Systems and methods for detecting stroke

Abstract

A system for detecting a stroke includes a sensor device configured to obtain physiological data, such as brain activity data, from a patient. The sensor device may include electrodes configured to be placed at the back of the patient's neck or skull base. The electrodes may detect electrical signals corresponding to brain activity in the P3, Pz, and / or P4 brain regions or other brain regions. A computing device communicatively coupled to the sensor device is configured to receive the physiological data and analyze it to indicate whether the patient has suffered a stroke.

Description

[0001] This application claims priority to U.S. Patent Application No. 17 / 006,444, filed on August 28, 2020, and U.S. Patent Application No. 17 / 176,504, filed on February 16, 2021, both of which claim the benefit of U.S. Provisional Application No. 62 / 977,503, filed on February 17, 2020. The entire contents of each of these applications are incorporated herein by reference. Technical Field

[0002] The present technology relates to medical devices, and more particularly to systems and methods for detecting stroke. Background Art

[0003] Stroke is a serious medical condition that may lead to permanent neurological damage, complications and death. A stroke can be characterized as a rapidly developing loss of brain function due to a disturbance in the blood vessels that supply blood to the brain. The loss of brain function may be the result of ischemia (lack of blood supply) caused by thrombosis, embolism or hemorrhage. A reduction in blood supply may lead to dysfunction of brain tissue in the area.

[0004] Stroke is the second leading cause of death and the first leading cause of disability worldwide. Speed ​​of treatment is a critical factor in stroke care as an average of 1.9 million neurons are lost every minute during a stroke. Stroke diagnosis and the time between the event and therapy delivery are major barriers to improving treatment effectiveness. There are 3 main causes of stroke; i) ischemic stroke (approximately 65% ​​of all strokes), ii) hemorrhagic stroke (approximately 10% of all strokes), and iii) cryptogenic stroke (including TIA, which accounts for approximately 25% of all strokes). Stroke can be considered to have a neurogenic and / or cardiogenic origin.

[0005] There are multiple methods for treating patients who have suffered apoplexy. For example, clinicians can use anticoagulants such as warfarin, or intravascular interventions such as thrombectomy can be performed to treat ischemic stroke. For example, clinicians can use antihypertensive drugs, such as beta blockers (for example, labetalol) and ACE inhibitors (for example, enalapril), or intravascular interventions such as coil embolization can be performed to treat hemorrhagic stroke. Finally, if the symptoms of stroke subside on their own, and the neurological examination is negative, clinicians can use long-term cardiac monitoring (external or implantable) to determine the potential cardiac origin of cryptogenic stroke. However, due to failure to identify in time whether the patient is suffering from or has recently suffered from apoplexy, this treatment may not be fully utilized and / or relatively ineffective. This has a specific risk for making the patient relatively active after a rough assessment of the milder stroke. Summary of the invention

[0006] For example, the present invention is exemplified according to the various aspects described below. For convenience, various embodiments of various aspects of the present invention are described as numbered clauses (1, 2, 3, etc.). These clauses are provided as examples but not limiting the present invention. It should be noted that any of the dependent clauses can be combined in any combination and placed in the corresponding independent clauses. Other clauses can be presented in a similar manner.

[0007] 1. A stroke detection system, comprising:

[0008] a sensor device configured to obtain physiological data from a patient; and

[0009] a computing device, the computing device being communicatively coupled to the sensor device, the computing device being configured to:

[0010] receiving the physiological data from the sensor device;

[0011] analyzing the physiological data; and

[0012] Based on the analysis, a patient stroke indicator is provided.

[0013] 2. A system according to clause 1, wherein the physiological data includes brain activity data.

[0014] 3. A system according to any of the clauses herein, wherein the sensor device comprises a plurality of electrodes configured to detect brain activity data corresponding to activity in at least one of the P3, Pz and P4 brain regions.

[0015] 4. A system according to any of the clauses herein, wherein the sensor device comprises a plurality of electrodes configured to detect brain activity data corresponding to activity in each of the P3, Pz and P4 brain regions.

[0016] 5. A system according to any of the clauses herein, wherein the sensor device is configured to be placed on or adjacent to the posterior portion of the patient's neck or base of the head or skull.

[0017] 6. A system according to any of the clauses herein, wherein the sensor device is configured to be placed above the shoulder of the patient.

[0018] 7. A system according to any of the clauses herein, wherein the sensor device is configured to be positioned at or below the occipital bone of the patient.

[0019] 8. A system according to any of the clauses herein, wherein the sensor device comprises a housing configured to be implanted in the patient.

[0020] 9. A system according to claim 9, wherein the shell is configured to be implanted subcutaneously.

[0021] 10. A system according to any of the clauses herein, wherein the sensor device comprises a housing configured to be placed on the skin of the patient.

[0022] 11. The system of clause 10, wherein the sensor device comprises electrodes configured to contact the patient's skin.

[0023] 12. The system of clause 11, wherein the electrode comprises a protrusion configured to at least partially penetrate the patient's skin.

[0024] 13. The system of clause 12, wherein the protrusions comprise microneedles.

[0025] 14. A system according to any of the clauses herein, wherein the sensor device comprises an EEG array.

[0026] 15. The method of clause 14, wherein the EEG array comprises at least 2, at least 3, at least 4 or at least 5 electrodes.

[0027] 16. The method of clause 14, wherein the EEG array comprises fewer than 6, fewer than 5, fewer than 4, or fewer than 3 electrodes.

[0028] 17. A system according to any of the clauses herein, wherein the sensor device includes a housing having a volume less than about 1.5cc, about 1.4cc, about 1.3cc, about 1.2cc, about 1.1cc, about 1.0cc, about 0.9cc, about 0.8cc, about 0.7cc, about 0.6cc, about 0.5cc or about 0.4cc.

[0029] 18. A system according to any of the clauses herein, wherein the sensor device includes a shell having a lower surface configured to face the patient's tissue, an upper surface opposite the lower surface, and a thickness extending between the lower surface and the upper surface.

[0030] 19. The system of clause 18, wherein the housing comprises a plurality of sub-housings coupled together by flexible members or conductors.

[0031] 20. The system of clause 19, wherein electrodes of the sensor device are distributed between the sub-housings.

[0032] 21. The system of clause 18, wherein the housing is flexible such that the lateral ends can move forward relative to a central portion of the housing.

[0033] 22. The system of clause 21, wherein one electrode is disposed at each of the lateral ends.

[0034] 23. A system according to clause 21 or 22, wherein at least one electrode is disposed in a central portion of the shell.

[0035] 24. A system according to any one of clauses 18 to 23, wherein the thickness is less than about 10 mm, about 9 mm, about 8 mm, about 7 mm, about 6 mm, about 5 mm, about 4 mm or about 3 mm.

[0036] 25. The system of any of clauses 18 to 24, wherein the electrodes of the sensor device are exposed along the lower surface.

[0037] 26. The system of clause 25, wherein the sensor device comprises at least three electrodes exposed along the lower surface, and wherein the three electrodes are not aligned along a common axis.

[0038] 27. A system according to claim 26, wherein the shell is elongated along a horizontal axis, and wherein the first electrode is disposed substantially centrally along the horizontal axis, the second electrode is spaced apart from the first electrode along the horizontal axis in a first direction, and wherein the third electrode is spaced apart from the first electrode along the horizontal axis in a second direction opposite to the first.

[0039] 28. The system of clause 26, wherein the shell is elongated along a first horizontal axis, and wherein the first electrode is disposed substantially centrally along the horizontal axis and the second electrode is disposed along the horizontal axis toward a first end of the shell.

[0040] 29. The system of any of clauses 17 to 28, wherein the upper and lower surfaces form a substantially boomerang or chevron shape.

[0041] 30. A system according to any of the clauses herein, wherein one or more electrodes are disposed on a first major surface of the device and one or more electrodes are disposed on an opposing second major surface of the device.

[0042] 31. A system according to any of the clauses herein, wherein the sensor device comprises a housing configured to be delivered through a trocar introducer.

[0043] 32. A system according to any of the clauses herein, wherein the sensor device and the computing device are enclosed in a common housing.

[0044] 33. A system according to any of the clauses herein, wherein the physiological data includes at least three EEG signal channels.

[0045] 34. A system according to any of the clauses herein, wherein the physiological data includes brain electrical activity data and heart electrical activity data, and wherein analyzing the physiological data includes filtering the physiological data to separate the brain electrical activity data from the heart electrical activity data.

[0046] 35. A system according to any of the clauses herein, wherein the physiological data comprises electrical signals detected by electrodes of the sensor device, and wherein analyzing the physiological data comprises analyzing the electrical signals to detect brain activity.

[0047] 36. A system according to clause 35, wherein analyzing the electrical signals to detect brain activity data includes filtering the electrical signals to reduce the contribution of electrical signals generated by cardiac activity.

[0048] 37. A system according to clause 35, wherein analyzing the electrical signals to detect brain activity data includes filtering the electrical signals to reduce the contribution of electrical signals generated by muscle activity.

[0049] 38. A system according to any of the clauses herein, wherein the physiological data includes motion data, and wherein the computing device is further configured to analyze the motion data to make a fall determination.

[0050] 39. A system according to any of the clauses herein, wherein providing the patient stroke indicator comprises classifying an identified stroke as ischemic or hemorrhagic.

[0051] 40. The system of any of the clauses herein, wherein providing the patient stroke indicator comprises determining whether the patient is suffering from a stroke.

[0052] 41. The system of any of the preceding clauses, wherein providing the patient stroke indicator comprises determining a risk that the patient will suffer a stroke.

