel in a time-central-parasagittal (TCP) montage. To analyze the data using a restricted 10-lead circumferential montage, the EEG processing pipeline was re-run with the parasagittal channel removed before artifact identification and rejection.

[0104] Calculation of the Correlation of Injury to the Nervous System (COIN) Index The raw EEG data is passed through a Fast Fourier Transform to generate a power spectrum matrix A, where frequency i=1, 2, ..., L represents the L number of power bands examined, and channel j=1, 2, ..., M represents the M number of channels examined. Thus, a ij represents the power for frequency band i on channel j.

[0105] Note that frequency i is also referred to herein as frequency f.

[0106] The letter "a" with subscript "j" may refer to the electrode / channel at position "j." Additionally, the letter "a" with subscript "k" is used interchangeably herein with the letter "a" with subscript "j*," which refers to the position contralateral to the electrode / channel at position "j."

[0107] The R and S matrices are calculated (i and ii) and passed through a cubic function using element-wise matrix multiplication to produce the COIN matrix Q(iii).

number

[0108] Q passes through a gating function to ij and s ij Remove instances where has opposite polarity (iv).

[0109] Figure 7A shows the q ij r ij and s ij Figure 7B shows a three-dimensional graph of q ij r ij and s ij is a function of r ij to s ij When the number multiplied by is less than zero, q ij is set to zero, i.e., to show a three-dimensional graph as shown in equation (iv) above.

[0110] Figures 1A-1D show sample 4-second epochs of EEG (A) used to generate the COIN matrix (B). The mean values ​​within each channel are mapped to their corresponding channel locations on a topographic representation of the head to generate the visualizer (C). All channels with negative values ​​are summed to generate a summary value, C, for each 4-second epoch. The values ​​for C are smoothed over time using a 5-minute moving average (D).

number

[0111] BSI and ADR calculations The calculation of BSI was adapted from vanPutten et al., (12) by applying the following equation to the power spectrum matrix:

number

[0112] where R and L are the right and left channels, respectively. The calculation for ADR was adapted using the front channels (F7-T7, F8-T8) and rear channels (P7-O1, P8-O2) as described by Classen et al. 8、16 The channels were selected on a temporal chain for the findings to be applicable using a restricted circumferential montage. For the analysis, we used the absolute value of the difference between the ADR calculated on the left and right.

[0113] ADR—the power ratio between the alpha range (8–13 Hz) and the delta range (1–4 Hz)—is a useful tool for monitoring cerebral ischemia because it increases delta power and relatively attenuates alpha power under ischemic conditions. ADR calculations were performed using anterior channels (F7–T7, F8–T8) and posterior channels (P7–O1, P8–O2) as described by Claassen et al. (2004; Rosenthal et al., 2018). Channels were selected along the temporal chain so that our findings could be applied using portable circumferential EEG devices. For analysis, the absolute value of the difference between the ADRs calculated on the left and right sides was used.

[0114] Data processing and statistical analysis IRB approval was obtained for this project, and all EEGs were de-identified prior to analysis. Raw EEG preprocessing and artifact rejection were performed in MATLAB (MathWorks, Natick, MA) using Fieldtrip. 14 Statistical analysis with two-tailed t-tests for statistical significance and logistic regression for receiver operating characteristic (ROC) analysis were performed in MATLAB using the Statistics Toolbox. Time window analysis was accomplished in Python using a random forest classifier (RFC) model.

[0115] The first 6 hours of EEG data for each subject were used for statistical analysis, and the median COIN value over the first 6 hours of recording was used as the individual data point for statistical comparison. Representative neuroimaging (CT or MRI) was compared with topographic visualization of COIN. A two-tailed t-test was used to compare continuous variables and logistic regression to derive the area under the receiver operating characteristic (AUROC) curve. The Youden J statistic (sensitivity + specificity - 1) was calculated to determine the ideal cutoff point for optimizing test performance. Sensitivity, specificity, and test accuracy were calculated by applying the optimal cutoff point to the median COIN value from each recording. Analysis was performed in MATLAB 2021b using the Statistics Toolbox.

[0116] One of the secondary goals of this study was to determine the amount of EEG data required for stroke prediction using COIN. First, we created a randomly sampled 5-minute clip from each subject's EEG recording. For a range of COIN values ​​between -5 and -35 for this 5-minute epoch, we generated a vector containing the percentage of time spent below the COIN value. The vectors from each stroke and control subject were compiled to generate a matrix that was used as input data for a random forest classifier (RFC) model to predict stroke diagnosis (Python version 3.9.1 and scikit-learn) (Pedregosa et al., 2011). This process was repeated 200 times, using a different randomly selected time clip from each patient per iteration. Summary performance data (sensitivity, specificity, accuracy, and area under the receiver operating curve [AUROC]) were compiled. This process was systematically repeated using clips of increasing length, starting with a 5-minute duration and going up to 30 minutes (i.e., the first 200 RFC models were EEG clips each with a 5-minute duration, the following 200 models with a 6-minute length, etc., until a 30-minute clip was reached). This procedure was performed to assess how model performance changed as longer clips of EEG data were used. The mean and standard deviation of accuracy, AUROC, sensitivity, and specificity were reported for each time window.

