Medical device and method for diagnosing and treating diseases

Abstract

A medical device includes a memory, a processor communicatively coupled to the memory, and the processor is configured to execute instructions to evaluate one or more patient data inputs related to a first specific disease, compare the one or more data inputs to a set of values ​​from at least one database using at least one computing algorithm, train at least one computing algorithm to estimate a diagnosis of the patient based on the first specific disease, determine a first diagnostic score for the patient for the specific disease using the at least one computing algorithm, diagnose the patient as having the specific disease when the first diagnostic score for the first specific disease is above a first value, and output a diagnosis provided to the patient.

Description

[0001] Cross-references to related applications / priorities

[0002] This application claims priority to U.S. Provisional Patent Application No. 62 / 723,593 filed on August 28, 2018 and U.S. Provisional Patent Application No. 62 / 816,239 filed on March 11, 2019, both of which are incorporated by reference into the present disclosure as if fully restated herein. If there is any conflict between the incorporated material and the specific ideas of the present disclosure, the latter shall prevail. Similarly, if there is any conflict between the definition of a word or phrase as understood in the technical field and the definition of the word or phrase specifically set forth in this disclosure, the latter shall prevail.

[0003] background

[0004] Proper diagnosis of patients is essential for healthcare professionals so that they can prescribe medications that are appropriate to treat the patient's illness and manage the condition appropriately. Some illnesses can be difficult to diagnose accurately, especially when a patient is first seen by a doctor. This problem is made more difficult when the illness has a limited window for effective treatment, such as stroke and heart attack. Because of the time required to diagnose the illness, emergency treatment, such as the "thrombolytic" recombinant tissue plasminogen activator (rtPA), is rarely given to different types of acute stroke, especially ischemic stroke. Despite the fact that stroke is the leading cause of disability and the second most common cause of death worldwide, nearly 800,000 strokes occur each year in the United States, costing healthcare professionals approximately $37 billion. More than 17 million strokes occur each year, of which 5.7 million are fatal, but healthcare professionals do not have the tools to diagnose and treat them quickly and effectively. Neurological emergencies, such as acute stroke, cause time-dependent brain damage. For the reasons stated above, there is an urgent but seemingly unsolvable need to find a device and method that can quickly and effectively diagnose and treat disease.

[0005] Overview

[0006] It is therefore an object of embodiments of the present invention to overcome the above-mentioned deficiencies and drawbacks associated with the current technology.

[0007] Currently, ischemic stroke can be treated with recombinant tissue plasminogen activator in the pre-hospital setting. When a patient is diagnosed with acute stroke and sent to the hospital, the patient only needs to undergo routine neuroimaging to determine the final diagnosis, and then the patient can be treated with recombinant tissue plasminogen activator infusion therapy. In the case where there is a strong correlation between earlier treatment and better prognosis, the device disclosed herein can establish a diagnosis on the spot without the need for a resident doctor to evaluate, thereby saving an average of 83 minutes for recombinant tissue plasminogen activator infusion therapy. The time saved means that the effectiveness of recombinant tissue plasminogen activator treatment is improved by 37%, and the number of patients treated can be increased by 2 times. Because the medical device disclosed herein is scalable, it can be used in, for example, approximately 81,000 ambulances in the United States and 178,000 ambulances in Europe.

[0008] According to one embodiment, the artificial intelligence diagnosis (AID) medical device of the present disclosure has natural language and visual recognition / computer vision, is a cloud-based artificial neural network, and diagnoses neurological emergencies by communicating directly with the patient. After the diagnosis is completed, the medical device will convey treatment information to medical staff and guide the transportation of the patient to the appropriate hospital. Patients who do not establish a clear diagnosis may be sent to the on-duty neurologist for evaluation.

[0009] The term "disease" as used herein is a general synonym and is used interchangeably with the terms "disorder" and "condition" (medical condition) because they all reflect an abnormal condition manifested by impairment of the normal function of a human or animal or a part thereof, usually manifested by obvious signs and symptoms, and resulting in a reduction in the survival time or quality of life of the human or animal, and may also include dysfunction or abnormal function of an organ, part, structure or system of the body caused by genetic or developmental errors, infection, poisoning, nutritional deficiency or imbalance, toxic or adverse environmental factors, illness, discomfort, and pain.

[0010] The invention disclosed herein relates to methods, systems and medical devices, the device comprising a memory, a processor connected to the memory and capable of communication; and the processor is configured to execute instructions to evaluate one or more input patient data related to a first specific disease; the processor uses at least one computing algorithm to compare the one or more input data with a set of values ​​from at least one database; train at least one computing algorithm to estimate the patient's diagnosis based on the first specific disease; use at least one computing algorithm to determine a first diagnostic score for the patient for the specific disease; when the first diagnostic score for the first specific disease is higher than the first value, diagnose the patient as having the specific disease, and display or provide the patient's diagnosis as an output. According to another embodiment, the processor is also configured so that when the diagnostic score is lower than a second value and the second value is lower than the first value, execute instructions to diagnose the patient as not having the specific disease. According to another embodiment, the processor is also configured so that when the diagnostic score is between the first value and the second value, execute instructions to display that the patient's diagnosis is inconclusive. According to yet another embodiment, the medical device further comprises the following: a reflector capable of measuring a reflective area of ​​at least one square foot; a clock; a refrigerator; a toilet; a chair; a bed; a television; a microwave; a floor lamp or a counter lamp; and a ceiling mounted lamp. According to another embodiment, the medical device further comprises straps for securing the medical device on a wrist. According to another embodiment, the input data is one of demographic data, symptoms, medical history elements, examination results, and / or diagnostic test results, or a combination thereof. According to another embodiment, the input data is input by one of the patient or a third party and automatically acquired by the medical device. According to another embodiment, the medical device interacts with the patient via voice prompts. According to another embodiment, when a positive symptom of a specific disease is present, the likelihood of being diagnosed with the disease increases, and when different symptoms of a similar disease are present, the likelihood of being diagnosed with the disease decreases. According to another embodiment, the first specific disease is one of a nervous system disorder, congestive heart failure, asthma, myocardial infarction, and infection. According to another embodiment, the specific disease is a neurological abnormality including one of acute ischemic stroke, transient ischemic attack, epileptic seizure, demyelinating disease, multiple sclerosis, brain trauma and brain tumor, or some combination thereof. According to another embodiment, the medical device automatically assesses the initial signs of one or more specific diseases of the patient and automatically triggers a more comprehensive assessment when initial signs are detected.According to another embodiment, abnormal body temperature is evaluated as an initial sign of infection, one or more changes in gait, speech and limb movement are evaluated as initial symptoms and signs of abnormal nervous system, one or both of changes in respiratory rate and pauses in speech are evaluated as initial symptoms and signs of worsening congestive heart failure, one or both of shortness of breath and dyspnea are evaluated as initial symptoms and signs of an impending or ongoing asthma attack, and one or more of grasping the chest, facial expression indicating pain, shortness of breath, flushing and / or sweating are evaluated as initial symptoms and signs of myocardial infarction. According to another embodiment, when the device diagnoses the patient with the first specific disease, the device determines the appropriate medication for the patient to take and notifies the patient, or determines the appropriate medication for the patient to take, notifies the patient, and then also directly dispenses the medication. According to another embodiment, the processor is further configured to execute instructions to assess the likelihood that the patient has a second disease that is similar to the first disease, and when the likelihood of diagnosing the similar symptoms as the other disease is greater than 25%, 50%, 75% or 90%, the diagnosis of the patient having the first disease is accompanied by an alarm. According to another embodiment, at least one of the computational algorithms includes one of an artificial neural network, a support vector machine (SVM), a Nu-SVM, a linear SVM, a naive Bayes (NB) algorithm, a Gaussian NB, a polynomial NB computational algorithm, a multi-class SVM, a directed acyclic graph SVM (DAGSVM), a structured SVM, a least squares SVM (LS-SVM), a Bayesian SVM, a transductive SVM, a support vector clustering algorithm (SVC), a classification SVM type 1 (C-SVM classification), a classification SVM type 2 (nu-SVM classification), a regression SVM type 1 (ε-SVM regression) and a regression SVM type 2 (nu-SVM regression). According to another embodiment, the processor is further configured to send the diagnosis to a medical facility via a wired or wireless network, and the medical device also includes a device for communicating the diagnosis via the network. According to another embodiment, the processor is further configured to execute a second calculation algorithm to determine a second diagnostic score for the patient when it is determined that the first diagnostic score is lower than the first value, or lower than the first value and higher than the second value, or lower than both the first value and the second value. According to another embodiment, the processor is further configured to execute multiple calculation algorithms, each of which uses data from multiple databases, and determines the diagnostic score of the patient's first specific disease for each calculation algorithm, and when the diagnostic scores of most or all calculation algorithms for a specific disease are higher than the first value, the patient is diagnosed with the specific disease.According to another embodiment, the processor is further configured to input one or more syndrome elements covered by a historical definition of a classic syndrome associated with a first specific disease, assign a syndrome element score proportional to a known or documented prevalence in a population of patients determined to have a classic syndrome, determine whether the patient has the syndrome element, calculate a diagnostic probability that the patient has a classic syndrome by dividing the total score representing the syndrome elements identified in the patient by the total score including all syndrome elements covered by the historical definition of the classic syndrome, and input the diagnostic probability of the classic syndrome as data input associated with the first disease. According to another embodiment, the first calculation algorithm and the second calculation algorithm are part of a plurality of calculation algorithms that improve the diagnosis of the patient in a serial manner. According to another embodiment, the calculation of the first calculation algorithm is based on a set of common data from a plurality of databases, and wherein the calculation of the second calculation algorithm is based on all data from a single database. According to another embodiment, the second calculation algorithm is selected from a plurality of calculation algorithms, wherein each of the plurality of calculation algorithms is based on all data from a different database, and wherein the selection of the second calculation algorithm is based on similarities between the patient's data input and the database data used by the second calculation algorithm.

[0011] According to another embodiment, the presently disclosed invention relates to devices, systems and methods, which include generating a numerical value representing a syndrome associated with a medical disease, wherein the syndrome can be a collection of symptoms, medical history elements, examination results and diagnostic test results associated with the medical disease; generating a set of biometric values ​​representing a patient, providing each value representing the syndrome and the set of biometric values ​​to a machine learning system to provide an output value representing the likelihood that the patient has the medical disease, and generating an output derived from the output value to a user.