[0053] 42. A system according to any of the clauses herein, wherein providing the patient stroke indicator comprises determining the location of the stroke.

[0054] 43. The system of any of the clauses herein, wherein providing the patient stroke indicator comprises providing a confidence score associated with determining that the patient has a stroke.

[0055] 44. The system of any of the clauses herein, wherein providing the patient with a stroke indicator comprises providing a recommended therapeutic action accompanying a stroke determination.

[0056] 45. An apparatus comprising:

[0057] at least one housing configured to rest on a posterior portion of a patient's neck or base of a skull; and

[0058] A plurality of electrodes carried by the housing are configured to detect electrical signals corresponding to brain activity in at least P3, Pz, and P4 brain regions of the patient.

[0059] 46. ​​The device of clause 45, wherein the device further comprises processing circuitry configured to analyze the detected electrical signal to provide a patient stroke indicator.

[0060] 47. The apparatus of clause 46, wherein the processing circuitry comprises a tensor processing unit.

[0061] 48. A device according to any of the clauses herein, wherein the device is configured to be implantable.

[0062] 49. A device according to clause 48, wherein the device is configured to be implanted subcutaneously.

[0063] 50. A device according to any of the clauses herein, wherein the device is configured to be placed on the skin of the patient.

[0064] 51. A device according to any of the clauses herein, wherein the device is configured to be placed above the shoulder of the patient.

[0065] 52. A device according to any of the clauses herein, wherein the device is configured to be placed at or below the occipital bone of the patient.

[0066] 53. A device according to any of the clauses herein, wherein the electrode is configured to contact the patient's skin.

[0067] 54. A device according to clause 53, wherein the electrode comprises a protrusion configured to at least partially penetrate the patient's skin.

[0068] 55. A device according to clause 54, wherein the protrusions comprise microneedles.

[0069] 56. A device according to any of the clauses herein, wherein the shell has a volume less than about 1.5cc, about 1.4cc, about 1.3cc, about 1.2cc, about 1.1cc, about 1.0cc, about 0.9cc, about 0.8cc, about 0.7cc, about 0.6cc, about 0.5cc or about 0.4cc.

[0070] 57. A device according to any of the clauses herein, wherein the shell has a lower surface configured to face the patient's tissue for sensing, an upper surface opposite the lower surface, and a thickness extending between the lower surface and the upper surface.

[0071] 58. A device according to clause 57, wherein the thickness is less than about 10 mm, about 9 mm, about 8 mm, about 7 mm, about 6 mm, about 5 mm, about 4 mm or about 3 mm.

[0072] 59. A device according to clause 57 or 58, wherein the electrode is exposed along the lower surface.

[0073] 60. A device according to any of the clauses herein, wherein at least three electrodes are exposed along the lower surface, and wherein the three electrodes are not aligned along a common axis.

[0074] 61. An apparatus according to claim 60, wherein the shell is elongated along a horizontal axis, and wherein a first electrode is disposed substantially centrally along the horizontal axis, a second electrode is spaced apart from the first electrode along the horizontal axis in a first direction, and wherein the third electrode is spaced apart from the first electrode along the horizontal axis in a second direction opposite to the first.

[0075] 62. An apparatus according to clause 61, wherein the shell is elongated along a first horizontal axis, and wherein the first electrode is disposed substantially centrally along the horizontal axis and the second electrode is disposed along the horizontal axis toward the first end of the shell.

[0076] 63. A device according to any of the clauses herein, wherein the upper surface and the lower surface form a substantially boomerang or chevron shape.

[0077] 64. A device according to any of the clauses herein, wherein one or more electrodes are disposed on a first major surface of the device and one or more electrodes are disposed on an opposing second major surface of the device.

[0078] 65. A device according to any of the clauses herein, wherein one or more electrodes are disposed on a first surface of the device and one or more electrodes are disposed on an opposing second surface of the device.

[0079] 66. A device according to any of the clauses herein, wherein the device is configured to be delivered to a target site via a trocar introducer.

[0080] 67. A method for detecting and / or predicting stroke, the method comprising:

[0081] obtaining physiological data from a patient via a sensor device;

[0082] analyzing the physiological data; and

[0083] Based on the analysis, a patient stroke indicator is provided.

[0084] 68. A method according to any of the clauses herein, wherein the sensor device comprises a device according to any of the preceding clauses.

[0085] 69. A method according to any of the clauses herein, wherein obtaining the physiological data includes detecting brain activity data.

[0086] 70. A method according to any of the clauses herein, wherein obtaining physiological data includes detecting electrical activity through electrodes of the sensor device, the electrical activity corresponding to activity in at least one of the P3, Pz and P4 brain regions of the patient.

[0087] 71. A method according to any of the clauses herein, wherein obtaining physiological data includes detecting electrical activity through electrodes of the sensor device, the electrical activity corresponding to activity in each of the P3, Pz and P4 brain regions of the patient.

[0088] 72. A method according to any of the clauses herein, wherein obtaining physiological data includes obtaining the physiological data when the sensor device is positioned at or adjacent to a posterior portion of the patient's neck or base of the skull.

[0089] 73. A method according to any of the clauses herein, wherein obtaining physiological data includes obtaining the physiological data while the sensor device is positioned above a shoulder of the patient.

[0090] 74. A method according to any of the clauses herein, wherein obtaining physiological data includes obtaining the physiological data when the sensor device is positioned at or below the occipital bone of the patient.

[0091] 75. A method according to any of the clauses herein, wherein obtaining physiological data includes obtaining the physiological data when the sensor device is subcutaneously implanted in the patient.

[0092] 76. A method according to any one of clauses 67 to 75, wherein obtaining the physiological data further comprises obtaining additional physiological data from the patient using one or more additional sensor devices, the one or more additional sensor devices comprising at least one of the following: an accelerometer, a heart rate monitor, a blood pressure monitor, a respiratory monitor, an electrocardiogram (ECG) sensor, a galvanic skin sensor, or a thermometer.

[0093] 77. A method according to any of the clauses herein, wherein obtaining physiological data from the patient comprises:

[0094] providing prompts to the patient to perform one or more actions; and

[0095] Patient physiological data is recorded while the patient attempts to perform the one or more actions.

[0096] 78. A method according to any of the clauses herein, wherein the one or more actions include at least one of: raising a limb, moving a hand or finger, speaking, blinking, or making a facial expression.

[0097] 79. A method according to any of the clauses herein, wherein providing the patient stroke indicator comprises classifying an identified stroke as ischemic or hemorrhagic.

[0098] 80. The method of any of the clauses herein, wherein providing the patient stroke indicator comprises determining whether the patient has suffered a stroke.

[0099] 81. A method according to any of the clauses herein, wherein providing the patient stroke indicator comprises determining a risk that the patient will suffer a stroke.

[0100] 82. A method according to any of the clauses herein, wherein providing the patient stroke indicator comprises determining the location of the stroke.

[0101] 83. A method according to any of the clauses herein, wherein providing the patient stroke indicator comprises providing a confidence score associated with determining that the patient has a stroke.

[0102] 84. A method according to any of the clauses herein, wherein providing the patient with a stroke indicator comprises providing a recommended therapeutic action accompanying a stroke determination.

[0103] 85. The method of any of the clauses herein, wherein providing the patient stroke indicator comprises transmitting an alert to an emergency healthcare provider.

[0104] 86. A method for detecting stroke during at least one of intraoperative or perioperative periods in a patient, the method comprising:

[0105] obtaining physiological data from the patient during at least one of intraoperative or preoperative periods of the patient via a sensor device positioned over a shoulder of the patient;

[0106] analyzing the physiological data; and

[0107] Based on the analysis, a patient stroke indicator is provided.

[0108] 87. A non-transitory computer-readable medium storing instructions that, when executed by one or more processors of a computing device, cause the computing device to perform operations comprising:

[0109] A method according to any of the preceding clauses.

[0110] 88. A computing device comprising:

[0111] one or more processors; and

[0112] A non-transitory computer-readable medium as described in clause 79.

[0113] Other features and advantages of the present invention will be explained in the following description, and part of it will be obvious from the description, or can be understood through the practice of the subject technology. The advantages of the present invention are realized and obtained by the structures specifically pointed out in the written description and its claims and the attached drawings.

[0114] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory and are intended to provide further explanation of the present technology as claimed. BRIEF DESCRIPTION OF THE DRAWINGS

[0115] Many aspects of the present disclosure may be better understood with reference to the following drawings. The components in the drawings are not necessarily to scale. Instead, emphasis is placed on clearly illustrating the principles of the present technology. For ease of reference, throughout this disclosure, the same reference numerals may be used to identify the same or at least generally similar or similar components or features.

[0116] Figure 1 is a schematic diagram of a stroke detection system configured in accordance with an embodiment of the present technology.

[0117] Figure 2A Depicted is a top view of a sensor device in accordance with an embodiment of the present technology.

[0118] Figure 2B Describes the technology Figure 2A Side view of the sensor device shown in .

[0119] Figure 2CDepicted is a top view of another embodiment of a sensor device in accordance with the present technology.

[0120] Figure 2D Depicted is a side view of another embodiment of a sensor device in accordance with the present technology.

[0121] Figure 2E Depicted is a side view of another embodiment of a sensor device in accordance with the present technology.

[0122] Figure 2F Depicted is a side view of another embodiment of a sensor device in accordance with the present technology.

[0123] Figure 3 Another sensor device in accordance with an embodiment of the present technology is depicted.

[0124] Figure 4 Depicted are exemplary target areas for sensor devices of the present technology.