[0117] result Ninety patients from the SIPS cohort were screened, and 23 met the inclusion criteria. One patient was excluded due to the presence of hemorrhage, and two patients were excluded due to insufficient EEG quality. The ages of 20 stroke patients (5.2 ± 5.5 years, 20% female) and 29 control patients (4.9 ± 5.8 years, 52% female) were analyzed.

[0118] X-ray stroke localization was anterior circulation (AC) stroke in 60% (n = 12) and posterior circulation (PC) stroke in 40% (n = 8). RIV was greater than 5% in 35% (n = 7) and greater than 10% in 60% (n = 12). Strokes were left in 45% (n = 9), right in 30% (n = 6), and bilateral in 25% (n = 5). The mean RIV was 8.1%, ranging from 0.01% to 22%. Imaging was obtained before EEG in 55% (n = 11) of patients.

[0119] The mean time to EEG placement was 17 ± 13 hours. 55% (n = 11) of patients had a seizure within 14 days of the stroke, and 35% (n = 7) had an EEG-captured seizure. EEGs were read as normal in 15% (n = 3) of patients, abnormal with epileptiform findings in 35% (n = 7), and abnormal with non-epileptiform findings in 50% (n = 10). Antiseizure medications (ASMs) were given in 55% (n = 11) of patients; 10% (n = 2) received one, 20% (n = 4) received two, and 25% (n = 5) received three or more ASMs. 45% (n = 9) of patients did not receive ASMs. Imaging was obtained before EEG in 55% (n = 11) of patients. Patient characteristics are summarized in Table 1. [Table 1-1] [Table 1-2]

[0120] COIN topographic visualization Figure 2 shows a comparison of representative neuroimaging data for each stroke patient using full and restricted circumferential montages, along with a topographic representation of the COIN. Patients with EEG-captured seizures and patients imaged prior to EEG are shown.

[0121] Summary COIN, BSI, and ADR test characteristics Median values ​​from the first 6 hours of EEG recordings from stroke and control patients were included for statistical analysis (Figures 3A-3D). The mean median COIN score was -34.7 for the primary cohort of all stroke patients and -9.5 for controls (p = 0.003), with an area under the receiver operating characteristic (ROC) of 0.88, sensitivity and specificity of 70% and 100%, and an optimal COIN cutoff value of -15. Figures 4A-4C show performance statistics for the primary cohort and subgroups consisting of patients with anterior circulation stroke, posterior circulation stroke, stroke with >5% RIV, and stroke with >10% RIV. Table 2 summarizes the results of using COIN in the full and circumferential montages, BSI, and ADR. Table 2 is presented in two parts: Table 2A and Table 2B. [Table 2] [Table 3]

[0122] As shown in the table above, strokes were detected with 70% sensitivity, 100% specificity, and 88% accuracy (see Table 2B: COIN complete montage: all strokes). Additionally, strokes with relative infarct volumes greater than 5% were detected with 92% sensitivity, 100% specificity, and 98% accuracy (Table 2B: COIN complete montage: relative infarct volume >5%). The threshold for C used to achieve these detections was -15 in both cases (see Table 2A: optimal cutoff column).

[0123] Time window analysis COIN data for stroke and controls were expressed as the time-independent proportion of studies spent below a COIN cutoff ranging from -35 to -5, which was used as input data for a random forest classification model (Figure 5A). Accuracy, AUROC, sensitivity, and specificity were measured using 200 random time windows ranging in length from 5 to 30 minutes. The results of the time window analysis are shown in Figures 5B-5E.

[0124] Consideration This study demonstrates that a continuous, quantitative measure of cerebral asymmetry derived from scalp EEG data has high sensitivity and specificity for detecting acute stroke with a relative infarct volume greater than 5% in hospitalized children. Validation of QEEG algorithms for stroke detection in children is limited, and this study demonstrates that the discriminatory performance of COIN is superior to ADR and comparable to BSI, with the additional advantage of providing interpretable visualization for localizing asymmetric regions within the brain. Furthermore, COIN's performance in detecting stroke using both full and limited (10-electrode) EEG montages is comparable to anterior circulation stroke, suggesting that this diagnostic technique may be deployed using portable EEG monitors that do not require the deployment of specialized technicians.

[0125] Although the topographic visualizer demonstrated agreement with imaging in localizing infarct size for most patients, in some patients there was a discrepancy between the size of the COIN signal on imaging and the infarct volume. For subjects with small infarcts and large COIN signals on imaging, imaging tended to be performed before EEG, suggesting that the infarct area may have expanded since imaging. Conversely, for patients with large infarcts but small COIN signals on imaging, imaging tended to be performed after EEG was removed. This may also be explained by the expansion of the infarct area that occurred from the first 6 hours of recording to the time of imaging. This result highlights the dynamic nature of COIN and warrants future evaluation.

[0126] Using the Youden J statistic, it was determined that a COIN cutoff value of -15 could be used to distinguish stroke from non-stroke. This finding was consistent across stroke volume and vascular territory subgroups, with COIN values ​​being more pronounced in larger stroke volumes. BSI demonstrated comparable performance in distinguishing COIN from control strokes but had higher variability in optimal cutoff values ​​across stroke volume and vascular territory subgroups. Furthermore, BSI does not provide localization or lateralization data and therefore cannot be used to generate topographical displays. 12、17 .