[0012] The preferred embodiments of the medical device will be described in detail below, and the accompanying drawings will be attached to make the various objects, features, aspects and advantages of the present invention more apparent, in which the same numerical values ​​represent the same components. The present invention can solve one or more problems and deficiencies of the current technology discussed above. However, it is conceivable that the present invention may prove to be useful in solving other problems and deficiencies in many technical fields. Therefore, the claimed invention should not be construed as being limited to solving any specific problem or deficiency discussed herein. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate various embodiments of the invention and, together with the general description of the invention given above and the detailed description of the drawings given below, explain the principles of the invention. It should be understood that the drawings are not necessarily drawn to scale, as emphasis is placed on illustrating the principles of the invention. The invention will now be described by way of example with reference to the accompanying drawings, in which:

[0014] Figure 1 Schematic diagram of the external symptoms and internal neurological damage of the lateral medullary syndrome, a classic ischemic stroke syndrome. Damage to the lateral portion of the medulla oblongata (shown as the darker portion on the left side of the medulla in a cross-section) damages certain neuroanatomical structures, resulting in some unique symptoms and abnormalities on physical examination. Acute damage can be caused by blockage of a specific artery, also known as an ischemic stroke;

[0015] Figure 2 This is a table that summarizes various stroke-like episodes, excluding the symptoms of hemorrhagic stroke (intracranial hemorrhage, subarachnoid hemorrhage);

[0016] Figure 3 is a flow chart of the domain of computational algorithms serially employed in the diagnostic evaluation of a patient according to one embodiment of the presently claimed invention. In this example of one embodiment of the medical device, in order to identify ischemic stroke patients who are suitable for emergency treatment, patients who may be suffering from a certain neurological emergency are evaluated by a set of computational algorithms that operate in a logical sequence to refine the diagnosis. "Stroke-like episodes" cover neurological conditions that are often mistaken for strokes, such as epileptic seizures. AI = Artificial Intelligence. CA = Computational Algorithm. TIA = Transient Ischemic Attack;

[0017] 4A to 4D These are the potential steps involved in weighting / probability calculations for classic syndromes using scores, which can be used as discrete data inputs for the disclosed medical device artificial intelligence-based diagnostic medical device embodiments. 4A to 4D The logic of using score assessment to identify classic syndromes such as the lateral medullary ischemic stroke syndrome is described;

[0018] Figure 5The data collection tasks of the patient interface of the device are shown (left), and the order of which will be adjustable based on the importance of the data input. Then, one or more computational algorithms process the data inputs (right), where the strongest data may be the key definitions. Using the stroke diagnosis device as an example, the key definitions will include acuity of attack, persistence of symptoms, classic stroke syndrome, and stroke-like attack (gray box in the upper left center of the figure). Key definitions may require examination results as well as symptoms and precipitating events to be established; other data may have an impact, although to a lesser extent. Based on experience, the weights (W) of the data inputs are refined to improve the accuracy of the diagnosis rather than the "gold standard" doctor's diagnosis. The calculated probability (netj) can then determine the diagnosis of, for example, acute ischemic stroke (AIS), which can determine the treatment decision. EMS = Emergency Medical Service; EKG = Electrocardiogram; rtPA = recombinant tissue plasminogen activator.

[0019] Figure 6 A specific example of a hierarchical multi-level analysis device process having a primary computing algorithm group and three secondary computing algorithm groups is shown, wherein the primary computing algorithm first evaluates the patient and, if a diagnosis cannot be generated, then employs, applies or complies with a secondary computing algorithm to perform a different evaluation of the patient to arrive at a diagnosis;

[0020] Figure 7 A flow chart is shown that describes an example of a process for a domain of a computational algorithm used serially in a diagnostic evaluation of a patient in accordance with one embodiment of the presently claimed invention;

[0021] Figure 8 A schematic diagram showing a serial method for calculating a patient diagnosis using multiple databases and calculation algorithms;

[0022] Fig. 9 A schematic diagram showing a group approach to computing a patient diagnosis using multiple databases and computational algorithms;

[0023] Fig.10 A flow chart is shown that describes an example of a process for a domain of a computational algorithm serially employed in a diagnostic evaluation of a patient in accordance with one embodiment of the presently claimed invention;

[0024] Fig.11 An example of a neural network architecture is shown;

[0025] Fig.12 A specific schematic example of a neural network having four neural network layers is shown;

[0026] Fig.13 shows a schematic example of a computing device according to an embodiment of the disclosed subject matter;

[0027] Fig.14 An example block diagram of a medical device including a plurality of sensors coupled to a neural network via an interface is shown according to an embodiment of the disclosed subject matter;

[0028] Fig.15 The present invention shows a diagnostic medical device embodiment based on artificial intelligence, and potential steps involved in another embodiment of performing score weighting / probability calculation for classic syndromes as discrete data input;

[0029] Fig.16 and Fig.17 shows a specific example of a middle-aged African American male from New York City and the method of selecting a superior diagnostic algorithm based on the demographic characteristics of the patient's medical records similar to other patients in the database used to train the algorithm, taking advantage of non-geographic demographic similarity ( Fig.16 ) and geographical similarity ( Fig.17 );as well as

[0030] Fig.18 and Fig.19 are schematic diagrams of components in one embodiment of the diagnostic medical device ( Fig.18 ) and a swim-lane diagram of the process flow through these elements when the device is functioning ( Fig.19 ). API = Application Programming Interface; LUIS = Language Understanding Intelligence Service; CA = Computational Algorithms; NLP = Natural Language Processing.

[0031] Detailed description

[0032] The present invention is understood by reference to the following detailed description, which should be interpreted in conjunction with the accompanying drawings. It should be understood that the detailed description of the following various embodiments is only an example and is not intended to limit the scope of the present invention in any way. In the above overview, in the following detailed description, in the following claims, and in the accompanying drawings, specific features (including method steps) of the present invention are shown as reference. It should be understood that the disclosure of the present invention in this specification includes all possible combinations of these specific features, not just the contents of these explicit descriptions. For example, when a specific feature is disclosed in the context of a specific aspect or embodiment of the present invention or a specific claim, the feature may also be used or combined in the context of other specific aspects and embodiments of the present invention and in the entire invention, where possible. The term "comprising" and its grammatically equivalent synonyms used herein refer to the presence of other components, ingredients, steps, etc. in optional cases. For example, the term "comprising" (or "including") components A, B and C can be composed of (i.e., only) components A, B and C, or can contain not only components A, B and C, but also one or more other components. When reference is made herein to a method comprising two or more defined steps, the defined steps may be performed in any order or simultaneously (unless the context excludes such a possibility), and the method may include one or more additional steps performed before any defined step, between two defined steps, or after all defined steps (unless the context excludes such a possibility).

[0033] The term "at least" followed by a number is used herein to indicate the beginning of a range that begins with that number (which may be a range with an upper limit or without an upper limit, depending on the variables defined). For example, "at least 1" means 1 or greater than 1. The term "at most" followed by a number is used herein to indicate the end of a range that ends with that number (which may be a range with a lower limit of 1 or 0, or a range with no lower limit, depending on the variables defined). For example, "at most 4" means 4 or less than 4, and "at most 40%" means 40% or less than 40%. In this specification, when a range is specified as "(first number) to (second number)" or "(first number)-(second number)", this means a range with a lower limit of the first number and an upper limit of the second number. For example, 25 to 100 millimeters means a range with a lower limit of 25 millimeters and an upper limit of 100 millimeters. The embodiments set forth below represent the necessary information to enable those skilled in the art to practice the invention, and show the best way to practice the invention. In addition, the present invention does not require all advantageous features and all advantages to be incorporated into every embodiment of the present invention.

[0034] Now introduce Figure 1-19, a brief description of the various components of the present invention will be discussed. The inventors disclose that a computational algorithm (CA) 2, such as an artificial neural network (ANN), can be used as part of a medical device 3 to predict a diagnosis 20 of a patient having a disease 4 based on one or more individual symptoms 6, medical history elements 8, examination results 10, and / or diagnostic test results 12 (collectively referred to as "data inputs 14"), preferably each data input 14 receives its own prediction weight 16 for analysis.

[0035] The inventors have observed that certain diseases 4 can facilitate identification based on a collection of highly concurrent symptoms 6, e.g., a syndrome 18. While the technical definition of "syndrome 18" includes only symptoms 6, the term is used herein to include a broader, more colloquial sense as a collection of multiple individual data inputs 14 including, for example, symptoms 6, medical history elements 8, examination results 10, and / or diagnostic test results 12. This broader definition is more closely related to the practice of medicine.

[0036] In particular, the diagnosis of certain disorders4 in the neurological domain may be predicted20 by identifying a syndrome18, which itself may be used as a single data input14 to a computational algorithm2 or to supplement the diagnostic assessment of a computational algorithm2. In clinical neuroscience, focal brain damage typically involves multiple discrete neural structures involved in functional networks where these structures and networks are in close proximity (if not overlapping) in physical space and may otherwise have little or no functional relationship. For example, the brainstem serves to connect the larger forebrain to the rest of the body directly via cranial nerves and indirectly via projections to the spinal cord, and where most non-cognitive neural functions are located. Thus, even small focal injuries to the brainstem may cause significant symptoms, but in a unique manner based on the location of the brainstem injured. A specific example is a unilateral injury to the lateral portion of the lower (medulla) brainstem, which causes a collection of symptoms6, examination findings10, and diagnostic test results12 known as the lateral medullary syndrome18, such as Figure 1As shown. The lateral medullary syndrome 18 is closely associated with vertebral artery or posterior inferior cerebellar artery occlusion and, therefore, it almost always represents an ischemic stroke 4, the condition that the patient 22 suffers from and will be correctly identified as having the syndrome 18. Therefore, it will be considered by clinicians to be a typical stroke syndrome 18. Other typical neurological syndromes 18 are more strongly associated with other conditions 4, such as epileptic seizures, demyelinating diseases such as multiple sclerosis or head trauma. Then, a computational algorithm 2, such as an ANN, for the purpose of diagnosing 20 an ischemic stroke 4 can identify the diagnosis 20 of an ischemic stroke 4, thereby excluding the diagnosis 20 of an ischemic stroke 4 for any individual 22 diagnosed by the medical device 3 as having typical symptoms 18 of a non-ischemic stroke condition (referred to as a "stroke-like" condition 24).