[0125] Figure 5 is a graph of 10-20 maps measured by an electroencephalogram (EEG) sensor.

[0126] Figure 6 Depicted are example EEG spectral power data from a sensor device in accordance with embodiments of the present technology.

[0127] Figure 7 Depicted is the output of a gradient boosted classifier for stroke determination in accordance with an embodiment of the present technology.

[0128] Figure 8 is a flow chart of a method for making a stroke determination according to an embodiment of the present technology.

[0129] Fig. 9 is a flow chart of another method for making a stroke determination in accordance with an embodiment of the present technology.

[0130] Fig.10 is a flow chart of another method for making a stroke determination in accordance with an embodiment of the present technology.

[0131] Fig.11 is a flow chart of another method for making a stroke determination in accordance with an embodiment of the present technology.

[0132] Fig. 12A and 12B is a conceptual diagram depicting a sensor device including a housing and a plurality of flexible electrode extensions extending from the housing. DETAILED DESCRIPTION

[0133] It can be difficult to determine whether a patient is having a stroke or has had a stroke. Current diagnostic techniques typically involve evaluating a patient for visible symptoms, such as paralysis or numbness of the face, arms, or legs and difficulty walking, speaking, or understanding. However, these techniques can result in strokes going undiagnosed, particularly milder strokes that leave the patient relatively mobile after a cursory assessment. Even for relatively mild strokes, it is important to treat the patient as quickly as possible because the effectiveness of treatment for stroke patients is very time-dependent. Therefore, improved methods for detecting stroke are needed.

[0134] Embodiments of the present technology enable detection of stroke by obtaining patient physiological data using a sensor device and analyzing the physiological data to provide an indication of a stroke, as described in more detail below. For example, the monitoring device can be equipped with electrodes (e.g., electroencephalogram (EEG) electrodes) that can be used to sense and record the patient's brain electrical activity. The monitoring device can be implantable (e.g., subcutaneous) or configured to be placed on the patient's skin.

[0135] Conventional EEG electrodes are typically positioned over a large portion of a user's scalp. While electrodes in this area work well for detecting electrical activity in a patient's brain, there are certain disadvantages. Sensors in this location interfere with the patient's movement and daily activities, and therefore cannot be monitored for extended periods of time. In addition, implanting electrodes under a patient's scalp is difficult and may cause significant discomfort to the patient. To address these and other shortcomings of conventional EEG sensors, embodiments of the present technology include a sensor device configured to record electrical signals in an area adjacent to a patient's neck or a posterior portion of the base of the patient's skull. In this location, implantation under the patient's skin is relatively simple, and temporary application of a wearable sensor device (e.g., coupled to a bandage, clothing, strap, or adhesive member) does not unduly interfere with the patient's movement and activities.

[0136] However, EEG signals detected by electrodes placed at or near the back of a patient's neck may be relatively noisy. For example, electrical signals associated with brain activity may be mixed with electrical signals associated with cardiac activity (e.g., ECG signals) and electrical signals associated with muscle activity (e.g., EMG signals), as well as other artifacts. Therefore, in some embodiments, the sensor data may be filtered or otherwise manipulated to separate brain activity data (e.g., EEG signals) from other electrical signals (e.g., ECG signals, EMG signals, etc.).

[0137] As described in more detail below, in some embodiments, the sensor data may be analyzed to make a stroke determination, including using a classification algorithm, which itself may be derived using machine learning techniques applied to a database of known stroke patient data. The one or more detection algorithms may be passive (involving measurements of a patient at complete rest) or active (involving prompting the patient to perform a potentially impaired function, such as moving a specific muscle group (e.g., lifting an arm, moving a finger, moving a facial muscle, etc.) and / or speaking while the electrical response is recorded).

[0138] Example System

[0139] The following discussion provides a brief general description of the environment suitable for implementing the present technology therein. Although not required, the various aspects of the present technology are described in the general context of computer executable instructions such as routines performed by general-purpose computers. The various aspects of the present technology can be embodied in a special-purpose computer or a data processor, and the special-purpose computer or the data processor is specifically programmed, configured or constructed to perform one or more computer executable instructions in the computer executable instructions explained in detail herein. The various aspects of the present technology can also be put into practice in a distributed computing environment in which a task or module is performed by a remote processing device, and the remote processing device is connected by a communication network (e.g., a wireless communication network, a wired communication network, a cellular communication network, the Internet, a short-range radio network (e.g., via Bluetooth)). In a distributed computing environment, a program module can be located in both a local memory storage device and a remote memory storage device.

[0140] Computer-implemented instructions, data structures, screen displays, and other data under various aspects of the technology may be stored or distributed on computer-readable storage media including magnetic or optically readable computer disks, as microcode on semiconductor memories, nanotechnology memories, organic or optical memories, or other portable and / or non-transitory data storage media. In some embodiments, various aspects of the technology may be distributed over the Internet or over other networks (e.g., Bluetooth networks) over a propagating signal on a propagating medium (e.g., one or more electromagnetic waves, sound waves) for a period of time, or may be set on any analog or digital network (packet switching, circuit switching or other schemes).

[0141] Figure 1 1 is a schematic diagram of a system 100 configured according to an embodiment of the disclosed technology. Although for the purpose of explanation, the system 100 is shown as having certain devices, in various examples, Figure 1 Any one or more of the devices shown in the figure may be omitted. Figure 1The devices shown in are shown to include certain components, but in various examples, any one or more specific components within these devices may be omitted (e.g., sensor device 110 may omit accelerometer 115). In addition, any device may include additional components not specifically shown here.

[0142] System 100 can be configured to sense physiological patient data and analyze the data to make a stroke determination. In one example, system 100 includes sensor device 110, which is configured to be implanted in the target site of the patient or placed on the patient's skin at the target site. The sensor device can be a relatively small device and can be placed (e.g., inserted) under or on the skin at the back of the patient's neck or skull base. Other target sites that can locate the sensor device include other locations on the head, such as above the temporal bone. As described in more detail below, sensor device 110 can detect one or more physiological parameters of the patient (e.g., electrical activity, heart rhythm data, motion data, etc. corresponding to brain activity in a specific area of ​​the patient's brain). Sensor device 110 can be communicatively coupled to external device 150, such as by wireless connection. In some embodiments, external device 150 can be a mobile device (e.g., a smart phone, a tablet computer, a smart watch, etc.) or other computing devices that the patient can interact with. In some instances, for example, when the sensor device 110 is used intraoperatively or preoperatively, the external device 150 may be another medical device, such as a ventilator, a heart-lung machine, an EKG machine, or other operating room equipment, or another patient monitoring or therapy device or computing device in an operating room or elsewhere in a hospital. In operation, the patient may receive an output or instruction from the external device 150 that is based at least in part on data received from the sensor device 110 at the external device 150. For example, the external device 150 may provide an alert to the patient or another entity (e.g., a call center) based on a stroke indication provided by the sensor device 110. Additionally or alternatively, the external device 150 may output a user prompt that may be synchronized with data collection by the sensor device 110. For example, the external device 150 may instruct the user to raise an arm, make a facial expression, etc., and the sensor device 110 may record physiological data when the user performs the requested action. Additionally, the external device 150 itself can analyze the patient (e.g., the patient's activity or condition in response to such prompts), such as using a camera to detect facial drooping, a microphone to detect slurred speech, or any other signs of stroke. In some embodiments, such signs can be compared to pre-stroke input (e.g., a stored baseline facial image or voiceprint with a baseline voice recording).

[0143] The sensor device 110 and / or the external device 150 may also be communicatively coupled to one or more external computing devices 180 (e.g., via the network 170). In some instances, the external computing device 180 may take the form of a server, a personal computer, a tablet computer, or other computing device associated with one or more healthcare providers (e.g., a hospital, a medical data analysis company, a device manufacturer, etc.). These external computing devices 180 may collect data recorded by the sensor device 110 and / or the external device 150. In some embodiments, this data may be anonymized and aggregated to perform large-scale analysis (e.g., using machine learning techniques or other suitable data analysis techniques) to develop and improve stroke detection algorithms using data collected by a large number of sensor devices 110. In addition, the external computing device 180 may transmit data to the external device 150 and / or the sensor device 110. For example, an updated algorithm for making a stroke determination may be developed by the external computing device 180 (e.g., using machine learning or other techniques), and then provided to the sensor device 110 and / or external device 150 over a network (e.g., as a wireless update), and installed on the sensor device 110 and / or external device 150.

[0144] In some embodiments, the system 100 may also include additional implantable devices, such as an implantable cardiac monitor, an implantable pacemaker, an implantable cardioverter defibrillator, a cardiac resynchronization therapy (CRT) device (e.g., a CRT-D defibrillator or a CRT-P pacemaker), a neurostimulator, a deep brain stimulation device, a neurostimulator, a drug pump (e.g., an insulin pump), a blood glucose monitor, or other devices. Other devices that can support and enhance a personal ecosystem to reduce the risk of stroke include fitness monitors, nutrition devices, and the like. Additionally or alternatively, the stroke detection device can be used in conjunction with other disease therapies that have a high risk of stroke as an adverse event (e.g., LVAD devices, TAVI / TAMR surgery, bariatric / gastric surgery, etc.). Another example of an adjunctive therapy for a high risk of stroke is ventilation, such as during treatment of COVID-19 or other infections or acute respiratory distress syndrome (ARDS).