[0127] The quantitative stroke metric ADR relies on a loss of alpha power in conjunction with an increase in delta power, and the resulting decrease in ADR signifies ischemia of the underlying brain tissue. ADR has been found to perform poorly in children, likely due in part to differences in the EEG background frequency mixture in infants compared with older children and adults. The relative underrepresentation of activity above 8 Hz in younger children under 1 year of age precludes the use of ADR, which relies on the presence of electrical activity in the alpha (8-12 Hz) range. 18 ADR also showed a worsening ability to distinguish large strokes from control patients. This is likely explained by the loss of delta power that occurs after a complete loss of regional cortical perfusion, leading to an increase (or pseudonormalization) of ADR. ADR has been shown to be useful for the dynamic detection and trending of cerebral ischemia in subarachnoid hemorrhage. 8、16 , the results suggest that it may not be a useful screening tool for detecting large strokes.

[0128] The results of the time window analysis demonstrated improved accuracy and sensitivity of COIN as the recording period increased from 5 to 20 minutes, with performance remaining stable thereafter. This finding suggests that changes in COIN can be detected rapidly after the onset of stroke, allowing for confirmatory testing and therapeutic intervention. Although improvements in accuracy and sensitivity were observed with increasing time windows, a slight decrease in test specificity was observed over time. This reflects the random forest classification model, which is designed to optimize accuracy, hypothesized to result in a larger COIN cutoff with a shorter time window that minimizes the false positive rate. For strokes with RIV greater than 5%, the model was able to maintain a specificity of greater than 95% with 5-minute recordings, suggesting that very large COIN values ​​can be used to diagnose large strokes very quickly with a low false positive rate.

[0129] The accuracy of COIN using circumferential montage demonstrates its potential as a rapid triage tool using restricted EEG settings. The use of restricted circumferential EEG has been shown to reduce the time to diagnosis of subclinical seizures and has been associated with reduced costs and patient length of stay in acute and intensive care settings. 19、20 A similar approach can be used in the prehospital setting or after cardiac catheterization. 21 and mechanical circulatory support 6、22 This technology could potentially be developed for rapid triage and identification of stroke in specific clinical scenarios such as:

[0130] Point-of-care brain monitoring algorithms for stroke detection can streamline workflow for acute stroke management in hospitalized children, particularly those at high risk for stroke after cardiac procedures or on extracorporeal life support, who often have limited or unreliable neurological examinations due to sedation or muscle blockade. Importantly, the high sensitivity and specificity of COIN for large strokes also aids decisions regarding the risks and benefits of transporting critically ill patients to a CT scanner and exposing them to radiation, a common dilemma faced by intensivists and nursing teams managing patients at risk for stroke and other catastrophic neurological emergencies.

[0131] The rationale for EEG as a sensitive marker for cerebral ischemia has been clearly explained, 23 Despite extensive studies in adult populations, there has been limited research into the deployment of continuous EEG monitoring for stroke detection in children. 9 Although there are no prospective trials of EEG detection of stroke, a small pilot study in adult patients demonstrated the feasibility of EEG for detecting ischemic stroke in the emergency department. 24 EEG is considered an ideal platform for stroke detection with high diagnostic accuracy. 13 Although it is commonly suggested for patients with neurological impairments known to be at higher risk for ischemic exacerbation, 25 However, it is limited by the requirement for expert interpretation. The implementation of COIN in various clinical contexts may help overcome this feasibility barrier by making the detection of ischemia in EEG accessible to non-epileptic specialists.

[0132] This pilot study was possible in a pediatric population due to the high prevalence of seizures in pediatric stroke and the resulting frequent use of EEG. 14The prevalence of seizures and ASM in the stroke cohort may have influenced the background EEG, but because of the short duration of treatment, seizures likely contributed to a small portion of the background recording. The performance of the COIN algorithm for stroke detection was not significantly affected by the presence of seizures or the use of ASM. The SIPS EEG dataset was not designed for real-time capture of stroke, as EEG was placed after stroke onset in all cases. Furthermore, the timing of stroke depended on the time the patient was last seen; normal or positive diagnostic imaging was obtained, neither of which was an accurate reflection of stroke onset time.

[0133] Because the results did not take into account the patient's neurological examination, it was not possible to evaluate the utility of COIN compared with the clinical diagnosis of stroke or how COIN related to the severity of any abnormal neurological examination. The use of EEG to confirm the diagnosis of stroke in patients presenting with overt stroke syndromes is likely to be of limited value. Future studies of COIN should evaluate how it functions within specific clinical scenarios where it is expected to enhance clinical diagnosis, such as in sedated or neuromuscularly blocked patients in the perioperative window or in the intensive care unit.