[0037] According to one embodiment of the medical device 3 currently claimed, is a computational algorithm 2, such as an ANN, wherein a syndrome 18 is rated as a separate data input 14 and is accompanied by one or more symptoms 6, medical history elements 8, examination results, and / or diagnostic test results 12 experienced by an individual 22. The weight 16 of the typical syndrome 18 input is preferably high (e.g., 0.9 or greater) and has a positive predictive effect on the target disease. Then, the definition of satisfying or not satisfying the classic syndrome 18 will be considered as a "key definition 26" for the ANN diagnostic assessment. Taking acute ischemic stroke (AIS) as an example, the classic stroke syndrome 18 will be used as a key definition 26 for evaluating the patient 22, such as Figure 5 shown.

[0038] Similarly, other key definitions 26 may include definitions / diagnostic criteria / characteristics of medical conditions or syndromes that are similar to the target condition. Using an acute ischemic stroke diagnostic algorithm as an example, stroke-like symptoms 24 would include conditions that are commonly misdiagnosed as acute ischemic stroke in the diagnostic setting, such as epileptic seizures, hemorrhagic stroke, migraine, or traumatic brain injury. See Figure 2 The classic stroke-like attack syndrome 24 as a key definition 26 will preferably have a high weight 16 (eg, 0.9 or greater), but will have a negative predictive effect on the target disease 4 (here, acute ischemic stroke).

[0039] In another embodiment, the presently claimed invention includes a computational algorithm 2, such as an ANN, wherein for diagnostic purposes, other key definitions 26 are considered as data inputs 14. These include predetermined definitions of meeting or not meeting the following conditions: (i) acute / sudden onset; (ii) persistence or remission of symptoms 6, examination results, and / or diagnostic test results.

[0040] In another embodiment, the calculation algorithm 2 may rank the weights 16 of the data inputs 14 relative to one another as follows: the key definition 26 is greater than a single symptom 6, which is greater than a single test result 10, which is greater than a medical history element 8. In some embodiments, a diagnostic test result 12 may be used as the key definition 26, e.g., the absence of intracranial hemorrhage on a CT scan or other diagnostic test assessment is a default or rule-out definition of "ischemia."

[0041] In other embodiments, meeting or failing to meet the classic syndrome 18 definition is not used by the computational algorithm 2 as a data input 14, but rather confirms the computational algorithm's diagnosis 20 or invalidates the computational algorithm's diagnosis 20, or generates an uncertain diagnosis 20 for a condition requiring further evaluation by the duty physician 28. In other embodiments, the classic syndrome 18 definition must first be met before the data input 14 can be evaluated by the computational algorithm 2 to arrive at a diagnosis 20. In still other embodiments, meeting or failing to meet the classic syndrome 18 definition determines the computational algorithm 2 employed in the patient evaluation 30, wherein multiple classic syndromes 18 are selectively employed by the computational algorithm 2 from multiple computational algorithms 2.

[0042] Now introduce Figure 3-10 , showing another embodiment of the currently disclosed medical device 3, namely, an artificial intelligence medical device 3 that adopts multiple computing algorithms 2.

[0043] The artificial intelligence based diagnostic medical device 3 (including "software as medical device 3") may include a computing algorithm 2, which is structured as, for example, an artificial neural network, a support vector machine, a Bayesian algorithm, etc. A single computing algorithm 2 may be used to predict a diagnosis 20 of a patient having a disease 4 based on data input 14 of, for example, symptoms 6, medical history elements 8, examination results 10, and / or diagnostic test results 12.

[0044] According to another embodiment of the disclosed medical device 3, an artificial intelligence-based medical device 3, which will be further described below, has one or more computing algorithms 2 that coordinately predict a diagnosis 20 of having a disease 4, such as by consensus, majority, or other predefined thresholds. In some embodiments of the medical device 3, if all of the computing algorithms 2 employed do not reach the diagnosis 20, then the diagnosis 20 will not be provided to the patient 22 or healthcare provider 28, or the diagnosis 20 provided will be accompanied by an alarm 32 or warning message, such as that all of the computing algorithms 2 did not reach the diagnosis 20. And, in some embodiments, if all of the computing algorithms 2 employed do reach the diagnosis 20, then the diagnosis 20 will be provided to the patient 22 or healthcare provider 28. In other embodiments, if at least a majority of the computing algorithms 2 do not reach the diagnosis 20, then the diagnosis 20 will not be provided to the patient 22 or healthcare provider 28, or the diagnosis 20 will be provided with an alarm 32 or warning message. And in other embodiments, if at least a majority of the computing algorithms 2 do reach the diagnosis 20, then the diagnosis 20 will be provided to the patient 22 or healthcare provider 28. In yet other embodiments, if one or more computing algorithms 2 acting as "gatekeepers" 34 provide specific outputs 38, this will allow other computing algorithms 2 to determine a diagnosis 20, or will allow other computing algorithms 2 to provide a diagnosis 20, or in the case where the computing algorithm 2 acts as a "poison pill defense 36", provide specific outputs 38 that can prevent other computing algorithms 2 from determining a diagnosis 20 that might otherwise be determined. In such cases where "gatekeepers" 34 and "poison pill defenses" 36 are present, certain computing algorithms 2 or other computer processes may have specific purposes in addition to identifying a target primary disease 4 (e.g., stroke), such as detecting head trauma, identifying epileptic seizure activity on an EEG, measuring increased intracranial pressure, or detecting a motor vehicle accident in an emergency, similar seizure diseases 24 will call into question the diagnosis 20 of stroke, while in other cases a diagnosis 20 of stroke would be reached through a different computing algorithm 2 in the artificial intelligence medical device 3. In other embodiments, the computational algorithm 2 may provide 2, 3, 4, 5 or more diagnoses 20, each diagnosis 20 with its own probability of being true, and preferably, with data inputs 14 in favor of each diagnosis 20 and data inputs 14 questioning each diagnosis 20, and preferably, with data inputs 14 that can address the absence of any or all of the diagnoses 20, and particularly data inputs 14 indicating whether the diagnosis 20 is correct or incorrect, such as gatekeepers 34 and poison defenses 36.

[0045] Each of the multiple computational algorithms 2 of the proposed artificial intelligence-based diagnostic medical device 3 can be trained and / or used for reference using their respective data inputs 14. Alternatively, some or all of the computational algorithms 2 can be computed on the basis of a common set of data inputs 14. In some embodiments of the medical device 3, the multiple computational algorithms 2 can have substantially the same original structure and / or code, but differ due to different trained data sets, and therefore have different weights 16 for the same data input 14, for example, two initially identical computational algorithms 2 (e.g., artificial neural networks) trained with different patient databases 40 will become different. The two algorithms will adjust the corresponding weights 16 assigned to the data inputs 14 based on the predicted weights 16 of the single data inputs 14 in the single patient database 40 for the specific disease 4.

[0046] In some embodiments of the proposed machine learning medical device 3, multiple computational algorithms 2 and / or analysis processes are engaged in a serial manner to improve diagnosis 20, such as Figure 5 The improvement of diagnosis 20 may continue, for example, by determining, by one computational algorithm 2, that stroke is the cause of the neurological emergency of patient 22, then determining, by a second computational algorithm 2, that the stroke is acute / sudden onset within a predetermined time limit, then determining, by a third computational algorithm 2, that the acute onset is caused by cerebral ischemia, and then determining, by a fourth computational algorithm 2, whether patient 22 is suitable for medical treatment or surgical / endovascular intervention based on the indications and contraindications for the treatment. In some embodiments, the coordinated activities of the computational algorithms 2 are substantially like a decision tree or flow chart with multiple independent or semi-independent decision points Figure 1 In the same way, the decision calculation is performed at the decision point. If the multiple computing algorithms 2 of the medical device 3 cannot come up with a high diagnostic value (e.g., high probability / high certainty) to determine the diagnosis 20 for the patient 22, other embodiments of the medical device 3 will forward the evaluation 30 of the patient 22 to the physician 28 or other healthcare provider. In some embodiments of the medical device 3, the diagnosis 20 implemented by the artificial intelligence-based diagnostic medical device 3 is considered to have a sufficiently high probability / certainty only when the diagnostic accuracy / consistency with the diagnosis 20 of the physician 28 is at least 85%, 90%, 92%, 94%, 95%, or 96%.

[0047] In other embodiments of the medical device 3, the involvement of the computational algorithms 2 of the individual 22 is flexible and can be adjusted so that those computational algorithms 2 with high diagnostic confidence will be used in the patient assessment 30 and diagnostic decision process, while in addition, less efficient computational algorithms 2 will only be trained on the patient 22 data for possible future use and will not help in the diagnostic assessment. In such an embodiment, the use of a particular computational algorithm 2 for diagnostic purposes as part of a plurality of computational algorithms 2 can be changed or adjusted over time according to which computational algorithm 2 has the most desired sensitivity, specificity, positive predictive value, negative predictive value and / or consistency / concordance rate with other computational algorithms 2. In one such embodiment of the medical device 3, a computational algorithm 2 with machine learning capabilities trained on the records of the patient 22 collected retrospectively is replaced with a computational algorithm 2 trained on the records of the patient 22 collected prospectively. This embodiment can gradually replace the retrospectively trained computational algorithm 2 with the prospectively trained computational algorithm 2, such as in a manner proportional to the number of records of the patient 22 trained, or suddenly with the prospectively trained computational algorithm to reach a predetermined threshold. In some embodiments, different computing algorithms 2 proactively collect different data inputs 14 to build a proactive database 40 that they and / or other computing algorithms 2 can use in training and / or decision calculations.

[0048] Although the technical definition of “syndrome 18” refers only to symptoms 6 reported by the patient 22, in this article, we use the term in a broad sense as a collection of multiple symptoms 6, historical elements 8, examination findings 10, and / or diagnostic test results 12 (collectively, “syndrome elements 42”). Definitions broader than technical definitions are more appropriate and more closely related to medical practice.

[0049] Certain diseases are particularly well identified by the presence of a syndrome18, or even pathologically by a syndrome18 (“classic syndrome18”). The diagnosis of certain medical diseases4 in the field of neurology is particularly well identified in this way20. In clinical neuroscience, focal brain damage / dysfunction often involves multiple discrete neural structures that participate in important aspects of anatomically distributed functional networks, where these neural structures are physically proximal (if not overlapping) but have little or no functional relationship. Take the brainstem, for example, which is the part of the cerebrum that connects the larger forebrain to the body directly via cranial nerves and indirectly via projections to the spinal cord. Most non-cognitive neural functions can be located in the brainstem. Thus, even a small focal injury to the brainstem can produce a number of neurological abnormalities in ways that are unique to the part of the brainstem that is damaged and the nature of the pathophysiological mechanism that causes the injury. An example is that unilateral damage to the lateral portion of the lower brainstem (medulla oblongata) can cause a group of syndromic elements known as lateral medullary syndrome1842, such as Figure 1 As shown in Figure 3, lateral medullary syndrome 18 is closely associated with occlusion of the vertebral artery or the posterior inferior cerebellar artery. Therefore, it represents an ischemic stroke. Therefore, clinicians may consider it as “classic stroke syndrome 18” or “classic ischemic stroke syndrome 18”, which is likely an ischemic stroke that does not require further diagnostic evaluation 30.