[0145] As previously described, the sensor device 110 is configured to be coupled to the patient to record physiological data related to the determination of stroke. For example, the sensor device 110 can be implanted in the patient, can be directly placed on the patient's skin (for example, maintained in place by an adhesive or fastener), or can be removably worn by the patient. The sensor device 110 includes a sensing component 111, which can include many different sensors and / or sensor types. For example, the sensing component 111 can include multiple electrodes 113, an accelerometer 115, and optional other sensors 117. Examples of other sensors 117 include blood pressure sensors, pulse oximeters, ECG sensors or other heart recording devices, EMG sensors or other muscle activity recording devices, temperature sensors, skin galvanometers, hygrometers, altimeters, gyroscopes, magnetometers, proximity sensors, Hall effect sensors, or any other suitable sensors for monitoring the physiological characteristics of patients. These specific sensing components 111 are exemplary, and in various embodiments, the sensors used can vary.

[0146] Electrode 113 may be configured to detect electrical activity, such as brain activity (e.g., EEG data), cardiac activity (e.g., ECG data), and / or muscle activity (e.g., EMG data). Electrode 113 may be formed of any one or more suitable conductive materials to enable the electrode to perform electrical measurements on a patient. In some embodiments, sensor device 110 may be configured to analyze data from electrode 113 to extract both brain activity data (e.g., EEG signals) and cardiac activity data (e.g., ECG signals). Brain activity data may be evaluated to provide a stroke determination or other brain condition assessment, while cardiac activity data may be evaluated to provide a cardiac condition assessment or to detect certain cardiac events (e.g., heart rate variability, arrhythmias (e.g., tachyarrhythmias or bradycardia), episodes of ventricular or atrial fibrillation, etc.

[0147] In some embodiments, the sensor device 110 is configured to analyze data from the electrodes 113 to extract brain activity data and discard or reduce any contribution from cardiac or muscle activity. In some embodiments, the electrodes 113 are configured to be placed on the patient's skin. In such embodiments, the electrodes 113 may include protrusions (e.g., microneedles or other suitable structures) configured to at least partially penetrate the patient's skin to improve detection of subcutaneous electrical activity. In some embodiments, the sensor device 110 may be configured to be implanted in the body (e.g., subcutaneously), so the electrodes 113 may include a conductive surface exposed along at least a portion of the sensor device 110 to detect electrical activity in the body.

[0148] The sensor device 110 can be configured to calculate physiological characteristics related to one or more electrical signals received from the electrodes 113. For example, the sensor device 110 can be configured to determine the presence or absence of a stroke or other neurological condition based on the electrical signals through an algorithm. In some embodiments, the sensor device 110 can make a stroke determination for each electrode 113 (e.g., channel), or can make a stroke determination using electrical signals obtained from two or more selected electrodes 113.

[0149] In various embodiments, the number and configuration of electrodes 113 can vary. For example, the sensor device 110 can include at least 2, at least 3, at least 4, at least 5 or more electrodes 113 in the array. In some embodiments, the sensor device 110 includes less than 6, less than 5, less than 4 or less than 3 electrodes 113 in the array. As described in more detail below, although conventional EEG arrays include a large number of electrodes placed above the patient's head, some embodiments of the present technology include a relatively small number of electrodes (e.g., three electrodes) configured to be placed in the posterior portion of the patient's neck or skull or another target area of ​​the patient. In this position, the electrical data collected by these electrodes 113 can correspond to brain activity in the area determined to be of interest for stroke determination (e.g., in the case of the posterior portion of the patient's neck or skull, the P3, Pz and / or P4 areas).

[0150] In some embodiments, the electrodes 113 may all be located within a single housing of the sensor device 110. In some embodiments, the electrodes 113 may extend away from the housing of the sensor device 110 and be connected by leads or other connecting components. For example, the sensor device 110 may include a housing that covers certain components (e.g., power supply 119, communication link 121, processing circuit system 123 and / or memory 125), and the electrodes 113 (and / or other sensing components 111) may be connected to the housing by electrical leads or other suitable connections. In such a configuration, the electrodes 113 may be positioned at a position spaced apart from the housing of the sensor device 110. In some embodiments, the electrodes 113 may be placed in a separate housing, which in turn is coupled to the housing containing other components of the sensor device 110. Such a configuration in which a plurality of housings (or sub-housings) are coupled together by flexible or other connectors may help place the sensor device 110 in a desired position to improve the comfort of the patient. In addition, this may help place the electrodes 113 in a desired position for detecting clinically useful brain activity data.

[0151] The accelerometer 115 can be configured to detect patient movement. In some embodiments, the patient movement data collected by the accelerometer 115 can be used to make a fall determination. Fall detection may be particularly valuable when evaluating potential stroke patients because it has been found that a large proportion of patients admitted to the hospital for ischemic or hemorrhagic stroke have significant falls within 15 days after the stroke event. Therefore, in some embodiments, the sensor device 110 can be configured to initiate monitoring of brain activity through the electrodes 113 when a fall is detected using the accelerometer 115. In some embodiments, the sensing performed by the electrodes 113 can be modified in response to the fall determination, for example by increasing the sampling rate or making other modifications. In addition to fall detection, the accelerometer 115 (or similar sensor) can also be used to determine potential physical trauma caused by sudden acceleration and / or deceleration (e.g., vehicle accidents, sports collisions, concussions, etc.). These events may be thrombolytic and are precursors to stroke.

[0152] Sensor device 110 may also include power source 119 (e.g., battery, capacitor). In some embodiments, power source 119 may be rechargeable, for example using inductive charging or other wireless charging technology. Such rechargeability may facilitate long-term placement of sensor device 110 on or in a patient.

[0153] The communication link 121 enables the sensor device 110 to transmit data to and / or receive data from an external device (e.g., the external device 150 or the external computing device 180). The communication link 121 may include a wired communication link and / or a wireless communication link (e.g., Bluetooth, near field communication, LTE, 5G, Wi-Fi, infrared, and / or another radio transmission network).

[0154] Processing circuit system 123 may include one or more CPUs, ASICs, digital signal processing circuit systems, or any other suitable electronic components configured to process data from sensing component 111 and control the operation of sensor device 110. In some embodiments, processing circuit system 123 includes hardware particularly suitable for artificial intelligence and / or machine learning applications, such as a tensor processing unit (TPU) or other such hardware. In some embodiments, the processing circuit system of sensor device 110 may include one or more input protection circuits to filter electrical signals and may include amplifier / filter circuit systems to remove DC and high frequency components, one or more analog-to-digital (A / D) converters, or any other suitable components.

[0155] Sensor device 110 may further include memory 125, which may take the form of one or more computer-readable storage modules configured to store information (e.g., signal data, subject information or profiles, environmental data, data collected from one or more sensing components, media files) and / or executable instructions that may be executed by processing circuit system 123. Memory 125 may include, for example, instructions for analyzing patient data to determine whether a patient is experiencing or has recently or previously experienced a stroke. In some embodiments, memory 125 stores data used in the stroke detection techniques disclosed herein (e.g., signal data acquired from sensing component 111).

[0156] As described above, in some embodiments, the sensor device 110 can also communicate with an external device 150. The external device 150 can be, for example, a smart watch, a smart phone, a laptop, a tablet, a desktop PC, or any other suitable computing device, and can include one or more features, applications, and / or other elements commonly found in such devices. For example, the external device 150 can include a display 151, a communication link 153 (e.g., a wireless transceiver, which can include one or more antennas for wireless communication with, for example, other devices, websites, and the sensor device 110). Communication between the external device 150 and other devices can be performed through, for example, a network 170 (which can include the Internet, public and private intranets, local or extended Wi-Fi networks, cellular towers, plain old telephone systems (POTS) and the like), direct wireless communication, and the like. The external device 150 can additionally include well-known input components 131 and output components 133, including, for example, a touch screen, a keypad, a speaker, a camera, and the like.

[0157] In operation, the patient may receive an output or instruction from the external device 150 that is based at least in part on data received at the external device 150 from the sensor device 110. For example, the sensor device 110 may generate a stroke indication based on an analysis of data collected by the sensing component 111. The sensor device 110 may then instruct the external device 150 to output an alert to the patient (e.g., via the display 151 and / or output 157) or another entity. In some embodiments, the alert may be displayed to the user (e.g., via the display 151 of the external device) and may also be transmitted to appropriate emergency medical response services (e.g., a 9-1-1 call may be placed using location data from the external device 150, which is used to direct responders to locate the patient) and / or other healthcare provider entities or individuals (e.g., a hospital, emergency room, or physician). In some embodiments, embedded circuitry (e.g., a GPS unit) that provides location data may be included within the sensor device 110.

[0158] Additionally or alternatively, the external device 150 may output user prompts that may be used in conjunction with the collection of physiological data by the sensor device 110. For example, the external device 150 may instruct the user to perform an action (e.g., raise an arm, make a facial expression, etc.), and the sensor device 150 may record physiological data as the user performs the requested action. In some embodiments, the external device 150 itself may analyze the patient's physiological parameters, such as using a camera to detect facial droop or other signs of stroke. In some embodiments, such physiological data collected by the external device 150 may be combined and analyzed together with data collected by the sensing component 111 to make a stroke determination.

[0159] As previously described, the external computing device 180 can take the form of a server or other computing device associated with a healthcare provider or other entity. The external device can include a communication link 181 (e.g., a component that facilitates wired or wireless communication with other devices directly or through the network 170), a memory 183, and a processing circuit system 185. These external computing devices 180 can collect data recorded by the sensor device 110 and / or the external device 150. In some embodiments, this data can be anonymized and aggregated to perform large-scale analysis (e.g., using machine learning techniques or other suitable data analysis techniques), thereby using data collected by a large number of sensor devices 110 associated with a large number of patients to develop and improve stroke detection algorithms. In addition, the external computing device 180 can transmit data to the external device 150 and / or the sensor device 110. For example, an updated algorithm for making a stroke determination can be developed by the external computing device 180 (e.g., using machine learning or other techniques), then provided to the sensor device 110 and / or the external device 150 through the network 170, and installed on the recipient device 110 / 150.