[0134] References 12.Vanputten M.Extended BSI for continuous EEG monitoring in carotid endarterectomy.Clinical Neurophysiology.2006;117(12):2661-2666.doi:10.1016 / j.clinph.2006.08.007

[0135] 13.van Meenen LCC,van Stigt MN,Siegers A,et al.Detection of Large Vessel Occlusion Stroke in the Prehospital Setting:Electroencephalography as a Potential Triage Instrument.Stroke.2021;52(7).doi:10.1161 / STROKEAHA.120.033053

[0136] 14.Fox CK,Mackay MT,Dowling MM,et al.Prolonged or recurrent acute seizures after pediatric arterial ischemic stroke are associated with increasing epilepsy risk.Dev Med Child Neurol.2017;59(1):38-44.doi:10.1111 / dmcn.13198

[0137] 15.Oostenveld R,Fries P,Maris E,Schoffelen JM.FieldTrip:Open Source Software for Advanced Analysis of MEG,EEG,and Invasive Electrophysiological Data.Computational Intelligence and Neuroscience.2011;2011:1-9.doi:10.1155 / 2011 / 156869

[0138] 16.Claassen J,Hirsch LJ,Kreiter KT,et al.Quantitative continuous EEG for detecting delayed cerebral ischemia in patients with poor-grade subarachnoid hemorrhage.Clinical Neurophysiology.2004;115(12):2699-2710.doi:10.1016 / j.clinph.2004.06.017

[0139] 17.Wilkinson CM,Burrell JI,Kuziek JWP,Thirunavukkarasu S,Buck BH,Mathewson KE.Predicting stroke severity with a 3-min recording from the Muse portable EEG system for rapid diagnosis of stroke.Sci Rep.2020;10(1):18465.doi:10.1038 / s41598-020-75379-w

[0140] 18.A Kaminska,M Eisermann,P Plouin.Child EEG(and maturation).Handbook of Clinical Neurology.2019;160:125-142.doi:10.1016 / B978-0-444-64032-1.00008-4

[0141] 19.Ney JP,Gururangan K,Parvizi J.Modeling the economic value of Ceribell Rapid Response EEG in the inpatient hospital setting.Journal of Medical Economics.2021;24(1):318-327.doi:10.1080 / 13696998.2021.1887877

[0142] 20.LaMonte MP.Ceribell EEG shortens seizure diagnosis and workforce time and is useful for COVID isolation.Epilepsia Open.2021;6(2):331-338.doi:10.1002 / epi4.12474

[0143] 21.Harrar DB,Salussolia CL,Vittner P,et al.Stroke After Cardiac Catheterization in Children.Pediatric Neurology.2019;100:42-48.doi:10.1016 / j.pediatrneurol.2019.07.005

[0144] 22.Di Gennaro JL,Chan T,Farris RWD,Weiss NS,McMullan DM.Increased Stroke Risk in Children and Young Adults on Extracorporeal Life Support with Carotid Cannulation.ASAIO Journal.2019;65(7):718-724.doi:10.1097 / MAT.0000000000000912

[0145] 23.Jordan KG.Emergency EEG and Continuous EEG Monitoring in Acute Ischemic Stroke.Journal of Clinical Neurophysiology.2004;21(5):12.

[0146] 24. Shreve L, Kaur A, Vo C, et al. Electroencephalography Measures are Useful for Identifying Large Acute Ischemic Stroke in the Emergency Department. Journal of Stroke and Cerebrovascular Diseases.2019;28(8):2280-2286.doi:10.1016 / j.jstrokecerebrovasdis.2019.05.019

[0147] Example 2: Differentiating between large and small strokes in adults Method overview This retrospective cohort evaluates the performance of COIN in adult patients with ischemic stroke from a single university-affiliated hospital using a convenience dataset containing 8 hours of EEG data per subject. Subjects are classified as having large or small strokes based on a threshold volume of 100 mL. COIN is calculated in separate 4-second epochs by cross-referencing the power ratio in each channel across the entire region and to the contralateral homolog. COIN data are used to visualize stroke regions, and random forest classification with a 10-fold cross-validation task is used to obtain test performance metrics. To assess the EEG length required to optimize performance, the analysis is repeated using restricted samples pooled from random time windows ranging from 5 to 30 minutes.

[0148] Summary of results Thirty-five patients with a mean age of 67 (SD 17) years and a median NIHSS score of 14 (IQR 8-19) were analyzed. Ten patients had large strokes and 25 had small strokes. The mean (SE) median COIN scores were -53 (13) in the large group and -16 (1.2) in the small group (p = 0.0001). Average test performance using 8 hours of recording per patient was 86% accuracy, 90% sensitivity, and 84% specificity.

[0149] Summary of conclusions COIN can distinguish large (core volume >100 mL) from small ischemic strokes with good accuracy and high specificity. Prospective implementation and evaluation are needed to determine the utility of COIN as an aid in stroke diagnosis in patients with limited testing.

[0150] Study population This was a retrospective secondary analysis of EEG and clinical data from patients over 18 years of age who underwent at least 8 hours of continuous EEG with AIS at a single center from 2012 to 2019. Patients were excluded for lack of neuroimaging and EEG recordings with excessive line interference, muscle artifact, burst suppression, or severe voltage suppression. Neuroimaging was available as previously stored representative images for each patient. Clinical data included patient age and sex, stroke laterality and location, NIHSS score, TOAST score, and stroke volume calculated by ABC / 2.

