Systems and methods for detecting strokes

Abstract

A system for detecting strokes includes a sensor device configured to obtain physiological data from a patient, for example brain activity data. A computing device communicatively coupled to the sensor device is configured to receive the physiological data and compare it with reference data. The reference data can be patient data from an opposite brain hemisphere to the hemisphere being interrogated or the reference data can be non-patient data from stroke and normal patient populations. Based on comparison of the physiological data and the reference data, the system indicates whether the patient has suffered a stroke.

We claim: 
     
         1 . A system for detecting strokes, the system comprising:
 a sensor device comprising a sensor array including a plurality of electrodes and an accelerometer configured to acquire one or more physiological signals from the patient, the sensor device configured to obtain physiological data including at least brain activity data, cardiac activity data, and patient movement data from the physiological signals; and   a computing device communicatively coupled to the sensor device, the computing device configured to:   generate physiological analysis results by comparing the physiological data with reference data using a machine learning classifier, wherein the reference data comprises non-patient physiological data; and   based on the physiological analysis results, provide a patient stroke indicator.   
     
     
         2 . The system of  claim 1 , wherein the sensor device comprises at least one of a heart rate monitor or an electrocardiography (ECG) sensor configured to obtain the cardiac activity data from the patient. 
     
     
         3 . The system of  claim 1 , wherein the sensor device comprises at least one of an electroencephalogram (EEG) array, a magnetoencephalography (MEG) array, a functional magnetic resonance imaging (fMRI) device, a positron emission tomography (PET) scanner, or a computed tomography (CT) scanner configured to obtain the brain activity data from the patient. 
     
     
         4 . The system of  claim 1 , further comprising one or more additional sensor devices configured to obtain additional physiological data from the patient, the one or more additional sensors including at least one of: a near-infrared sensor, an ultrasound sensor, a blood pressure monitor, a respiration monitor, an electromyography (EMG) sensor, a galvanic skin sensor, a thermometer, or a camera. 
     
     
         5 . The system of  claim 1 , wherein the computing device is further configured to:
 generate the physiological analysis results by comparing passive physiological data of the patient with non-patient passive physiological reference data using the machine learning classifier.   
     
     
         6 . The system of  claim 5 , wherein the passive physiological data of the patient comprises brain activity data from a first brain hemisphere of the patient associated with an impaired functionality experienced by the patient. 
     
     
         7 . The system of  claim 1 , wherein the computing device is further configured to:
 generate the physiological analysis results by comparing active physiological data of the patient with non-patient active physiological reference data using the machine learning classifier.   
     
     
         8 . The system of  claim 7 , wherein the active physiological data comprises physiological data obtained while the patient attempts to perform one or more actions. 
     
     
         9 . The system of  claim 8 , wherein the one or more actions comprises at least one of: lifting a limb, moving a hand or fingers, speaking, or smiling. 
     
     
         10 . The system of  claim 1 , the non-patient physiological reference data comprising a library of physiological data obtained from a plurality of stroke patients. 
     
     
         11 . A computer-readable medium storing instructions that, when executed by one or more processors of a computing device, cause the computing device to perform operations, the operations comprising:
 acquiring one or more physiological signals from the patient from a sensor device configured to obtain physiological data including at least brain activity data, cardiac activity data, and patient movement data from the physiological signals;   generating physiological analysis results by comparing the physiological data with reference data using a machine learning classifier, wherein the reference data comprises non-patient physiological data; and   based on the physiological analysis results, providing a patient stroke indicator.   
     
     
         12 . The computer-readable medium of  claim 11 , wherein the sensor device comprises at least one of a heart rate monitor or an electrocardiography (ECG) sensor configured to obtain the cardiac activity data from the patient. 
     
     
         13 . The computer-readable medium of  claim 11 , wherein the sensor device comprises at least one of an electroencephalogram (EEG) array, a magnetoencephalography (MEG) array, a functional magnetic resonance imaging (fMRI) device, a positron emission tomography (PET) scanner, or a computed tomography (CT) scanner configured to obtain the brain activity data from the patient. 
     
     
         14 . A method for detecting strokes, comprising:
 acquiring one or more physiological signals from the patient from a sensor device configured to obtain physiological data including at least brain activity data, cardiac activity data, and patient movement data from the physiological signals; and   generating physiological analysis results by comparing the physiological data with reference data using a machine learning classifier, wherein the reference data comprises non-patient physiological data; and   based on the physiological analysis results, providing a patient stroke indicator.   
     
     
         15 . The method of  claim 14 , wherein acquiring one or more physiological signals from the patient comprises obtaining cardiac activity data from the patient with at least one of: a heart rate monitor or an electrocardiography (ECG) sensor. 
     
     
         16 . The method of  claim 14 , wherein acquiring one or more physiological signals from the patient comprises obtaining brain activity data from the patient with at least one of: an electroencephalogram (EEG) array, a magnetoencephalography (MEG) array, a functional magnetic resonance imaging (fMRI) device, a positron emission tomography (PET) scanner, or a computed tomography (CT) scanner. 
     
     
         17 . The method of  claim 14 , further comprising generating the physiological analysis results by comparing active physiological data of the patient with non-patient active physiological reference data using the machine learning classifier. 
     
     
         18 . The method of  claim 17 , wherein the active physiological data comprises physiological data obtained while the patient attempts to perform one or more actions. 
     
     
         19 . The method of  claim 18 , wherein the one or more actions comprises at least one of: lifting a limb, moving a hand or fingers, speaking, or smiling. 
     
     
         20 . The method of  claim 14 , the non-patient physiological reference data comprising a library of physiological data obtained from a plurality of stroke patients.