[0160] Example Sensor Device

[0161] Figure 2A 2 shows a plan view of an example sensor device 210. In some embodiments, the sensor device 210 may include the sensor device 210 described above. Figure 1 The sensor device 110 described and / or described below with respect to Figure 3 Some or all of the features of the sensor device 310 described herein and may include a combination of Figure 2AAdditional features described. In the illustrated example, the sensor device 210 includes a housing 201 that carries a plurality of electrodes 213a-c (collectively referred to as "electrodes 213") therein. In operation, the electrodes 213a-c can be placed in direct contact with tissue at a target site (e.g., in contact with the user's skin if placed on the user's skin, or in contact with subcutaneous tissue if the sensor device 210 is implanted). The housing 201 further encapsulates the electronic circuitry positioned within the sensor device 210 and protects the circuitry contained therein from bodily fluids. In various embodiments, the electrodes 213 can be placed along any surface of the sensor device 210 (e.g., a front surface, a rear surface, a left lateral surface, a right lateral surface, an upper surface, a lower surface, etc.), and the surface can then take any suitable form.

[0162] exist Figure 2A and 2B In an example of the present invention, the housing 201 can be a biocompatible material having a relatively planar shape, comprising a first major surface 203 configured to face the tissue of interest (e.g., facing forward when positioned at the back of the patient's neck), a second major surface 204 opposite the first major surface, and a depth D or thickness of the housing 201 extending between the first and second major surfaces. The housing 201 can define an upper side surface 206 (e.g., configured to face upward when the device 101 is implanted in or at the patient's neck) and an opposite lower side surface 208. The housing 201 can further include a central portion 205, a first lateral portion (or left side portion) 207, and a second lateral portion (or right side portion) 209. The electrodes 213 are distributed around the housing 201 such that the central electrode 213b is disposed within the central portion 205 (e.g., substantially centered along the horizontal axis of the device), the left electrode 213a is disposed within the left side portion 207, and the right electrode 213c is disposed within the right side portion 209. As shown, housing 201 may define a boomerang or chevron shape, wherein central portion 205 includes an apex, and first and second lateral portions 207, 209 extend laterally outward from central portion 205 and also extend at a downward angle relative to a horizontal axis of the device.

[0163] The configuration of the housing 201 can facilitate placement on the user's skin in a bandage-like form or for subcutaneous implantation. Therefore, a relatively thin housing 201 can be advantageous. Additionally, in some embodiments, the housing 201 can be flexible so that the housing 201 can be at least partially bent to correspond to the anatomical structure of the patient's neck (e.g., the left lateral portion 207 and the right lateral portion 209 of the housing 201 are bent forward relative to the central portion 205 of the housing 201).

[0164] In some embodiments, the housing 201 may have a length L of about 15-50 mm, about 20-30 mm, or about 25 mm. The housing 201 may have a width W of about 2.5-15 mm, about 5-10 mm, or about 7.5 mm. In some embodiments, the housing 201 may have a thickness of less than about 10 mm, about 9 mm, about 8 mm, about 7 mm, about 6 mm, about 5 mm, about 4 mm, or about 3 mm. In some embodiments, the thickness of the housing 201 may be between about 2-8 mm, about 3-5 mm, or about 4 mm. The housing 201 may have a volume of less than about 1.5 cc, about 1.4 cc, about 1.3 cc, about 1.2 cc, about 1.1 cc, about 1.0 cc, about 0.9 cc, about 0.8 cc, about 0.7 cc, about 0.6 cc, about 0.5 cc, or about 0.4 cc. In some embodiments, housing 201 may be sized to be implanted via a trocar introducer or any other suitable implantation technique.

[0165] As shown, the electrodes 213 carried by the housing 201 are arranged so that all three electrodes 213 are not located on a common axis. In such a configuration, the electrodes 213 can obtain a better signal vector than electrodes all aligned along a single axis. This is particularly useful in the sensor device 210, which is configured to be implanted in the neck while detecting electrical activity in the brain. In some embodiments, the electrode configuration also provides improved cardiac ECG sensitivity by integrating 3 potential signal vectors.

[0166] In the example shown in FIG. 2 , all three electrodes 213 are positioned on the first major surface 203 and are substantially flat and facing outward. However, in other examples, one or more electrodes 213 can utilize a three-dimensional configuration (e.g., bending around the edge of the device 210). Similarly, in other examples, one or more electrodes 213 can be placed on a second major surface relative to the first. Various electrode configurations allow configurations in which the electrodes 213 are positioned on both the first and second major surfaces. In other configurations, such as the configuration shown in FIG. 2 , the electrodes 213 are only placed on one of the major surfaces of the housing 201. The electrodes 213 can be formed by a variety of different types of biocompatible conductive materials (e.g., stainless steel, titanium, platinum, iridium or alloys thereof), and can utilize one or more coatings, such as titanium nitride or fractal titanium nitride. In some embodiments, the material selection of the electrode can also include a material with a high surface area (e.g., providing better electrode capacitance to obtain better sensitivity) and roughness (e.g., contributing to implant stability). Although the example shown in FIG. 2 includes three electrodes 213 , in some embodiments, sensor device 210 may include 1, 2, 4, 5, 6, or more electrodes carried by housing 201 .

[0167] Figure 2C Another exemplary embodiment is shown in which the electrodes 213 are not exposed along the first major surface 203 of the housing 201. Instead, the electrodes 213 can be exposed along the upper and lower surfaces (e.g., facing upward and facing downward when implanted at or on the patient's neck), such as Figure 2D and 2E shown. Figure 2F Another example is shown in which the housing 201 exhibits a curved configuration and in which electrodes may be placed along the upper and / or lower surface of the housing 201. In some embodiments, the curved configuration may improve patient comfort and more easily conform to the anatomy of the patient's neck region.

[0168] In operation, the electrodes 213 are used to sense electrical signals (e.g., EEG signals), which may be submuscular or subcutaneous. The sensed electrical signals may be stored in a memory of the sensor device 210, and the signal data may be transmitted to another device (e.g., Figure 1 The sensed electrical signals may be time-coded or otherwise associated with time data and stored in such a form that the recency, frequency, time of day, time span, or date (or a metric or statistic calculated based thereon) of a particular signal data point or data series may be determined and / or reported. In some examples, the electrodes 213 may additionally or alternatively be used to sense any biopotential signal of interest, such as an electrocardiogram (ECG), an intracardiac electrogram (EGM), an electromyogram (EMG), or a neural signal from any implanted location. Such data may be time-coded or time-related and stored in that form in the manner described above with respect to the EEG signal data.

[0169] Figure 3 Another example sensor device 310 is shown. In some embodiments, the sensor device 310 may include the sensor device 310 described above according to embodiments of the present technology. Figure 1 and 2. Some or all of the features of the sensor devices 110 and 210 described herein, and may include a combination of Figure 3 The additional features described. Figure 3In the example shown, the sensor device 310 can be embodied as a monitoring device having a housing 314, a proximal electrode 313a, and a distal electrode 313b (individually or collectively referred to as "electrode 313" or "electrode 313"). The housing 314 can further include a first major surface 318, a second major surface 320, a proximal end 322, and a distal end 324. The housing 314 encapsulates the electronic circuit system positioned inside the sensor device 310 and protects the circuit system contained therein from the influence of body fluids. The electrical feed-through provides electrical connection for the electrode 313. In one example, the sensor device 310 can be embodied as an external monitor, such as a patch that can be positioned on an external surface of a patient, or another type of medical device (e.g., instead of being an ICM), as further described herein.

[0170] exist Figure 3 In the example shown, the sensor device 310 is defined by a length "L", a width "W", and a thickness or depth "D". The sensor device 310 can be in the form of an elongated rectangular prism, wherein the length L is significantly greater than the width W, which in turn is greater than the depth D. In one example, the geometry of the sensor device 310 - specifically, the width W is greater than the depth D - is selected to allow the sensor device 310 to be inserted under the skin of a patient using minimally invasive surgery and to be maintained in a desired orientation during insertion. For example, Figure 3The device shown includes radial asymmetry (particularly rectangular shape) along the longitudinal axis, which keeps the device in the correct orientation after insertion. For example, in one example, the spacing between the proximal electrode 313a and the distal electrode 313b can be in the range of 30 millimeters (mm) to 55mm, 35mm to 55mm and 40mm to 55mm, and can be any range or a separate spacing of 25mm to 60mm. In some examples, the length L can be about 30mm to about 70mm. In other examples, the length L can be in the range of 40mm to 60mm, 45mm to 60mm, and can be any length or length range between about 30mm and about 70mm. In addition, the width W of the first major surface 18 can be in the range of 3mm to 10mm, and can be any single width or width range between 3mm and 10mm. The thickness of the depth D of the sensor device 310 can be in the range of 2mm to 9mm. In other examples, the depth D of the sensor device 310 can be in the range of 2 mm to 5 mm, and can be any single depth or depth range of 2 mm to 9 mm. In addition, the sensor device 310 according to the examples of the present disclosure has a geometry and size designed for ease of implantation and patient comfort. The examples of the sensor device 310 described in the present disclosure can have a volume of 3cc or less, 2cc or less, 1cc or less, 0.9cc or less, 0.8cc or less, 0.7cc or less, 0.6cc or less, 0.5cc or less, or 0.4cc or less, any volume between 3 and 0.4cc. In addition, in Figure 3 In the example shown, the proximal end 322 and the distal end 324 are rounded to reduce discomfort and irritation to surrounding tissue after insertion beneath the patient's skin.