[0151] EEG processing A bipolar montage was applied, and the study was divided into 4-second epochs. Artifact detection and rejection were performed using the Fieldtrip algorithm to detect clipping artifacts (amplitude threshold 0.05 μV) and threshold artifacts (>300 μV) with z-score rejection for muscle artifacts (50-99 Hz, z-threshold = 4) and movement artifacts (1-5 Hz, z-threshold = 4). Each epoch containing artifacts longer than 2.5 seconds was rejected, and epochs containing artifacts shorter than 2.5 seconds were partially rejected to allow analysis of the remaining usable data from the epoch. The 15 input EEG data were passed through a fast Fourier transform to generate power spectral data for each 4-second epoch for each channel in the bipolar montage.

[0152] Calculation of the Correlation of Injury to the Nervous System (COIN) Index The raw EEG data is passed through a fast Fourier transform to generate a power spectrum matrix A, which is input into equations (i)-(iv) below.

number

[0153] The mean values ​​within each channel are mapped to a topographic visualizer, and all channels with negative values ​​are summed to generate a summary COIN value, C, for each 4-second epoch. Values ​​for C are smoothed over time using a 5-minute moving average. In this study, we selected the 8-18 Hz frequency range to primarily reflect the attenuation effect of stroke in the alpha power band (8-13 Hz) in adults.

number

[0154] Data processing and statistical analysis Raw EEG preprocessing and artifact rejection were performed on MATLAB (MathWorks, Natick, MA) using Fieldtrip. Fifteen 8-hour EEG recordings were available for analysis from each patient. Median COIN values ​​across recordings were used as individual data points for two-sample t-tests and logistic regression.

[0155] result Patient characteristics and neuroimaging findings. Of 71 patients, 35 met the inclusion and exclusion criteria of mean (SD) age 67 (17) and NIHSS score 13.7 (6.6). Infarct volumes were <100 mL in 25 patients (69 ± 18 years, 48% female, NIHSS score 12.9 ± 6.8, mean stroke volume 44 ± 24) and >100 mL in 10 patients (62 ± 15 years, 50% female, NIHSS score 16.1 ± 5.5, mean stroke volume 233 ± 108). In patients with infarct volumes less than and greater than 100 mL, stroke laterality was left in 16 (64%) and 5 (50%) cases, right in 4 (16%) and 4 (40%) cases, bilateral in 2 (8%) and 1 (10%) cases, and unspecified in 3 (12%) and 0 (0%) cases, respectively. Vascular territory was anterior in 19 (76%) and 10 (100%) cases, posterior in 5 (20%) and 0 (0%) cases, and unspecified in 1 (4%) and 0 (0%) cases. Etiology was aortic disease in 8 (32%) and 6 (60%) cases, cardioembolism in 13 (52%) and 4 (40%) cases, small vessel in 0 (0%) and 0 (0%) cases, and other or undetermined in 4 (16%) and 0 (0%) cases.

[0156] COIN Analysis and Visualization In patients with infarct volumes less than 100 mL and greater than 100 mL, the mean (SD) median COIN values ​​were -15.7 (1.2) and -53.1 (13), respectively, and were significantly different (p = 0.0001). The laterality of the COIN signal on visualization was consistent with neuroimaging in 11 (44%) and 10 (100%) of cases, discordant in 2 (8%) and 0 (0%) of cases, and uncertain in 12 (48%) and 0 (0%) of cases. A representative comparison of neuroimaging and COIN visualization is shown in Figure 10. Logistic regression using the median COIN values ​​yielded a receiver operating characteristic area under 0.88, with a confidence interval of 0.59–0.98. A COIN cutoff value of -20 yielded a maximum Youden J statistic of 0.74, with a corresponding sensitivity of 90%, specificity of 84%, and accuracy of 86%. A specificity of 100% was observed at a COIN cutoff of -28, with a corresponding sensitivity of 60%. The median and interquartile range of COIN values ​​for stroke volume are shown in Figure 6A, and the results of the logistic regression are shown in Figures 6B and 6C.

[0157] Consideration This study demonstrates that the quantitative EEG metric COIN provides an intuitive and accurate means of distinguishing large strokes (core infarct volume >100 mL) from small strokes (core infarct volume <100 mL) in adults. We found that a cutoff value of -20 could be used to distinguish large from small strokes, and that COIN values ​​below -28 provided high specificity for the presence of large versus small volume strokes.

[0158] For all but one patient with a stroke less than 100 mL and a median COIN less than -20, the COIN visualizer matched the reported stroke area, which may be explained by a mismatch between the infarct core volume and downstream penumbral tissue, suggesting that the patient was at risk for or had already experienced an enlarged stroke area after imaging.

[0159] This finding demonstrates that abnormal COIN values ​​are most consistently found in anterior circulation strokes. A possible explanation is that surface EEG recordings capture cortical activity primarily from brain tissue supplied by the anterior circulation, which also supplies a higher proportion of the overall brain tissue. Indeed, all strokes were in the anterior circulation in the group with infarct volumes greater than 100 mL. Of the 35 patients analyzed, only five had strokes in the posterior circulation, and among these, the lowest COIN value observed was −16. The issue highlighted here is further exemplified by the finding that NIHSS scores were similar between the two groups. Small (<100 mL) infarct volumes in highly functional regions, such as the brainstem or internal capsule, may indicate severe neurological dysfunction, reflected by high NIHSS scores.

[0160] conclusion COIN can distinguish large (core volume >100 mL) from small ischemic strokes with good accuracy and high specificity. Prospective implementation and evaluation are needed to determine the utility of COIN as an aid in stroke diagnosis in patients with limited testing.