[0050] Given the high predictive value of the classic syndrome 18, its presence or absence may be used as a single data input 14 to one or more computational algorithms 2 in the artificial intelligence diagnostic medical device 3. However, all syndrome elements 42 may not be present in every typical patient 22 that is considered to have a classic syndrome 18. In order to weight 16 the single data input 14 of the classic symptoms 18 for computational analysis, or to assess the likelihood that a patient 22 has the classic symptoms 18, the inventors disclose calculations based on the number and prevalence of syndrome elements 42 present in a given patient 22 relative to the population average. For some embodiments of the medical device 3, the inventors disclose a calculation in which: 1) a syndrome element 42, for example recorded in the medical literature, is assigned a score that is proportional to the known / recorded prevalence in a patient population 22 confirmed to have the classic syndrome 18; 2) a determination is made as to whether the patient 22 being evaluated has the syndrome element 42; 3) then, the weight 16 of the data input or the probability of a diagnosis 20 of the classic syndrome 18 is calculated by dividing the score representing the syndrome element 42 identified in the patient 22 being evaluated by the total score of all elements of the syndrome 18 included in the historical definition of the syndrome 18, resulting in a percentage.

[0051] exist Figures 4A-4DAn example of a score-based evaluation case for a classic syndrome 18 is shown in . Other classic neurological syndromes 18 are more closely related to epileptic seizures, demyelinating diseases (such as multiple sclerosis), brain tumors or brain trauma, as are similar attack diseases 24 or counterexamples to stroke 4. In one embodiment of an artificial intelligence-based medical device 3, the purpose of which is to obtain a diagnosis 20 of a stroke 4, it may be necessary to consider, evaluate or identify, and thus exclude similar attack syndromes 24 (typical syndromes 18 of these non-stroke diseases) or stroke-like attacks 24. Similar attack syndromes 24 may become negative factors, thereby reducing the possibility of obtaining a diagnosis 20 of a specific disease 4 (such as ischemic stroke). The presence of a similar attack syndrome 24 may also be a period that terminates subsequent calculations, thereby completely preventing the diagnosis 20 of a specific disease 4 from being obtained. Alternatively, or in addition, the medical device 3 may arrive at a diagnosis 20 of disease 4, but include an alert 32 if the likelihood of another diagnosis 20 (eg, a similar onset disease 24) is above a certain level, such as 15%, 25%, 50%, 75% or 90%.

[0052] In some embodiments of the medical device 3, the patient 22 is assessed as having classic syndrome 18 only if a predetermined number or proportion of syndrome elements 42 are identified during the initial screening of the patient 22, for example, if one, two, or three of the most common syndrome elements 42 are found in patients 22 diagnosed with syndrome 18, or if one quarter or even one half of the syndrome elements 42 are found in patients 22 diagnosed with syndrome 18. Figure 4A -D, similarly, can be used to trigger a more comprehensive assessment of whether the patient 22 has a typical syndrome 18, where the assessment 30 is only initiated when a certain percentage or majority of the scores representing the syndrome elements 42 are identified in the patient 22.

[0053] Reference Figure 6, showing a hierarchical computing algorithm 2 system of the medical device 3. As shown, in various databases 40 consisting of records of patients 22 suffering from the same disease 4, inconsistent data inputs 14 may exist. In the figure, "Y" indicates that data exists in the displayed box, while "N" indicates that data does not exist in the displayed box. Taking ischemic stroke 4 as an example, many data inputs 14 may exist in all databases 40 (e.g., age, atrial fibrillation) because they can strongly predict the fact that it is a known fact that a given disease 4 exists. Some other data inputs 14 may exist in some but not other databases 40, such as drinking or family history of stroke. This may be a challenge, that is, how to serially train the disclosed medical device 3 on databases 40 that do not all have the same data inputs 14. Imputation of missing data may be problematic. One way to train the computing algorithm 2 of the medical device 3 to improve the accuracy of diagnosis 20 is to repeatedly cycle through the patient database 40, including finding the database 40, accessing the database 40, using the database 40 for training, and then searching other databases 40. A benefit of this embodiment of the training schedule is that it maximizes machine learning capabilities, while a potential disadvantage is that it may lose / dilute / overwhelm existing training in successive training cycles.

[0054] The disclosed medical device 3 can utilize various databases 40. The FABS database 40 (related to the FABS scoring system), the FAST-MAG (field management of stroke treatment-magnesium) database 40, and the GWTG (use by guidelines) database 40 are shown only as examples. Additional and / or other databases 40 may be used based on availability and applicability for a particular disease.

[0055] The inventors disclose multiple embodiments to train the medical device 3 and derive a diagnosis 20 using various databases 40. The first embodiment is serial training using a database 40 that uses all data elements and allows dilution of uncommon data inputs 14. An advantage of this embodiment is that it is very simple to implement. A potential weakness is that it may dilute previous training efforts, may ignore the value of data inputs 14 that are difficult to collect, and may become infeasible due to the need to constantly estimate and delete data inputs 14. The second embodiment is to train only on common data inputs 14 available in all databases 40. The advantage of this embodiment is that it is very simple to implement. A potential weakness is that this method may discard potentially useful data inputs 14 even if the data input 14 has a strong or only weak predictive effect. The third embodiment is serial training for transfer learning with or without training from large to small databases 40. The advantage of this embodiment is that it protects previous training, and it probably starts with the most accurate single training estimate (i.e., the largest database 40) first, which limits the variance. The potential weakness of this embodiment is that it may rely on having the largest database 40 at the beginning and the uncertainty of the partial calculation algorithm 2 frozen during training may mean trial and error. The fourth embodiment is to estimate the value of the data input 14 missing in all databases 40. The advantage of this embodiment is that it is not complicated for some data inputs 14. The potential weakness of this embodiment is that it may produce mixed interference, such as some data inputs 14 cannot be estimated, such as it may dilute the value of previous training, and there may be a large number of missing data inputs 14, resulting in potential greater unreliability. The fifth embodiment is to estimate the missing data input 14 for the smaller database 40 based on the largest database 40, and then merge all databases 40 and train on the integrated database 40. The advantage of this embodiment is that it creates the largest database 40 and it has potentially less unreliability than the fourth embodiment. The potential disadvantage is that it assumes that the missing data input 14 has a lower diagnostic value and can estimate the missing data input 14. The sixth embodiment sets thresholds for data inputs, including data input 14, for example, data input 14 must exist in multiple databases 40 and / or must be an acceptable risk factor for a disease 4 (e.g., stroke). The advantage of this embodiment is that it is based on the knowledge of established risk factors. The potential weakness is that it has the potential to eliminate data elements that are ignored or whose current value is unknown. The seventh embodiment is trained on the largest database 40 and then validated on a smaller database 40. The advantage of this embodiment is that it does not exclude other possible designs. The disadvantage is that it may be necessary to estimate missing data inputs 14 in the validation database 40, thereby limiting the value of the validation process.An eighth embodiment is to combine rare data inputs 14 into groups (e.g., heart rate and temperature are grouped as "non-blood pressure vital signs"). The advantage of this embodiment is that it is simple. The potential weakness of this embodiment is that the unique predictive value of the rare data inputs 14 may be lost.

[0056] The group of computing algorithms 2 may include one or more ANNs, support vector machines (SVMs) (including NuSVM and linear SVM), and naive Bayes (NB) algorithms (including Gaussian NB) and polynomial NB computing algorithms. The group of computing algorithms 2 may include multi-class SVM, directed acyclic graph SVM (DAGSVM), structured SVM, least squares support vector machine (LS-SVM), Bayesian SVM, direct induction support vector machine, support vector clustering (SVC), classification SVM type 1 (also called C-SVM classification), classification SVM type 2 (also called nu-SVM classification), regression SVM type 1 (also called ε-SVM regression), regression SVM type 2 (also called nu-SVM regression). In one embodiment of the device, the NB algorithm and the SVM algorithm must be diagnosed consistently before a diagnosis can be established. In other embodiments of the device, different combinations of 2, 3, 4, 5 or more computing algorithms must be diagnosed consistently before a diagnosis can be established. During a first phase, the device can train a set of main computing algorithms 44 using an aggregate database 40 that includes only data inputs 14 common to all databases 40, see Figure 6 , i.e., inputs 14 with representative data in each database 40. In a second phase or second domain, the medical device 3 can use all data inputs 14 from the various databases 40 to train a separate set of calculation algorithms 2, wherein each database 40 trains its own set of calculation algorithms (secondary calculation algorithm set 46). The primary and secondary calculation algorithms 2 can be the same or different types of algorithms.

[0057] Reference Figure 7, in this case, the primary computing algorithm group 44 first attempts to arrive at a diagnosis 20 based on the common and highly predictive data inputs 14 of known disease 4 risk factors. If uncertain, or if the diagnostic score 48 does not reach a first value 50, then the patient 22 case may be referred to the secondary computing algorithm group 46. Additional data inputs 14 may then be obtained through a front-end patient interface 50 of the device, a third party 74, and / or by retrieving the patient 22 record as needed by the secondary computing algorithm group 46. The diagnosis 20 may then be re-evaluated, for example, using the consensus results of the secondary computing algorithm group 2 as a tiebreaker or by superimposed probability assessments based on the calculations of the primary computing algorithm group 44. If the process re-evaluation 30 does not conclusively render a diagnosis 20 of the presence of the disease 3, and / or does not conclusively render a diagnosis 20 of the absence of the disease 3, then the patient 22 may then be referred to a physician 28 (a neurologist in this embodiment) for evaluation 30.

[0058] Now introduce Figure 11-14 , further discussing embodiments of the disclosed medical device 3. In some embodiments, device input 56 obtained by sensors 62 (e.g., microphones and cameras) and direct interfaces 64 can be passed through a feature extraction module (also referred to as a feature extractor) that converts the device input 56 into "features" 58, which are valid digital representations of the device input for training the computing algorithm 2. The interface can be, for example, a keyboard or a touch screen. In addition to the device input 56 features 58, characteristic "labels" 60 can be provided for symptoms. "Labels" can include the degree of slurring of a voice sample, or the degree to which the front half of the face droops relative to the back half when smiling. In addition to data 14 such as user input, other types of labels of diagnostic symptoms 6 from visual or auditory images / videos or records of patients 22 can also be used, including binary or scaled or ranged values ​​directly input into the medical device 3. This includes "yes" or "no" binary answers to questions asked by the medical device 3, such as from "0-10" or simulated answers (such as real or virtual sliders), to range answers to questions asked by the medical device 3. The neural network 66 may be trained by receiving, processing, and learning from a plurality of device inputs 56 and their associated labels 60 or sets of labels 60 to allow the device to estimate the diagnosis 20 of the patient 22 .