[0171] exist Figure 3 In the example shown, once inserted into the patient, the first major surface 318 faces outward, toward the patient's skin, and the second major surface 320 is positioned opposite the first major surface 318. Thus, the first major surface and the second major surface can face in a direction along the patient's sagittal axis, and this orientation can be consistently achieved upon implantation due to the size of the sensor device 310. Additionally, the accelerometer, or the axis of the accelerometer, can be oriented along the sagittal axis.

[0172] The proximal electrode 313a and the distal electrode 313b are used to sense electrical signals (e.g., EEG signals), which may be submuscular or subcutaneous. The electrical signals may be stored in a memory of the sensor device 310, and the signal data may be transmitted via the integrated antenna 326 to another medical device, which may be another implantable device or an external device, such as the external device 150 ( Figure 1). In some examples, electrodes 313a and 313b may additionally or alternatively be used to sense any biopotential signal of interest from any implant location, such as an electrocardiogram (ECG), an intracardiac electrogram (EGM), an electromyogram (EMG), or a neural signal.

[0173] exist Figure 3 In the example shown, the proximal electrode 313a is proximal to the proximal end 322, and the distal electrode 313b is proximal to the distal end 324. In this example, the distal electrode 313b is not limited to a flat, outwardly facing surface, but can extend from the first major surface 318 around the rounded edge 328 or end surface 330 and onto the second major surface 320, so that the electrode 313b has a three-dimensional curved configuration. Figure 3 In the example shown, the proximal electrode 313a is positioned on the first major surface 318 and is substantially flat and outward facing. However, in other examples, the proximal electrode 313a can utilize a three-dimensional curved configuration of the distal electrode 313b, thereby providing a three-dimensional proximal electrode (not shown in this example). Similarly, in other examples, the distal electrode 313b can utilize a substantially flat outward facing electrode positioned on the first major surface 18, which is similar to the surface shown with respect to the proximal electrode 313a. Various electrode configurations allow for configurations in which the proximal electrode 313a and the distal electrode 313b are positioned on both the first major surface 18 and the second major surface 320. In other configurations, such as Figure 3 In the configuration shown, only one of the proximal electrode 313a and the distal electrode 313b is positioned on both major surfaces 318 and 320, and in still other configurations, both the proximal electrode 313a and the distal electrode 313b are positioned on one of the first major surface 318 or the second major surface 320 (e.g., the proximal electrode 313a is positioned on the first major surface 318 and the distal electrode 313b is positioned on the second major surface 320). In another example, the sensor device 310 can include electrodes 313 on both the first major surface 318 and the second major surface 320 located at or near the proximal and distal ends of the device, such that a total of four electrodes 313 are included on the sensor device 310. The electrodes 313 can be formed from a variety of different types of biocompatible conductive materials (e.g., stainless steel, titanium, platinum, iridium, or alloys thereof), and can utilize one or more coatings, such as titanium nitride or fractal titanium nitride. Although Figure 3 The example shown in includes two electrodes 313 , but in some embodiments, sensor device 310 may include 3, 4, 5, or more electrodes carried by housing 314 .

[0174] exist Figure 3In the example shown, the proximal end 322 includes a header assembly 332 that includes one or more of a proximal electrode 313a, an integrated antenna 326, an anti-migration protrusion 334, and a suture hole 336. The integrated antenna 326 is positioned on the same major surface (i.e., the first major surface 318) as the proximal electrode 313a and is also included as part of the header assembly 332. The integrated antenna 326 allows the sensor device 310 to transmit or receive data. In other examples, the integrated antenna 326 can be formed on a major surface opposite the proximal electrode 313a, or can be incorporated into the housing 314 of the sensor device 310. Figure 3 In the example shown, the anti-migration protrusion 334 is positioned adjacent to the integrated antenna 326 and protrudes from the first major surface 318 to prevent longitudinal movement of the device. Figure 3 In the example shown, the anti-migration protrusion 334 includes a plurality (e.g., six or nine) of small bumps or protrusions extending away from the first major surface 318. As described above, in other examples, the anti-migration protrusion 334 can be positioned on the major surface opposite the proximal electrode 313a or the integrated antenna 326. Figure 3 In the example shown, the head mount assembly 332 includes suture holes 336, which provide another means of securing the sensor device 310 to the patient to prevent movement after insertion. In the example shown, the suture holes 336 are positioned near the proximal electrode 313a. In one example, the head mount assembly 332 is a molded head mount assembly made of a polymeric or plastic material that can be integral with the main portion of the sensor device 310 or can be separated from the main portion.

[0175] Example Method

[0176] Figure 4 An exemplary target area 401 for positioning a sensor device (e.g., sensor devices 110, 210, 310 described elsewhere herein) is shown. As shown, target area 401 can be the back portion of the user's neck or skull. Target area 401 can be positioned above the patient's shoulder and at or below the patient's occipital bone. As previously described, the sensor device can be placed in this area by implantation (e.g., subcutaneously) or by placement on the patient's skin (wherein one or more electrodes of the sensor device are in direct contact with the patient's skin at or near target area 401). The sensor device can be similarly placed in other target areas, such as above the temporal bone or other skull areas.

[0177] While conventional EEG electrodes are placed on the patient's scalp, the present technology advantageously enables recording clinically useful brain activity data using electrodes positioned in a target area 401 (e.g., the back of the patient's neck). This anatomical region is well suited for implanting a sensor device and temporarily placing the sensor device on the patient's skin. In contrast, EEG electrodes positioned on the scalp are bulky and challenging to implant on a patient's skull and may cause significant discomfort to the patient. As described elsewhere in this document, conventional EEG electrodes are typically positioned on the scalp to more easily achieve a suitable signal-to-noise ratio for detecting brain activity. However, by using specific digital signal processing and dedicated classifier algorithms, clinically useful brain activity data can be obtained using sensors placed in the target area 401. Specifically, the electrodes can detect electrical activity corresponding to brain activity in the P3, Pz, and / or P4 regions (see Figure 5 ).

[0178] Although conventional methods for stroke detection using EEG rely on data from a large number of EEG electrodes, the inventors have found that clinically useful stroke determination can be performed using relatively few electrodes. In an experiment conducted by the inventors, data from a basic group of 56 patients (26 strokes and 30 non-strokes) were used. The EEG data were recorded at a sampling frequency of 500Hz over 3 minutes using a conventional EEG array between 1 and 22 hours after the event. The EEG data were detrended, then bandpass filtered (e.g., 6-40Hz was filtered to remove high-frequency noise), then re-referenced to Pz, wavelet denoised, and finally low-pass filtered below 25Hz. A total of 224 features were extracted using an EEG array with 16 contacts (Pz as ground) and 14 power boxes.

[0179] After feature extraction, a gradient boosting algorithm was trained on the data set to generate a classifier algorithm. The classifier was adjusted by reducing the features to only those related to stroke / non-stroke conditions. A sequential backward floating feature selection method was used, which used classifier performance indicators to sequentially remove individual features. The classifier was further adjusted by adjusting the frequency bins. The result of this analysis was five features that effectively distinguished between stroke and non-stroke conditions. These features were three frequency bins associated with the P3 electrode (5.5-7.5Hz, 8-9.5Hz, and 13.5-15Hz) and two frequency bins associated with the P4 electrode (5.5-7.5Hz and 13.5-15Hz). Figure 6 is the normalized power map of the relevant frequency intervals of the P3 and P4 electrodes. The relevant frequency intervals are shaded in the figure.

[0180] The resulting classifier successfully made stroke / non-stroke determinations with approximately 85% accuracy. Figure 7 These results are graphically presented using two features (P4 electrode in the range of 5.5-7.5 Hz along the x-axis and P3 electrode in the range of 7.5-10 Hz along the y-axis). The "+" and "-" symbols in the figure reflect the actual stroke / non-stroke condition, and the shaded area in the figure reflects the prediction made by the classifier. Figure 7 As shown, most of the "+" symbols are grouped within the predicted stroke region, while most of the "-" symbols are grouped within the predicted non-stroke region. Remarkably and surprisingly, the classifier achieves relatively high accuracy while relying on data from only three electrodes: P3, P4, and the ground electrode Pz. Thus, the inventors have successfully demonstrated that clinically useful stroke determination is possible without the need for data from the full array of 16 or more EEG electrodes found in conventional methods.

[0181] The accuracy of such a classifier can be improved by training the algorithm on a larger data set corresponding to stroke and non-stroke EEG readings. In addition, other physiological parameters can be added to the classifier model (e.g., fall detection determined using an accelerometer, a specific heart rhythm, gender, age, medical history, etc.). In addition, in some embodiments, the classifier can be used to distinguish between ischemic and hemorrhagic strokes. This distinction may be particularly useful because the interventions may be different. For example, an ischemic stroke may be treated with thrombectomy, while a hemorrhagic stroke may be treated with surgery or other suitable techniques.

[0182] Figure 8 8 is a flow chart of a method 800 for making a stroke determination. The process 800 may include, for example, a memory (e.g., Figure 1 125, 163 and / or 183) may be stored in memory 125, 163 and / or 183) and may be processed by one or more processors (e.g., Figure 1 In some embodiments, portions of process 800 are implemented by one or more hardware components (e.g., Figure 1 In some embodiments, portions of process 800 are performed by Figure 1 The method may be executed by a device external to the system 100.