[0161] Example 3: Real-time recognition of stroke A comatose patient on life support with a high risk of bleeding was monitored for possible seizures as an explanation for the altered mental status. A brain scan using magnetic resonance brain imaging before initiating EEG showed no evidence of brain abnormalities. The patient required mechanical ventilation and sedation.

[0162] EEG data was recorded and processed to obtain topographical brain maps, as shown in Figures 9A-C. The color of the brain maps corresponds to the "COIN value," also described as the m(j) value.

[0163] Figure 9A shows the results of monitoring while the patient was in a baseline state. The horizontal axis had a time scale from 0 to 4 seconds. The vertical axis shows the 16 different channels monitored. An epileptologist reviewing the EEG data itself did not notice any focal EEG abnormalities. The brain map showed mostly white areas with m(j) values ​​close to zero, thus suggesting a brain state without focal abnormalities.

[0164] However, the brain map in Figure 9B shows a dark blue color in the right posterior quadrant, indicating focal attenuation. In particular, the right brain, designated by the letter "R," is actually shown on the left side of the figure. Therefore, analysis of the EEG data suggested a possible stroke in the right posterior quadrant. The pattern in Figure 9B persisted for 8 hours. The epileptologist, for example, did not notice any focal EEG abnormalities during the time period shown in Figure 9B because the raw EEG data in Figure 9B resembled those in Figure 9A.

[0165] Approximately 8 hours later, there was a sudden change in the EEG data, as shown in Figure 9C. The brain map showed a dark blue area in the right posterior quadrant of the brain, and a dark red area located in the anterior left hemisphere. This analysis indicated a very high probability of stroke. After the sudden change from the pattern in Figure 9B to that in Figure 9C, the epileptologist noticed focal EEG abnormalities. Specifically, the epileptologist noticed that the EEG recording was very flat in Figure 9C, but had large amplitudes in both Figures 9A and 9B.

[0166] Computed tomography determined that the patient had experienced a large intracranial hemorrhage in the right posterior quadrant of the brain.

[0167] Thus, interpretation of EEG data by calculating COIN values ​​was able to detect abnormalities and a high risk of stroke approximately 8 hours before epileptologists detected any abnormalities in the EEG data themselves.

[0168] Although the foregoing invention has been described in some detail by way of illustration and example for clarity of understanding, it will be readily apparent to those skilled in the art in light of the teachings of this invention that certain changes and modifications can be made to the invention without departing from the spirit or scope of the appended claims.

[0169] Accordingly, the foregoing merely illustrates the principles of the present invention. Those skilled in the art will recognize that various configurations, not explicitly described or shown herein, can be devised which embody the principles of the present invention and are within its spirit and scope. Furthermore, all examples and conditional language recited herein are intended primarily to aid the reader in understanding the principles of the present invention and concepts contributed by the inventors to further the art, and should not be construed as limitations on such specifically recited examples and conditions. Furthermore, all statements herein reciting principles, aspects, and embodiments of the present invention, as well as specific examples thereof, are intended to encompass both structural and functional equivalents thereof. Additionally, such equivalents are intended to include both currently known equivalents and equivalents developed in the future, i.e., any elements developed that perform the same function, regardless of structure. Furthermore, nothing disclosed herein is intended to be offered to the public, regardless of whether such disclosure is expressly recited in the claims.

[0170] Accordingly, the scope of the present invention is not intended to be limited to the exemplary embodiments shown and described herein. Rather, the scope and spirit of the present invention is embodied by the appended claims. In the claims, 35 U.S.C. 112(f) is expressly defined to apply for a limitation in a claim only if the precise phrase "means for" or the precise phrase "step for" appears at the beginning of such limitation in the claim; if such precise phrase is not used in the limitation in the claim, 35 U.S.C. 112(f) does not apply.

Claims

1. 1. A method of assessing whether a subject has had a stroke and treating said subject accordingly, said method comprising: (a) recording electroencephalography (EEG) data from electrodes placed at scalp locations on the subject's head; (b) determining relative electrical activity from the EEG data, wherein said determining comprises: determining the relative electrical activity at each scalp location compared to the electrical activity at all scalp locations; determining the relative electrical activity at each scalp location compared to the electrical activity at its corresponding contralateral scalp location; (c) determining a high probability of stroke based on the determined relative electrical activity; and (d) treating said subject for a high probability of stroke.

2. The method of claim 1 , wherein the electrodes are placed at four or more scalp locations.

3. 3. The method of claim 1 or 2, wherein treating the subject for a high probability of stroke comprises performing an additional stroke detection measurement.

4. 4. The method of claim 3, wherein the additional stroke detection measurement is a brain magnetic resonance imaging (MRI) or a head computed tomography (CT) scan.

5. 5. The method of any one of claims 1 to 4, wherein treating the subject for a high probability of stroke comprises medical or surgical intervention to reverse or minimize the effects of stroke.

6. 6. The method of any one of claims 1-5, wherein treating the subject for an elevated probability of stroke comprises transporting the subject to a medical facility, notifying the medical facility of an elevated probability of stroke, or a combination thereof.