[0059] In some embodiments, a neural network architecture may be constructed having a sufficient number of layers 52 and nodes 60 within each layer 52 so that when trained with input data 14 obtained from sensors 62 (e.g., cameras and microphones) and a user interface (UI) 64, it may model a diagnosis 20 with sufficient accuracy. Fig.11An example of a neural network 66 architecture is shown, where features extracted therefrom are provided as neural inputs 68 (f1, f2, ..., fL) to one or more lower layers 52, one or more long short-term memory (LSTM) layers 52, and one or more deep neural network (DNN) layers 52 to estimate a diagnostic score 48. Various types of neural network layers 52 may be implemented within the scope of the present disclosure. For example, one or more convolutional neural network (CNN) layers 52 or LSTM layers 52 may be implemented instead of or in addition to the DNN layer 52. In some cases, various types of filters may be implemented in addition to or as part of one or more neural network layers, such as infinite impulse response (IIR) filters, linear prediction filters, Kalman filters, and the like.

[0060] Fig.12 A specific example of an embodiment of a neural network 66 is shown having four neural network layers 52, namely layers 1, 2, 3, and 4, for processing features 58 extracted from device input 56. Fig.12 , two graphs are shown, namely, Graph n and Graph n+L. It should be understood that within a given length of time, the device inputs 56 can be represented as multiple graphs, and the size of the graph can represent the diagnostic score 48. For Graph n, the first layer 52, i.e., Layer 1, includes device inputs for multiple data inputs 14 for each diagnosis, as shown in the first diagnostic score 48. Similarly, for Graph n+L, Layer 1 includes device inputs 56 from multiple data inputs 14 for each diagnosis, as shown in the second diagnostic score 48. Other information about the patient 22 may be included in Layer 1. In order to change the data inputs 14, Layer 1 of Graph n and Graph n+L may include device inputs 56 representing symptoms 6 that change over time. In one embodiment, when there are a sufficient number of nodes 54 or units in Layers 2 and 3, the neural network 66 will be able to acquire knowledge or diagnostic accuracy and predict a diagnosis 20.

[0061] At least one layer 52 of the neural network 66 may be required to process complex numbers. In one example, complex numbers may be processed in layer 2 of the neural network 66. The complex numbers may be in the form of real and imaginary parts, or alternatively in the form of magnitude and phase. For example, in layer 2 of the neural network, each unit or node 54 may receive a complex input and produce a complex output. In this example, a neural unit with complex inputs and complex outputs may be a relatively straightforward setup for layer 2. In one example, the net result U within the complex unit is given by: U = Σ i W i X i +V, where W iis a complex-valued weight 16 connecting the complex-valued input, and V is a complex-valued threshold. To obtain a complex-valued output signal, the net result U is converted into real and imaginary parts and these real and imaginary parts are passed through the activation function f R (x) to obtain the output f out , the formula is where f r (x) = 11 + e - x; for example, x∈R. Various other complex-valued calculations may also be implemented within the scope of the present disclosure.

[0062] In another embodiment, layers 1 and 2 of neural network 66 may involve computations on complex numbers, while layers 52 above, such as layers 3 and 4, may involve computations on real numbers. For example, each unit or node 54 in layer 2 of neural network 66 may receive a complex input and produce a real output. Various schemes may be implemented to generate a real output based on a complex input. For example, one approach is to implement a complex input-complex output formula and make the complex output real by simply taking the magnitude of the complex output: Alternatively, another approach is to apply an activation function to the absolute value of the complex sum, i.e. f out =fR(|U|). In another alternative approach, each complex input feature is decomposed into magnitude and phase or real and imaginary parts. These components can be viewed as real input functions. In other words, each complex number can be viewed as two separate real numbers representing the real and imaginary parts of the complex number, or, alternatively, two separate real numbers representing the magnitude and phase of the complex number.

[0063] Embodiments of the presently disclosed subject matter can be implemented and used with various components and network architectures. Fig.14 The illustrated medical device 3 neural network 66 may include one or more computing devices 70 for implementing the above-described subject matter embodiments. Fig.13 An example of a computing device 70 suitable for implementing an embodiment of the disclosed subject matter is shown. The computing device 70 may be, for example, a desktop or laptop computer, or a mobile computing device such as a smart phone, tablet computer, video conferencing / telemedicine system, etc. The computing device 70 may include a bus that interconnects the main components of a computer, such as a central processing unit, memory such as main memory (RAM), read-only memory (ROM), flash RAM, etc., a user display (e.g., a display screen), a user input interface, which may include one or more controllers and associated user input devices such as a keyboard, a mouse, a touch screen (which may be considered part of the interface 64), etc., storage devices such as a hard drive, flash memory, etc., removable media components operable to control and receive optical disks, flash drives, etc., and a network interface operable to communicate with one or more remote devices via a suitable network connection.

[0064] As previously mentioned, the bus allows data communication between the central processor and one or more memory components, which may include RAM, ROM, and other memory. Typically, RAM is the main memory where the operating system and application programs are loaded. The ROM or flash memory component may contain, among other code, the basic input and output system (BIOS), which controls basic hardware operations, such as interaction with peripheral components. Application programs resident in the computer are typically stored on and accessed through a computer-readable medium, such as a hard drive (such as fixed storage), an optical drive, a floppy disk, or other storage media.

[0065] The fixed storage may be integrated with the computer or may be separate and accessible through other interfaces. The network interface may provide a direct connection to a remote server through a wired or wireless connection. The network interface may provide such a connection using any suitable technology and protocol readily apparent to those skilled in the art, including digital cellular telephones, Wi-Fi, Near field, etc. For example, a network interface may allow a computer to communicate with other computers via one or more local area networks, wide area networks, or other communication networks, as will be described in further detail below.

[0066] Many other devices or components (not shown) may be connected in a similar manner (e.g., document scanners, digital cameras, etc.). Instead, to practice the present disclosure, Fig.13 Not all components shown in the figure need to be present. Components may be interconnected in a manner different from that shown. Fig.13 The operation of the computer shown is well known in the art and will not be discussed in detail in this application. The code implementing the present disclosure may be stored in a computer-readable storage medium, such as one or more memories, fixed storage, removable media, or in a remote storage location.

[0067] More generally, various embodiments of the presently disclosed subject matter may include or be embodied in the form of computer-implemented processes and devices for practicing those processes. Embodiments may also be embodied in the form of a computer program product having computer program code containing instructions, which are embodied in a non-transitory or tangible medium (e.g., a floppy disk, a CD-ROM, a hard drive, a USB (universal serial bus) drive, or any other machine-readable storage medium), such that when the computer program code is loaded into a computer and executed by the computer, the computer becomes a device for practicing the disclosed subject matter embodiments. Embodiments may also be embodied in the form of computer program code, for example, whether stored in a storage medium, loaded into a computer or executed by a computer, or transmitted through some transmission medium (e.g., by wire or cable, by optical fiber, or by electromagnetic radiation), such that when the computer program code is loaded into a computer and executed by the computer, the computer becomes a device for practicing the disclosed subject matter embodiments. When implemented on a general-purpose microprocessor, the computer program code segments configure the microprocessor to create a specific logic circuit.

[0068] In some configurations, a set of computer-readable instructions stored on a computer-readable storage medium may be implemented by a general-purpose processor that may convert a general-purpose processor or a device containing a general-purpose processor into a special-purpose device configured to implement or execute instructions. The embodiments may be implemented using hardware that may include a processor such as a general-purpose microprocessor or an application-specific integrated circuit (ASIC) that embodies all or part of the techniques according to the disclosed subject embodiments in hardware or firmware. The processor may be coupled to a memory such as a RAM, ROM, flash memory, a hard disk, or any other device capable of storing electronic information. The memory may store instructions suitable for execution by a processor to perform techniques according to the disclosed subject embodiments.

[0069] In some embodiments, Fig.14The microphone and camera shown may be implemented as part of a network of sensors 62. For example, these sensors 62 may include microphones for sound detection and cameras for visual detection, and may also include other types of sensors 62. In general, a "sensor 62" refers to any device that can obtain information about its environment. Sensors 62 may be described by the type of information they collect. For example, the types of sensors 62 disclosed herein may include waveforms, chemical emissions, motion, smoke, carbon monoxide, proximity, temperature, time, physical orientation, acceleration, position, entry, presence, pressure, light, sound, and the like. Sensors 62 may also be described in terms of specific physical devices that obtain environmental information. For example, an accelerometer may obtain acceleration information and may therefore be used as a general motion sensor 62 or an acceleration sensor 62. Sensors 62 may also be described in terms of specific hardware components used to implement sensors 62. For example, a temperature sensor 62 may include a thermistor, a thermocouple, a resistance temperature detector, an integrated circuit temperature detector, or a combination thereof. Sensors 62 may also be described in terms of the functions that the sensors 62 perform within an integrated sensor 62 network such as a smart home environment. For example, when the sensor 62 is used to determine security events such as unauthorized entry, it can be used as a security sensor 62. The sensor 62 can operate with different functions at different times, such as a motion sensor 62 used to control lighting in a smart home environment when an authorized user is present, and used to warn of unauthorized or unexpected motion when there is no authorized user or the alarm system is in an "armed" state, etc. In some cases, the sensor 62 can operate as multiple sensor 62 types sequentially or simultaneously, such as when a temperature sensor 62 is used to detect temperature changes and the presence or absence of a person or animal. The sensor 62 can also operate in different modes at the same or different times. For example, the sensor 62 can be configured to operate in one mode during the day and in another mode at night. Another example is that the sensor 62 can operate in different modes based on the state of the home security system or smart home environment, or under other instructions of such a system.

[0070] In general, the "sensor 62" disclosed herein may include multiple sensors 62 or sub-sensors 62, for example, where the location sensor 62 includes both a global positioning sensor 62 (GPS) and a wireless network sensor 62, which provides data that can be associated with a known wireless network to obtain location information. Multiple sensors 62 can be arranged in a single physical space, such as where a single device includes motion, temperature, magnetic or other sensors 62. Such a space may also be referred to as a sensor 62 or a sensor 62 device. For clarity, when the specific functions performed by the sensor 62 or the specific physical hardware used are necessary for understanding the embodiments disclosed herein, such description will be explained when describing the sensor 62.