[0183] As shown, process 800 begins at block 802 by placing a device at or near the back of the neck or base of the skull (e.g., Figure 4 In some embodiments, the EEG sensor data may include the use of electrodes as described above with respect to the target area 401). Figure 1-3 The electrical signals detected by the electrodes of the sensor device 110, 210 or 310 described herein. Such a device may be placed at the target area 401 ( Figure 4 ) at a specific location (e.g., implanted subcutaneously or positioned on the patient's skin).

[0184] Process 800 continues in box 804, filtering the EEG sensor data to remove ECG artifacts. Traditionally, EEG data is obtained by electrodes positioned on the scalp because it is a relatively noise-free signal acquisition location. Other anatomical locations (such as the back of the neck) are not used, not because there is no EEG signal, but because of the noisy environment and the overlap of frequency bands with other physiological signals (such as ECG). However, the latest techniques in machine learning / adaptive neural network processing enhance signal extraction capabilities (e.g., filtering out or reducing the contribution of ECG signals from EEG signals). One such method is described in "ECG Artifact Removal of EEG signal using Adaptive Neural Network" published in IEEE Xplore on May 27, 2019, which is incorporated herein by reference in its entirety. Similarly, electrical signals associated with muscle activity can also be filtered from EEG sensor data to remove such artifacts.

[0185] In block 806, a patient stroke indicator is provided. The patient stroke indicator may be, for example, a binary output of a stroke condition / non-stroke condition, a probabilistic indication of the likelihood of a stroke, or other output related to the patient's condition and the likelihood that a stroke has occurred. The stroke indicator may be calculated using a classifier model as described elsewhere herein. In addition to providing the patient stroke indicator, information or instructions may also be output to the patient or user. The patient may be displayed via a display device (e.g., Figure 1 151) outputs information or instructions. For example, if a stroke is identified in box 806, the system may provide instructions for sending the patient to a comprehensive stroke treatment center or otherwise marking the patient for treatment. In an embodiment where process 800 is performed while the patient is in an ambulance, process 800 may output information or instructions to emergency medical personnel (EMT) or other personnel behind the ambulance and / or to the ambulance driver. In some embodiments, the content displayed to the ambulance driver may include navigation information such as a map and instructions for taking the patient to a specific hospital or location with a stroke center. In an embodiment where process 800 is performed intraoperatively or before and after surgery, a stroke indicator may be provided by a sensor device or other external device in an operating room or hospital that wirelessly communicates with the sensor device.

[0186] In some embodiments, before, simultaneously with, or after providing the stroke indicator in block 806, method 800 may include triggering an automatic data transmission, such as an automatic data transmission of a stroke determination that may be output to the patient or another entity (e.g., a call center, emergency response personnel, etc.). The call center may contact the patient or the patient's designated contact to inquire about the patient's status and / or confirm that the patient has had a stroke. If the patient is confirmed to have had a stroke (or the call center is unable to contact the patient), the call center personnel may manually or automatically initiate a 9-1-1 emergency call.

[0187] Fig. 9 is a flow chart of another method 900 for making a stroke determination. The process 900 may include, for example, a memory (e.g., Figure 1 125, 163 or 183) may be stored in memory 125, 163 or 183) and may be processed by one or more processors (e.g., Figure 1 In some embodiments, portions of process 900 are executed by one or more hardware components (e.g., display 151, input 155, and / or output 157 of external device 150; sensing component 111 of sensor device 110; Figure 1 )) is performed. In some embodiments, portions of process 900 are performed by Figure 1 The method may be executed by a device external to the system 100.

[0188] In block 902, instructions are output to the patient to perform an action. For example, the patient may be directed to the external device 150 ( Figure 1 ) output instructions. The instructions may include patient prompts for the patient to perform specific actions or movements, such as raising an arm or leg, moving a hand or finger, speaking, smiling, identifying an image, clapping, etc. In some embodiments, these prompts may be provided continuously, and patient data may be obtained after each prompt while the patient responds (or fails to respond) to the specific instructions. In some embodiments, accelerometer data (e.g., an accelerometer within external device 150) may be used to monitor the patient's movement in response to the prompts provided.

[0189] In block 904, EEG sensor data is collected while the patient performs the actions contained in the instructions of block 902. In some embodiments, the EEG sensor data may be collected by placement at or adjacent to the back of the neck or base of the skull as described elsewhere herein (e.g., Figure 4 401) to collect EEG sensor data. In some embodiments, the EEG sensor data may include using the electrodes described above with respect to Figure 1-3The electrical signals detected by the electrodes of the described sensor device 110, 210, or 310. Such a device may be placed at a target area of ​​a patient (eg, implanted subcutaneously or positioned on the patient's skin).

[0190] In block 906, the sensor data is analyzed, and based on the analysis, the system can provide a patient stroke indicator. The analysis can include, for example, using a classifier algorithm as described elsewhere herein. The patient stroke indicator can be, for example, a binary output of a stroke condition / non-stroke condition, a probabilistic indication of the likelihood of a stroke, or other output related to the patient's condition and likelihood of having suffered a stroke. In some embodiments, if a stroke is indicated, the system can display a display device (e.g., Figure 1 2-4) outputs appropriate information or instructions. For example, if a stroke is identified in block 906, the system may provide instructions for sending the patient to a comprehensive stroke treatment center or otherwise marking the patient for treatment.

[0191] Fig.10 An example method 1000 for enhanced stroke detection is shown. As shown, process 1000 begins at block 1002 by placing a device at or near the back of the neck or base of the skull (e.g., Figure 4 In some embodiments, the EEG sensor data may include the use of electrodes as described above with respect to the target area 401). Figure 1-3 The electrical signals detected by the electrodes of the sensor device 110, 210 or 310 described herein. Such a device may be placed at the target area 401 ( Figure 4 ) at a specific location (e.g., implanted subcutaneously or positioned on the patient's skin).

[0192] Process 1000 continues at block 1004 by filtering the EEG sensor data to remove ECG artifacts, as described elsewhere herein. At block 1006, a classification algorithm is applied. The classification algorithm may be, for example, an algorithm suitable for artificial intelligence (e.g., machine learning, neural networks, etc.) applied to patient stroke data, for example, as described above with respect to Figure 6 and 7The algorithm described. Based on the classification algorithm, a stroke determination is made in the box, which can be binary or probabilistic. In box 1008, if a stroke is detected (e.g., the probability determination is below a predetermined threshold, such as a patient's stroke probability based on the classifier algorithm is less than 15%), the result can be output in box 1010. If a stroke is detected in box 1008 (e.g., the probability determination exceeds a predetermined threshold, such as an 85% probability of stroke), the process 1000 continues to box 1012 to apply a cause classifier. In some embodiments, such a cause classifier can determine (probabilistically or deterministically) the origin of the stroke (e.g., ischemic or hemorrhagic). Such a determination can be made based on the collected EEG sensor data alone or in combination with additional physiological parameters or patient data. In box 1014, a location classifier is applied. The classifier can determine the location of the stroke. For example, the location determination can include a left and right hemisphere determination (e.g., a binary output or a probabilistic result). In some embodiments, the location determination may include a more accurate mapping of brain regions assigned specific probabilities, such as a 70% probability that the stroke location is at a specific point on the patient's brain. The stroke location may be output along a spherical map or other suitable coordinate system to identify the location in the patient's brain. In block 1016, the results of the classifier may be output, such as by a graphical display, an automatic alert to a call center, the patient, or other entity, etc. In some embodiments, the output may include a graphical representation of the stroke location, for example as an overlay location on a graphical representation of the brain.

[0193] In addition to outputting results, information or instructions can also be output to the patient or user. Figure 1 The system may output information or instructions to the display 151 of the ambulance. For example, if a stroke is identified in block 1008, the system may provide instructions for sending the patient to a comprehensive stroke treatment center or otherwise marking the patient for treatment. In embodiments where process 1000 is performed while the patient is in an ambulance, process 1000 may output information or instructions to emergency medical personnel (EMTs) or other personnel behind the ambulance and / or to the ambulance driver. In some embodiments, the content displayed to the ambulance driver may include navigation information such as a map and instructions for taking the patient to a specific hospital or location with a stroke center.

[0194] Fig.11An example method 1100 for detecting a stroke and / or determining a patient's risk of stroke is shown. In frames 1102, 1104, and 1106, the method 1100 includes collecting EEG sensor data, ECG sensor data, and accelerometer sensor data, respectively. In some embodiments, less data may be collected, and in other embodiments, additional data (e.g., body temperature, blood pressure, etc.) may also be collected. In frames 1108, 1110, and 1112, a classification algorithm is applied to the collected corresponding data, and based on the classification algorithm, a pre-stroke pattern is identified in frames 1114, 1116, and 1118. In some embodiments, the classification algorithm may be generated by an adaptive neural network model or other machine learning techniques, which are trained on a large sample of patient stroke data to identify a specific pattern indicating a pre-stroke state. Such data may be more easily collected by using an implantable monitoring device as described herein.

[0195] In box 1120, the identified patterns can be integrated or otherwise combined, and a stroke risk parameter can be calculated. The stroke risk can be based on physiological data and other patient parameters (e.g., gender, age, history of stroke or heart condition, etc.), and can include a probability output of the patient's stroke risk. If no stroke risk is identified in box 1122 (e.g., the stroke risk parameter is below a predetermined threshold), no action is taken in box 11124. Optionally, a "no risk" or "low risk" result can be output to the patient or other entity. If a stroke risk is identified in box 1122 (e.g., the stroke risk parameter exceeds a predetermined threshold), an alarm can be output in box 1126. Such an alarm can be provided to the patient (e.g., via an external device 150), a call center, the patient's medical team, or any other suitable entity.