7. The method of any one of claims 1 to 6, wherein the subject had an elevated risk of stroke prior to said recording.

8. 8. The method of claim 7, wherein the elevated risk of stroke comprises a risk selected from the group consisting of altered mental status, loss of motor function, loss of touch in a body part, dizziness, headache, and difficulty speaking.

9. 9. The method of claim 7 or 8, wherein the elevated risk of stroke comprises a risk selected from the group consisting of head trauma, hematoma, and subarachnoid hemorrhage.

10. 10. The method of any one of claims 7 to 9, wherein the elevated risk of stroke comprises a risk selected from the group consisting of currently undergoing surgery and having undergone surgery within the past 30 days.

11. The method of any one of claims 1 to 10, wherein the subject is unconscious.

12. determining the relative electrical activity from the EEG data; generating a matrix A from the EEG data, the matrix A comprising elements a(t,j), where a(t,j) refers to the amplitude of electrical activity at time t and scalp location j; generating a matrix B from matrix A, wherein matrix B includes elements a(f,j), where a(f,j) indicates the power of electrical activity at frequency f and scalp location j; generating a lookup matrix R containing elements r(f,j), where r(f,j) indicates the relative power of electrical activity at frequency f and scalp location j compared to the power of electrical activity at frequency f at all scalp locations; 12. The method of claim 1, comprising generating a symmetric matrix S comprising elements s(f,j), where s(f,j) indicates the relative power of electrical activity at frequency f and scalp location j compared to the power of electrical activity at scalp location j* that is contralateral to frequency f and scalp location j.

13. 13. The method of claim 12, wherein each r(f,j) is generated by comparing a(f,j) to the average of the a(f) values ​​from all scalp locations j.

14. Each r(f,j) is described by the following equation: [Equation 1] 14. The method of claim 13, wherein M is the total number of scalp locations j.

15. 15. The method of claim 12, wherein generating each s(f,j) comprises dividing a(f,j) by a(f,j*), where j* is a scalp location that is contralateral to scalp location j.

16. 16. The method of claim 15, wherein each s(f,j) is described by the following equation: [Equation 2]

17. Determining a high probability of stroke generating a gating matrix G containing elements g(f,j) by combining matrix R with matrix S; generating a map matrix M containing elements m(j), where m(j) is generated by averaging g values ​​for the same scalp location j across different frequencies f; The method of any one of claims 12 to 16, comprising determining a high probability of stroke based on the map matrix M.

18. 20. The method of claim 17, wherein generating the gating matrix G comprises calculating each g(f,j) according to the following equation: g(f,j)=r(f,j) ２ ・s(f,j)

19. 19. The method of claim 17 or 18, wherein generating the gating matrix G further comprises setting each g(f,j) to zero if its corresponding r(f,j) has the opposite sign as its corresponding s(f,j).

20. The method of any one of claims 17 to 19, wherein a high probability of stroke is determined if one or more m(j) values ​​are within a predetermined threshold range.

21. 21. The method of claim 20, wherein the predetermined threshold range of one or more m(j) values ​​is less than −15.

22. 20. The method of any one of claims 17 to 19, wherein a high probability of stroke is determined if the sum of all negative m(j) values ​​is within the predetermined threshold range.

23. 21. The method of claim 20, wherein the predetermined threshold range of the sum of all negative m(j) values ​​is less than -15.

24. 24. The method of any one of claims 17 to 23, further comprising estimating the location of the stroke by identifying the scalp location with the most negative m(j) value.

25. 25. The method of any one of claims 1 to 24, wherein the probability of stroke is the probability of ischemic stroke.

26. 25. The method of any one of claims 1 to 24, wherein the probability of stroke is the probability of hemorrhagic stroke.

27. 27. The method of any one of claims 1 to 26, further comprising determining a size of the stroke based on the relative electrical activity.

28. 28. The method of claim 27, wherein the size of the stroke is determined to be 100 ml or greater based on an m(j) value of −20 or less, and the size of the stroke is determined to be less than 100 ml based on an m(j) value of greater than −20.

29. repeating the recording of the EEG data during a second time period; determining the relative electrical activity from the EEG data from one or more additional time periods; generating a combined relative electrical activity by combining the relative electrical activity from the EEG data from the first time period and the relative electrical activity from the EEG data from the second time period; 26. The method of any one of claims 1 to 25, further comprising determining a high probability of stroke based on the combined relative electrical activity.

30. 1. A controller for assessing whether a subject has had a stroke and communicating a result of the assessment, said controller comprising: a) acquiring electroencephalography (EEG) data from electrodes placed at scalp locations on the subject's head; b) determining and calculating relative electrical activity from said EEG data; determining the relative electrical activity at each scalp location compared to the electrical activity at all scalp locations; determining the relative electrical activity at each scalp location compared to the electrical activity at its corresponding contralateral scalp location; c) determining a probability that the subject has experienced a stroke near one or more scalp locations based on the relative electrical activity; d) a controller configured to electronically instruct the communication device to communicate the determined probability.

31. The controller of claim 30 , wherein communicating the determined probabilities includes providing a visual image indicative of one or more determined probabilities.

32. 32. The controller of claim 31, wherein the image includes a symbol representing a top or bottom view of the subject's head.

33. 33. The controller of claim 32, wherein the image uses different colors to indicate different relative probabilities that a stroke occurred near different scalp locations.