[0071] In addition to specific physical sensors 62 that obtain information about the environment, the sensor 62 may also include hardware. The sensor 62 may include an environmental sensor 62, such as a temperature sensor 62, a smoke sensor 62, a carbon monoxide sensor 62, a motion sensor 62, an accelerometer, a distance sensor 62, a passive infrared (PIR) sensor 62, a magnetic field sensor 62, a radio frequency (RF) sensor 62, a light sensor 62, a humidity sensor 62, a pressure sensor 62, a microphone, a weight scale, or any other suitable environmental sensor 62, which obtains the corresponding type of information about the environment in which the sensor 62 is located. The processor can receive and analyze the data obtained by the sensor 62, control the operation of other components of the sensor 62, and process the communication between the sensor 62 and other devices. The processor can execute instructions stored on a computer-readable memory. The memory or another memory in the sensor 62 can also store the environmental data obtained by the sensor 62. A communication interface, such as a Wi-Fi or other wireless interface, an Ethernet or other local network interface, etc., can allow the sensor 62 to communicate with other devices. The user interface can provide information to the user or receive input from the user or the sensor 62. The user interface 64 may include, for example, a speaker to sound an alarm, i.e., an output 38, when the sensor 62 detects an event. Alternatively, or in addition, the user interface 64 may also include a light that will be activated when the sensor 62 detects an event. The user interface may be relatively small, such as a display with limited output 38, or may be a full-featured interface, such as a touch screen. It is readily understood by those skilled in the art that the components within the sensor 62 may send and receive information back and forth between each other via an internal bus or other mechanism. In addition, the sensor 62 may include one or more microphones to detect sounds in the environment. One or more components may be implemented in a single physical arrangement, such as the case where multiple components are implemented on a single integrated circuit. The sensor 62 as disclosed herein may include other components or may not include all of the illustrative components shown.

[0072] As disclosed herein, the sensors 62 can operate within a communication network such as a conventional wireless network or within a sensor 62-specific network, through which the sensors 62 can communicate with each other or with other dedicated devices. In some configurations, one or more sensors 62 can provide information to one or more other sensors 62, to a central controller, or to any other device capable of communicating with one or more sensors 62 on a network. A central controller may have general or special purposes. For example, one type of central controller is a home automation network that can collect and analyze data from one or more sensors 62 in a home. Another example of a central controller is a dedicated controller dedicated to a subset of functions, such as a security controller, which mainly or exclusively collects and analyzes data from sensors 62 because it is related to various security considerations at a location. The central controller can be arranged locally relative to the sensors 62 that communicate with it, and the central controller obtains the data of the sensors 62 from the sensors 62, such as arranging the central controller in a house containing a home automation or sensor 62 network. Alternatively, or in addition, the central controller disclosed herein may be remote from the sensors 62, such as by arranging the central controller as a cloud-based system that communicates with a plurality of sensors 62, which may be located at a plurality of locations and which may be both local and remote to each other.

[0073] In addition, the smart home environment can infer which individuals 22 reside in the home and are therefore users, and which electronic devices are associated with those individuals 22. In this way, the smart home environment can "learn" who the users are (e.g., an authorized user) and allow electronic devices associated with those individuals 22 to control the networked smart devices of the smart home environment, including, in some embodiments, sensors 62 used by or within the smart home environment. Various types of notifications and other information can be provided to users via messages sent to one or more user electronic devices. For example, messages can be sent via email, short message service (SMS), multimedia message service (MMS), unstructured supplementary service data (USSD), and any other type of messaging service or communication protocol.

[0074] The smart home environment may include communication with devices outside the smart home environment but within the immediate geographic range of the home. For example, the smart home environment may communicate information about the detected motion or presence of people, animals, and any other objects to a central server or cloud computing system via a communication network or directly, and receive return commands for controlling ambient lighting accordingly.

[0075] In some embodiments of the medical device 3, the medical device 3 periodically evaluates the patient 22 for symptoms of disease or abnormality, and if symptoms are detected, a more comprehensive patient 22 device evaluation 30 may be automatically triggered to arrive at a diagnosis 20. In one embodiment of the medical device 3, the periodic evaluation 30 involves an evaluation of gait, speech, and limb movement to look for symptoms of neurological abnormalities, such as limp, slurred speech, weakness in one arm, or drooping of one part of the face, respectively. In this embodiment of the medical device 3, the detection of such neurological abnormalities will trigger a comprehensive evaluation of symptoms 6 and physical examination abnormalities consistent with focal brain injury such as stroke. In another embodiment of the device, a patient 22 known to have congestive heart failure will be evaluated for changes in respiratory rate or pauses in speech, which can indicate worsening congestive heart failure, and the patient 22's condition will be specifically evaluated, which may include weighing the patient 22. In this embodiment, the device can guide the patient 22's medication adjustment, such as the use of diuretics, based on the diagnosis 20 of worsening congestive heart failure. In another embodiment of the device, the device will recognize shortness of breath or difficulty breathing as evidence of an impending or ongoing asthma attack, at which point the patient 22 will be notified to use any available respiratory treatment measures, including an inhaler, and / or if the patient 22's condition is severe enough, the device will notify nursing and / or emergency medical services 28 to assist the patient 22. In another embodiment of the device, the device will recognize the following behaviors of the monitored target person 22, such as grasping the chest, facial expressions indicating pain, shortness of breath, flushing and / or sweating, which will indicate that the patient has a myocardial infarction, at which point the device will confirm other symptoms 6 and examination results 10 consistent with myocardial infarction, thereby instructing the patient 22 to take emergency treatment for myocardial infarction, and notify an ambulance 72 to receive the patient 22 based on the possible diagnosis 20 of myocardial infarction. In another embodiment of the device, a conventional thermal scan or spot temperature measurement of the patient 22 can be used to identify whether there is an abnormal body temperature, which will trigger the medical device 3 to evaluate 30 the patient 22 for symptoms consistent with infection 6; based on the diagnosis 20 of the medical device 3, the patient 22 can take the presumptive antibiotic treatment by himself or by a third party 74 before the patient 22 is evaluated 30 in the physician's 28 office or hospital. In other embodiments, when the medical device 3 determines that the patient 22 needs medication, the medical device 3 can also directly dispense the medication. In another embodiment, a same medical device 3 can evaluate whether the patient 22 has any or all of the above diseases 4.

[0076] In some embodiments of the medical device 3, any electronic device equipped with a sufficient combination of input devices 76 (e.g., microphones, thermometers, and one or more cameras), output devices 38 (e.g., speakers and lights), processors, and / or memory, for example, can be used alone or in conjunction to monitor the patient 22. In some embodiments of the medical device 3, various electronic devices are distributed throughout the home of the patient 22 to monitor the patient 22.

[0077] In some embodiments of the medical device 3, the medical device 3 passively assesses whether the patient 22 has evidence of a certain disease 4. When used in an ambulance 72, the medical device 3 can listen to the report provided by the ambulance / emergency medical service (EMS) dispatcher and interpret the report to indicate whether the next patient 22 to be seen by the ambulance personnel has a certain target disease 4, such as a stroke. In that case, for example, the medical device 3 can have the ability to activate itself and perform a comprehensive assessment 30 of the patient 22. In other cases, the medical device 3 can request an opportunity to evaluate the patient 22 (e.g., in person or over the phone), by allowing a conversation between the patient 22 and the emergency medical technician or paramedic, and / or by having a health care provider of such ambulance 72 perform a physical examination assessment 30 of the patient 22.

[0078] The first role of one embodiment of the present medical device 3 is to diagnose ischemic stroke in a pre-hospital setting. Based on this diagnosis 20, treatment options can be immediately obtained. A preferred embodiment of the medical device 3 is to determine that the ischemic stroke symptoms 6 and test results 10 have resolved, indicating that the ischemic stroke has subsided and the patient 22 has had a transient ischemic attack. In this case, the medical device 3 will instruct the patient 22 to take aspirin or other antiplatelet drugs before further diagnostic evaluation 30 or arrival at the hospital. The administration can be completed under the guidance of the medical device 3 before any healthcare provider or any professional medical service including nurses, paramedics or ambulances arrives. In another embodiment, the treatment of ischemic stroke, such as a facial nerve stimulator, is safe enough in the event of a hemorrhagic stroke, so it can be used on an undifferentiated stroke patient 22 without a prior neuroimaging assessment. The present medical device 3 can diagnose the patient 22 as having a stroke and then instruct the use of a transcranial magnetic stimulation (TMS) facial nerve stimulator for treatment. In another embodiment, the medical device 3 itself may provide TMS to the patient 22 after the patient 22 has been diagnosed with a stroke. In other embodiments, after the patient has been given an initial therapeutic TMS stimulation, the medical device 3 will assess whether the patient's symptoms 6 and test abnormalities have improved, and determine whether there has been a recurrence or new onset of symptoms 6 and test abnormalities that would be appropriately treated / benefit from repeated therapeutic TMS stimulation.

[0079] Establishing a diagnosis of stroke 20 in the ambulance 72 or elsewhere prior to arrival at the hospital allows the patient 22 to undergo neuroimaging immediately upon arrival at the hospital to identify intracranial hemorrhage, which would establish a diagnosis of hemorrhagic stroke 20, thereby avoiding treatment with established therapies for ischemic stroke, such as intravenous tissue plasminogen activator (rtPA) or endovascular catheterization. Such a system would bypass the emergency department evaluation 30, thereby facilitating treatment of the ischemic stroke patient 22.