[0196] Intraoperative monitoring for stroke has been implemented for high-risk surgeries such as transcatheter valve replacement and carotid endarterectomy. Technologies used for intraoperative stroke monitoring include monitoring EEG and / or ultrasound. Typically, intraoperative EEG monitoring for stroke is performed using large cranial electrode arrays (e.g., 12 to 64 electrodes) and large benchtop / rack-mounted signal acquisition and processing systems.

[0197] Using the signal processing and analysis techniques described herein, the systems and sensor devices described herein are configured to provide stroke indications using a relatively small number of electrodes (e.g., three electrodes) and a relatively small sensor device package. In some instances, the sensor device as described herein can be externally positioned in one of the target locations described herein for intraoperative and / or preoperative stroke monitoring during the surgical operation or preoperative and postoperative period of the patient. A possible target location is the back of the head or neck. Other possible target locations include the forehead and / or neck behind the ear, which can eliminate the need for shaving before placing the sensor device. In addition, such sensor devices, systems and techniques can be used for temporary stroke monitoring during other periods when the patient may be at a relatively high risk of stroke, such as when using a ventilator or during certain periods of arrhythmia caused by complications caused by Covid-19 or other infections.

[0198] Some exemplary sensor devices may include one or more flexible electrode extensions or leads attached to their housing to allow the housing to be positioned at one of these target locations, as well as electrodes on the extensions at another of these target locations. The electrode extensions are flexible in nature and can conform to the anatomical structures of the neck and / or head. In addition, the length and flexibility of one or more electrode extensions can allow the electrodes on the extensions to be advantageously positioned near certain brain structures or locations, vascular structures, or other anatomical structures or locations, which can also help to improve signal quality, for example when the signal originates from or is affected by the structure. In addition, the electrode extensions can extend upward and / or downward from the sensor device housing to improve brain signal and / or cardiac signal sensing and detection. The improved signal quality can result in improved performance of algorithms that use such signals to predict or detect stroke.

[0199] For example, Fig. 12A and 12B A sensor device 1200 is depicted, which includes a housing 1202 and a plurality of flexible electrode extensions 1204a-1204e (collectively referred to as "electrode extensions 1204") extending from the housing. The sensor device 1200 includes electrodes 213a-213c distributed on the housing 1202. The housing 1202 and the electrodes 213a-213c can be substantially as described above with respect to Figures 2A-2F Each electrode extension 1204 includes a respective one of electrodes 213d-213h. In some examples, electrode extension 1204 may include more than one electrode and / or include other sensing elements instead of or in addition to electrodes. Fig. 12A and 12B The number and configuration of electrode extensions 1204 and electrodes 213 shown in FIG. 1 are examples only. Fig. 12BAs shown, sensor device 1200 may be implanted at a target location 401, which may be at the back of a patient's neck or skull, a temporal location, or other location on the patient as described herein.

[0200] In some examples, one or more electrode extensions 1204 may include a paddle having one or more electrodes 213 distributed thereon. In some examples, the electrodes 213 on the electrode extensions 1204 may include a ring electrode or a segmented ring electrode. Fig. 12A and 12B In the example shown, electrode extensions 1204a-1204c extend from housing 1202 in a first upward direction, and electrode extensions 1204d and 1204e extend from housing 1202 in a second downward direction opposite the first direction. When positioned at target site 401, the first direction may be toward the patient's upper skull and scalp, for example, to better sense brain signals, and the second direction may be toward the patient's neck and / or shoulders, for example, to better sense cardiac signals.

[0201] In some instances, the sensor device can take the form of a wearable patch, e.g., attached to the patient with an adhesive. In some instances, the patch can be configured to adhere to the patient during and / or before and after surgery. Depending on the target location for placement of the sensor device, a "wet electrode" (e.g., containing a conductive gel at the electrode-skin interface) may require shaving the hair on the head. "Dry electrodes" (e.g., not containing a conductive gel) can provide sufficient signal quality and are used in some instances. Dry electrodes can be integrated into a sensor device having a hat-like form factor or otherwise worn on the head, e.g., a baseball cap with dry electrodes on the back of the head.

[0202] in conclusion

[0203] The present disclosure is not intended to be exhaustive or to limit the present invention to the precise form disclosed herein. Although specific embodiments are disclosed herein for illustrative purposes, various equivalent modifications are possible without departing from the present invention as will be appreciated by those skilled in the relevant art. In some cases, well-known structures and functions have not yet been shown and / or described in detail to avoid unnecessarily obscuring the description of the embodiments of the present invention. Although the steps of the method may be presented in a particular order herein, in alternative embodiments, the steps may have another suitable order. Similarly, in other embodiments, certain aspects of the present invention disclosed in the context of a particular embodiment may be combined or omitted. In addition, although the advantages associated with those embodiments have been disclosed in the context of certain embodiments, other embodiments may also exhibit these advantages, and not all embodiments necessarily exhibit such advantages or other advantages disclosed herein to fall within the scope of the present invention. Therefore, the present disclosure and associated technologies may encompass other embodiments that are not explicitly shown and / or described herein.

[0204] Unless otherwise stated, all numerical values ​​used in the specification and claims should be understood to be modified by the term "about" in all cases. Therefore, unless otherwise indicated, the numerical parameters set forth in the following specification and the appended claims are approximate values ​​that may be attempted to change by the desired properties obtained by the present technology. At a minimum, and without attempting to limit the scope of the application of the principle of equivalents to the claims, each numerical parameter should at least be interpreted according to the number of reported significant digits and by applying general rounding techniques. In addition, all ranges disclosed herein should be understood to cover any and all sub-ranges contained therein. For example, the range "1 to 10" is included in any and all sub-ranges between (and including) a minimum value of 1 and a maximum value of 10, that is, any and all sub-ranges with a minimum value equal to or greater than 1 and a maximum value equal to or less than 10, such as 5.5 to 10.

[0205] Throughout this disclosure, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" include plural referents. Similarly, unless the word "or" is explicitly limited to mean only a single item other than other items in a list with two or more items, the use of "or" in such a list should be interpreted as including any single item in the (a) list, all items in the (b) list, or any combination of items in the (c) list. In addition, the use of the term "including" and the like throughout this disclosure means at least including the features described, so that any greater number of the same features and / or one or more other types of features are not excluded. Directional terms such as "upper", "lower", "front", "back", "vertical" and "horizontal" can be used in this article to express and clarify the relationship between the various elements. It should be understood that such terms do not represent absolute orientation. References to "one embodiment", "embodiment" or similar expressions herein mean that the specific features, structures, operations or characteristics described in conjunction with the embodiment may be included in at least one embodiment of the present invention technology. Therefore, the appearance of such phrases or expressions in this article does not necessarily all refer to the same embodiment. Furthermore, the various specific features, structures, operations, or characteristics may be combined in any suitable manner in one or more embodiments. For example, a master-slave configuration may utilize a mature chest implant location to obtain cardiac ECG information and a back of the head / neck implant location to obtain neuro-EEG information. These slave devices may be aggregated into a master device, which may be an external smart watch or smartphone, to provide stroke detection capabilities.

Claims

1. A stroke detection system, comprising: a sensor device configured to obtain physiological data from a patient, wherein the sensor device comprises a housing carrying a plurality of electrodes, wherein the housing is configured to be implanted subcutaneously above a shoulder of the patient, and the physiological data comprises brain electrical activity data and heart electrical activity data of the patient sensed by the plurality of electrodes; as well as a processing circuit system enclosed by the housing and configured to: analyzing the physiological data; as well as Based on the analysis, a patient stroke indicator is provided.

2. The system of claim 1, wherein the sensor device is configured to detect brain activity data corresponding to activity in at least one of the P3, Pz, and P4 brain regions through the plurality of electrodes.

3. A system according to claim 1 or 2, wherein the sensor device is configured to be placed at or adjacent to a posterior portion of the patient's neck or skull.

4. The system of any one of claims 1 to 3, wherein the processing circuit system is configured to: filtering the physiological data to separate the brain electrical activity data from the heart electrical activity data; and Both the brain electrical activity data and the heart electrical activity data are analyzed, and a stroke indicator is provided to the patient based on the analysis of both the brain electrical activity data and the heart electrical activity data.

5. The system of claim 4, wherein the processing circuit system is configured to apply a respective classification algorithm to each of the brain electrical activity data and the heart electrical activity data, and to provide the patient stroke indicator based on the classification performed by the algorithm.

6. The system of any one of claims 1 to 3, wherein the physiological data includes motion data, and wherein the processing circuitry is further configured to analyze the motion data and provide the patient a stroke indicator based on the analysis of the motion data.

7. The system according to any one of claims 1 to 3, further comprising a computing device configured to: communicating with the sensor device; and instructing the patient to perform one or more actions, Wherein the sensor device is configured to obtain at least a portion of the physiological data during an attempt by the patient to perform the one or more actions.

8. The system of any one of claims 1 to 3, wherein the patient stroke indicator is probabilistic.

9. The system of any one of claims 1 to 3, wherein the processing circuitry is further configured to provide a probabilistic etiology classification of the stroke.

10. The system of any one of claims 1 to 3, wherein the sensor device comprises at least one housing configured to be positioned on a posterior portion of a patient's neck or skull, wherein the at least one housing has a volume of less than about 1.2 cc.

11. The system according to any one of claims 1 to 3, wherein: The housing is configured for subcutaneous implantation at or below the patient's occipital bone.

Images