34. 34. The controller of claim 33, wherein the different colors include a gradient between three or more colors.

35. 35. The controller of claim 34, wherein the image uses a blue-white-red color gradient to indicate a gradient of high-medium-low stroke probability.

36. 36. The controller of any one of claims 30 to 35, wherein the controller is further configured to electronically instruct an alert device to provide a visual alert, an audible alert, or a combination thereof if the determined probability near one or more scalp locations is within a predetermined threshold range.

37. determining the relative electrical activity from the EEG data; generating a matrix A from the EEG data, the matrix A comprising elements a(t,j), where a(t,j) refers to the amplitude of electrical activity at time t and scalp location j; generating a matrix B from matrix A, where matrix B includes elements a(f,j), where a(f,j) indicates the power of electrical activity at frequency f and scalp location j; generating a lookup matrix R containing elements r(f,j), where r(f,j) indicates the relative power of electrical activity at frequency f and scalp location j compared to the power of electrical activity at frequency f at all scalp locations; 37. The controller of claim 30, comprising generating a symmetric matrix S comprising elements s(f,j), where s(f,j) indicates the relative power of electrical activity at frequency f and scalp location j compared to the power of electrical activity at scalp location j* that is contralateral to frequency f and scalp location j.

38. 38. The controller of claim 37, wherein each r(f,j) is generated by comparing a(f,j) to the average of the a(f) values ​​from all scalp locations j.

39. Each r(f,j) is described by the following equation: [Equation 3] 39. The controller of claim 38, wherein M is the total number of scalp locations j.

40. 40. The controller of any one of claims 37 to 39, wherein generating each s(f,j) comprises dividing a(f,j) by a(f,j*), where j* is a scalp position that is contralateral to scalp position j.

41. 41. The controller of claim 40, wherein each s(f,j) is described by the following equation: [Equation 4]

42. Determining a high probability of stroke generating a gating matrix G containing elements g(f,j) by combining matrix R with matrix S; generating a map matrix M containing elements m(j), where m(j) is generated by averaging g values ​​for the same scalp location j across different frequencies f; 42. The controller of any one of claims 37 to 41, comprising determining the probability of stroke based on a map matrix M.

43. 37. The controller of claim 36, wherein generating the gating matrix G comprises calculating each g(f,j) according to the following equation: g(f,j)=r(f,j) ２ ・s(f,j)

44. 44. The controller of claim 42 or 43, wherein generating the gating matrix G further comprises setting each g(f,j) to zero if its corresponding r(f,j) has the opposite sign as its corresponding s(f,j).

45. A controller as claimed in any one of claims 42 to 44, wherein a high probability of stroke is determined if one or more m(j) values ​​are within a predetermined threshold range.

46. 46. ​​The controller of claim 45, wherein the predetermined threshold range of one or more m(j) values ​​is less than -15.

47. 46. ​​A controller as claimed in any one of claims 42 to 45, wherein a high probability of stroke is determined if C is within a predetermined range, C being the sum of all negative m(j) values.

48. 48. The controller of claim 47, wherein the predetermined threshold range for C is less than −15.

49. A controller according to any one of claims 37 to 48, wherein steps a), b), c), and d) are repeated within two minutes or less.

50. 50. The controller of any one of claims 30 to 49, wherein the probability of stroke is the probability of ischemic stroke.

51. 50. The controller of any one of claims 30 to 49, wherein the probability of stroke is the probability of hemorrhagic stroke.

52. 52. The controller of any one of claims 30 to 51, wherein the controller is configured to detect stroke with a sensitivity of 70% or greater and a specificity of 90% or greater.

53. 52. The controller of any one of claims 30 to 51, wherein the controller is configured to detect strokes with an infarct volume of 5% or greater with a sensitivity of 90% or greater and a specificity of 90% or greater.

54. The controller of any one of claims 30 to 53, wherein the controller is further configured to determine a size of the stroke based on the relative electrical activity.

55. 55. The controller of claim 54, wherein the size of the stroke is determined to be 100 ml if C is less than or equal to −20, and the size of the stroke is determined to be less than 100 ml if C is greater than −20, where C is the sum of all negative m(j) values.

56. 56. The controller of claim 54 or 55, wherein the controller is configured to detect whether the size of the stroke is greater than or less than 100 ml with a sensitivity of 80% or greater and a specificity of 85% or greater.

57. The controller: repeating said recording of said EEG data during a second time period; determining the relative electrical activity from the EEG data from one or more additional time periods; generating a combined relative electrical activity by combining the relative electrical activity from the EEG data from the first time period and the relative electrical activity from the EEG data from the second time period; 57. The controller of any one of claims 30 to 56, further configured to determine the probability of stroke based on the combined relative electrical activity.

58. 1. A system for assessing whether a subject has had a stroke and communicating the results of said assessment, said system comprising: A controller according to any one of claims 30 to 57; one or more of the communication device, the alert device, and the electrodes.

59. 1. A kit for assessing whether a subject has had a stroke and communicating the results of said assessment, the system comprising: A controller according to any one of claims 30 to 57. A kit comprising a package containing the controller.

60. 60. The kit of claim 59, further comprising one or more of the communication device, alert device, and electrodes contained in the package.

Images