[0080] Now introduce Fig.15 , shows another embodiment in which a diagnosis of a classic syndrome 18 is determined based on a cumulative probability of symptoms 6 and signs for a patient 22. In this embodiment, the proportion of patients 22 having a classic syndrome 18, i.e., the presence of individual symptoms 6, examination abnormalities 10, or diagnostic test results 12 ("components of syndrome 18"), are compiled into a database 40 or library. A bias factor may also be assigned equally or unequally to each syndrome element 42, where the unequal bias factor may be determined by a patient 22 informed survey, impact on quality of life, or decision of one or more healthcare providers. Assuming that all syndrome elements 42 of a classic syndrome 18 are required to have a bias factor of 1.0, the weight 16 of the data input for the classic syndrome 18 may be adjusted based on the proportion of the syndrome elements 42 for the patient 22. As shown Fig.15 In the example shown, the target patient 22 being evaluated has ipsilateral limb and gait ataxia, ipsilateral facial numbness, and ipsilateral Horner's syndrome 18, but no contralateral hemibody numbness, dysphagia and dysarthria, and no nausea, vomiting, vertigo, and nystagmus. The relative proportions of the various components of the syndrome 18 are 90%, 50%, 50%, 90%, 20%, and 10%. The device will calculate the power of symptoms 6 and test results 10 by subtracting the product of the probabilities from 1, as well as the percentage of patients 22 who have all the components of the syndrome 18 for the patient 22. For example, if the patient 22 only has ipsilateral limb and gait ataxia and contralateral hemibody numbness, and the bias factor 42 of each syndrome element is 1, then the diagnostic score 48 is 1–(0.9*-0.9)=1-(0.81)=0.19. If Fig.15If all symptoms 6 and test results 10 are present, the diagnostic score 48 according to this embodiment will be 0.996. When the higher diagnostic value 78 of the diagnostic score 48 is reached, a diagnosis 20 of having a specific disease 4 can be obtained, for example, the diagnostic value can be a value greater than 0.80, greater than 0.85, greater than 0.90, and greater than 0.95. For example, if these higher diagnostic values ​​78 are between 80% and less than 100% (for example, for the embodiment of the higher diagnostic value 0.80, between 0.64 and 0.80, and for the embodiment of the higher diagnostic value 0.90, between 0.72 and 0.90), the diagnostic score 48 can be marked as a diagnosis 20 of a specific disease that cannot be made due to lack of specificity or the diagnosis 20 of a specific disease is not completely accurate, and / or triggers a referral process to refer the patient to a medical professional for diagnosis 20. Such a medium diagnostic score 48 will be referred to as having a medium diagnostic value 80, and it is uncertain whether the specific disease 4 is present. Other lower limits of the intermediate diagnostic value may be 75%, 85%, 90% and 95% of the value of the higher diagnostic value 78. A lower diagnostic score 48, such as a value lower than four fifths or 80% of the high diagnostic value 78 (e.g., lower than 0.64 for an embodiment of a higher diagnostic value of 0.80, lower than 0.72 and 0.90 for an embodiment of a higher diagnostic value of 0.90), may be diagnosed by the medical device 3 as having a lower diagnostic value 82 and being negative for a specific disease 4. Other upper limits of the lower diagnostic value 82 may be 75%, 85%, 90% and 95% of the value of the higher diagnostic value 78.

[0081] Some embodiments of the present medical device 3 aggregate or compile the syndrome elements 42 of each classic syndrome 18 into a database 40 or one or more libraries. For example, multiple libraries can be used to distinguish the classic stroke syndrome 18 from the classic stroke-like attack syndrome 24. In such embodiments of the present medical device 3, the front-end patient interface 64 of the device can collect the initial symptoms 6 reported by the patient 22, and then the medical device 3 can use these initial symptoms 6 to identify the classic syndrome 18 from the library containing these initial symptoms 6. The medical device 3 can then evaluate the distribution of other syndrome elements 42 from the selected classic syndrome 18 and sort the syndrome elements 42 according to the number of times they are found in the selected typical syndrome 18. The next step is that the medical device 3 can then ask the patient 22 whether the syndrome element 42 is most frequently encountered. Based on the patient 22's answer, the subset of classic syndromes 18 selected for including the initial symptoms 6 will be reduced, and only the classic syndromes 18 that also contain the most common syndrome elements 42 will be retained. This process is repeated until a single classic syndrome 18 remains, or until there may still be a small group of classic syndromes 18, all of which are of a single type, such as a classic stroke syndrome 18, at which point the diagnosis 20 may be output by a medical device or communicated to the patient 22 and / or health care provider via a network. In other embodiments, the patient 22 may be asked whether they have uncommon symptom elements 42 that are present in only a few or one of the initially selected classic syndromes 18, thereby reducing the number of possible classic syndromes 18. In other embodiments, the test results 10 or other data inputs 14 may be used to select among various classic syndromes as well as to exclude various classic syndromes.

[0082] Now introduce Fig.16 and 17 In the figure, other embodiments of a computational algorithm 2 for selective or targeted use of databases 40 and for training with certain databases 40 are shown. In the figure, lighter grey cells indicate that a value / data input 14 is present in that database 40, while darker grey cells indicate that no value is present for different elements of different databases 40. For example, in the embodiment shown, the FABS database 40 is used and there are values ​​representing glucose levels, but no values ​​representing medication use. The database 40 to be used that is particularly relevant to the patient 22 in the evaluation 30 made by the device can be selected based on demographic characteristics (age, gender, race, medical history, geographic environment, etc.). Two examples are given in the figure. In the Fig.16 In, non-geographic / personal demographic similarities are used to select the database 40 and its associated computational algorithm that is most relevant to the patient. Fig.17In the example of FIG. 3 , geographic similarity (e.g., geographic proximity) is used to select a database 40 and its associated computational algorithm that is most relevant to the patient. The demographic similarity or relevance of the database 40 to the target patient 22 of the evaluation 30 may cause the medical device 3 to employ a computational process that is adjusted or adapted to a certain database 40, even if the data value of the particular database 40 is less than the data value of another database 40.

[0083] Fig.18 and 19 Other embodiments of using the medical device 3 to diagnose a disease 4 are shown. In these embodiments, Services (e.g. ) are used as examples of functions or effects, but these services are considered to be examples of other similar services that can be used, for example, there are other companies, such as and other companies’ services.

[0084] In other embodiments, the medical device 3 can be integrated into one or more household devices 84 or household fixtures or appliances, such as mirrors, clocks, refrigerators, sinks, toilets, chairs, beds, televisions, microwave ovens, or electric lamps, such as lamps mounted on ceilings. In a preferred case, the household device 84 is a household device that the patient 22 will interact with or approach regularly or frequently. The household device 84 is preferably connected or connectable to a network such as WIFI, Bluetooth, cellular and / or the Internet of Things, and it includes a sensor 62 so that the patient 22's condition can be recorded passively or interactively, such as one or more microphones, cameras, thermal imagers or thermometers, electrocardiogram sensors, photoplethysmograph sensors (for measuring heart rate) and / or other input devices 76 such as electromagnetic pulse monitors and possible interactions with the patient 22. After the household device is integrated with the medical device 3, it can have an output element 38, such as a screen, a light and a speaker, so as to prompt the patient 22 to respond or otherwise interact with the patient 22, for example, with light, sound or voice questions. In an embodiment where the medical device 3 monitors the patient's reported stroke perceived symptoms 6 or physical signs 10, for example, the camera can capture video to determine whether there is a drooping phenomenon in a part of the face or body compared to other parts. The speaker can capture and determine whether the patient 22 is slurring his speech or whether the slurring becomes more severe, or whether the grammar, syntax or content of the language is disrupted. For example, this can also be a reply to the medical device, which sends a "good morning" greeting to the patient 22 in the form of spoken words through the speaker or in the form of written text through the interface 64. In addition, the medical device 3 can monitor the patient's 22 language regularly or continuously, and if it is detected that the patient is speaking slurred, the device can automatically trigger an assessment 30 and / or send an alarm 32 to a caregiver, medical staff or other emergency personnel and / or a central server or other display screen. The camera can also detect whether the patient 22 has a limp in his gait. If the patient 22 has a high risk factor for stroke, the medical device 3 can regularly screen for initial warning symptoms 6 or signs, and automatically trigger an assessment 30 and / or send an alarm 32 when the initial warning symptoms 6 or signs 10 are detected. If initial warning symptoms 6 are detected, or if a diagnosis 20 of a disease 4 is made, the medical device 3 may also instruct the patient 22 to take immediate treatment, or both instruct and provide immediate treatment to the patient 22. In the example of cardiac ischemia or transient ischemic attack, the device may instruct the patient 22 to take aspirin while waiting for a medical professional to arrive. In other embodiments, the medical device 3 may dispense aspirin or other appropriate emergency medication for a specific diagnosed disease 4. The medication may be dispensed from a reservoir containing medications for one or more different diseases.The medications may be in containers or bags that are color coded, numbered, named, or otherwise labeled so that if the patient 22 requires multiple types of medications, then it will be clearly indicated which medications to take. For example, the medical device 3 may state "You have been diagnosed with a possible transient ischemic attack. Please take an aspirin from the blue bag labeled "A" in the medicine container." The medical device 3 may be self-activating. It is expected that the medical device 3 will be used, for example, in nursing and long-term care facilities, as well as in residences for the elderly.

[0085] The medical device 3 preferably collects information directly or indirectly interacting with the patient 22, but also information from other sources such as third parties 74, such as witnesses of the accident, nursing home staff, family members, medical staff and the hospital / medical records of the patient 22. If there is an inconsistency in the input data, the medical device 3 can mark the inconsistent information and issue a warning 32 to the physician 28, and / or the medical device 3 can query the source of the inconsistent information and other information in an attempt to clarify. For example, if the patient 22 enters that symptom 6 began to appear six hours ago, and the nurse enters that symptom 6 began to appear two days ago, this inconsistent information can be marked by the medical device 3, and then the physician 28 or the medical device 3 can determine which is the true information, for example by further inquiring about the source of the inconsistent information. Alternatively or additionally, the accuracy of the answers given by each person can be weighted 16 based on the authenticity or accuracy of other answers given by the individual 22 to the given question or all questions asked to the individual 22, or based on the authenticity or accuracy of the answers given by a group of individuals such as nurses or ambulance personnel, or working in a particular hospital.

[0086] In some embodiments of the medical device 3, the medical device 3 will only talk to and visualize the patient 22 as a means of obtaining diagnostic information about the patient 22. In other more preferred embodiments, the medical device 3 will be able to interact with multiple third parties 74 (individuals other than the patient 22) and other information sources related to the patient 22, such as multiple witnesses when the patient 22 was injured in an accident, family members or healthcare providers of the patient 22, and sources of medical records of the patient 22. In some embodiments, the medical device 3 includes multiple device devices, one of which can be arranged in the ambulance 72, and the other devices can be small handheld devices that can transmit information such as voice, images and input data from users (including third parties 74) to the data collection process of the medical device 3. The ambulance rescuer can then hand the handheld device to the person 74 at the emergency scene, allowing them to interact with the medical device system 3, and the independent device asks the person 74 about the necessary information related to the patient 22's condition, while the rescuer treats the patient 22 or transports the patient 22 to the hospital. The benefit of having a standalone medical device device is that when the device is provided to the third party user 74, the device may already be connected to a network with the rest of the medical device system 3 and already loaded with questions to be asked, which is a critical benefit when time is of the essence. In other embodiments, the standalone medical device device connects directly to a communication-enabled device already available to the third party 74, such as their personal cell phone or other computing device 70, through which information related to the diagnosis 20 of the patient 22 can be queried and communicated. In a preferred embodiment of the present invention, the medical device 3 will collect information about the patient 22 from multiple sources including the patient, third parties and / or medical records in a parallel, simultaneous or overlapping manner.

[0087] The medical staff may leave one or more embodiments of the medical device 3, which are preferably small handheld devices, with a third party 74, such as a witness or family member, especially when the patient 22 is unconscious when the medical staff arrives. The medical device device will ask questions, enter answers, and send data to the medical device system host via a wireless network for analysis, storage, and diagnostic decisions. This can be forwarded to the physician 28 of the destination hospital and / or the nursing staff who transports the patient 22 to the destination hospital. This can save time and improve the quality of information. In one embodiment, they can be mailed out after the third party 74 completes the input data on the small handheld medical device device for return and reuse.

[0088] In another embodiment, a program for querying a third party for information may be downloaded to a third party's 74 smart phone or computing device 70, or the third party 74 may be taken to a website that queries the third party for information about the patient 22. The information will be sent over the network to a medical device central server for compilation and analysis, and then to a medical professional 28 or diagnostic machine. Alternatively, the information may be sent directly or indirectly to the medical professional 28 for compilation and analysis.

[0089] In another embodiment, each or one or more patients 22 diagnosed as having, not having, or including a particular disease 4 can be tracked based on different data inputs 14 to obtain result data. If the patient 22 receives a clear diagnosis 20 prescribed by a doctor 28 or medical professional in a hospital or other medical care center, the diagnosis 20 can be used to create a database 40 of patients 22 after use to improve the accuracy of the computational algorithm 2 used by the medical device. That is, using, for example, back propagation and other analysis and calculations, the accuracy of the diagnosis 20 of past users 22 of the medical device 3 can be used to train and / or update the computational algorithm 2 and determine the selection of weights 16 and data inputs 14 (including changes based on demographic characteristics) and the required deviation of the activation / transfer function of the node 54 when determining the diagnosis 20 of the current user 22 of the medical device 3. In some preferred embodiments, a portable computing and / or communication device is provided to the patient 22 during hospitalization and after discharge, which is capable of providing information about the patient's chronic disease to the database 40. In other embodiments, a portable computing and / or communication device provided to patient 22 may be used to monitor the condition of patient 22 and / or summon emergency medical services.

[0090] In the absence of any elements not specifically disclosed herein, the invention currently claimed for protection disclosed herein by way of example may be appropriately and explicitly implemented. Although various embodiments of the present invention have been described in detail, it is apparent that various modifications and changes of those embodiments will be apparent to those skilled in the art. However, it should be clearly understood that such modifications and changes belong to the scope and spirit of the invention currently claimed for protection, as set forth in the appended claims. In addition, the invention described herein can be applied to other embodiments and can be practiced or executed in various other related ways. In addition, it should be understood that the words and terms used herein are for descriptive purposes and should not be considered to be restrictive. The use of "including", "comprising" or "having" and their synonyms herein is intended to cover the items listed thereafter and their equivalents and other items, and only the terms "consisting of..." and "consisting only of..." have a restrictive meaning when interpreted.

Claims

1. A medical device, the medical device being an artificial intelligence-based diagnostic medical device for diagnosing a neurological disease in a patient, the medical device being configured for use in a pre-hospital setting, the medical device comprising: A memory; a camera configured to acquire image / video data of the patient; one or more sensors including a microphone for acquiring natural language data of the patient; as well as a processor communicatively connected to the memory, the sensor, and the camera, wherein the processor is configured to: executing computer instructions to obtain patient data input, wherein the patient data input includes demographic data, symptoms, medical history elements, and physical examination results of the patient, and the patient data input includes results obtained by the processor using visual recognition and / or computer vision and natural language recognition to recognize the image / video data and the natural language data; executing at least one computing algorithm trained on at least one patient database to estimate a diagnosis of a patient based on the neurological disease using the patient data input; performing at least one analytical process to assess whether a classic syndrome definition is met based on said patient data input; The fulfillment or non-fulfillment of the classic syndrome definition is used to confirm or invalidate the algorithmic diagnosis, or to generate an indeterminate diagnosis; and If the patient is determined to have the neurological disorder, treatment is provided to the patient.

2. The medical device according to claim 1, wherein The executing at least one computational algorithm trained on at least one patient database to predict a diagnosis of the patient suffering from the neurological disease using the patient data input comprises: determining a diagnostic score for the patient for the neurological disease using the at least one computational algorithm; When the diagnostic score is higher than a first value, it is predicted that the patient suffers from the neurological disease; when the diagnostic score is lower than a second value and the second value is lower than the first value, it is predicted that the patient does not suffer from the neurological disease.

3. The medical device according to claim 2, characterized in that The processor is further configured to make an inconclusive diagnosis for the patient when the diagnostic score is between the first value and the second value.

4. The medical device according to any one of claims 1 to 3, characterized in that Straps are also included to allow the patient to wear the medical device.

5. The medical device according to any one of claims 1 to 4, characterized in that The patient data input is data input automatically captured by the medical device and is received from the patient or a third party or both the patient and the third party.

6. The medical device according to any one of claims 1 to 5, characterized in that The medical device initiates interaction with the patient through voice prompts to provide the patient data input.

7. The medical device according to any one of claims 1 to 6, characterized in that The at least one computational algorithm and / or the at least one analytical process is further configured to increase the likelihood of being diagnosed with a specific neurological disease when positive symptoms of the specific neurological disease are present, and to decrease the likelihood of being diagnosed with the specific neurological disease when different symptoms of a similar disease are present.

8. The medical device according to any one of claims 1 to 7, characterized in that The nervous system disease is one or a combination of acute stroke, transient ischemic attack, epileptic seizure, demyelinating disease, multiple sclerosis, brain trauma and brain tumor.

9. The medical device according to any one of claims 1 to 8, characterized in that The at least one computing algorithm and / or the at least one analytical process is further configured to automatically assess whether the patient has developed initial signs of one or more of the neurological diseases and automatically trigger a more comprehensive assessment process when the initial signs are detected.

10. The medical device according to claim 9, characterized in that The at least one computational algorithm and / or the at least one analytical process is further configured to assess the presence of a change in gait, speech, vision, movement of a body part, or some combination thereof as an initial physical examination finding indicative of the neurological disease.

11. The medical device according to any one of claims 1 to 10, characterized in that In response to the medical device diagnosing the patient as having the neurological disease, the processor is further configured to determine a suitable medication to be taken by the patient, and the medical device is further configured to output the determined suitable medication to be taken by the patient.

12. The medical device according to any one of claims 1 to 11, characterized in that The at least one computational algorithm and / or the at least one analytical process is further configured to assess the likelihood that the patient has a second disease similar to the neurological disease, and the diagnosis of the neurological disease is accompanied by an alarm when the likelihood of other diagnoses of the second disease is higher than one of 25%, 50%, 75% and 90%.

13. The medical device according to any one of claims 1 to 12, characterized in that The at least one computing algorithm includes one or more of artificial neural network, support vector machine SVM, Nu-SVM, linear SVM, naive Bayes NB algorithm, Gaussian NB, polynomial NB computing algorithm, multi-class SVM, directed acyclic graph SVM, structured SVM, least squares support vector machine LS-SVM, Bayesian SVM, transductive support vector machine, support vector clustering SVC, C-SVM classification, nu-SVM classification, ε-SVM regression, and nu-SVM regression.

14. The medical device according to any one of claims 1 to 13, characterized in that The processor is further configured to transmit the diagnosis to a medical institution via a wired or wireless network.

15. The medical device according to any one of claims 1 to 14, characterized in that The processor is further configured to: Execute multiple computational algorithms, each using data from multiple databases; determining, for each computational algorithm, a diagnostic score for the patient having the neurological disease, and When the diagnostic scores of the neurological disease given by most of the calculation algorithms are higher than the first value, or the diagnostic scores of the neurological disease given by all of the calculation algorithms are higher than the first value, the patient is diagnosed as having the neurological disease.

16. The medical device according to any one of claims 1 to 15, characterized in that The processor is further configured to: receiving one or more syndromic elements encompassed by historical definitions of classic syndromes associated with said neurological disorder; assigning a score to a syndrome element that is proportional to the known or documented prevalence of that syndrome element in a population of patients diagnosed with the classic syndrome; determining whether the patient exhibits elements of the syndrome; calculating a probability of diagnosis of the patient having the classic syndrome as a ratio representing the total score of the syndrome elements identified in the patient divided by the total score of all syndrome elements covered by the historical definition of the classic syndrome; as well as The diagnostic probabilities of the classic syndromes are used as data input related to the neurological diseases.

17. The medical device according to any one of claims 1 to 16, characterized in that The at least one computational algorithm includes a plurality of computational algorithms that refine the diagnosis of the patient in a serial manner.

18. The medical device according to any one of claims 1 to 17, characterized in that The at least one calculation algorithm includes a first calculation algorithm and a second calculation algorithm, wherein calculations of the first calculation algorithm are based on a common set of data from a plurality of databases, and calculations of the second calculation algorithm are based on all data from a single database.

19. The medical device according to claim 18, characterized in that The second calculation algorithm is selected from a plurality of calculation algorithms, each of which is calculated based on all data from different databases, and the selection of the second calculation algorithm is based on the degree of similarity between the patient data input and the data of the database used by the second calculation algorithm.

20. The medical device according to any one of claims 1 to 19, characterized in that The medical device is arranged in an ambulance.

21. The medical device according to any one of claims 1 to 20, characterized in that The natural language data includes audio data acquired from the microphone, and the processor is further configured to evaluate the audio data for relevance to the neurological disease.

22. The medical device according to any one of claims 1 to 21, characterized in that The at least one computational algorithm and / or the at least one analytical process is further configured to evaluate the patient data input to assess at least one of: limb and gait ataxia, numbness, Horner's syndrome, dysphagia, dysarthria, nausea, vomiting, dizziness, nystagmus, limp, slurred speech, weakness, partial facial drooping, pain, shortness of breath, flushing, sweating, facial expression, chest clutching, respiratory rate, speech interruptions, shortness of breath, difficulty breathing, rapid breathing, or other syndrome elements documented in the medical literature.

Images