Systems and methods for analyzing data to assess a human medical condition - Patents.com

Abstract

A system for determining a medical condition in an individual, comprising: acquiring a first set of digital biomarkers from a subject from a sensor associated with the subject; generating a digital neural signature from the first set of digital biomarkers; generating a first digital neural fingerprint from the digital neural signature, the fingerprint being a comparison of biomarkers in the first digital neural signature to a threshold associated with the first set of biomarkers; identifying a first medical condition in the subject from the first digital neural fingerprint; acquiring a second set of biomarkers from one or more sensors and generating a second digital neural fingerprint. generating a digital neural signature and then generating a second digital neural fingerprint, the second digital neural fingerprint being a comparison of biomarkers in the second digital neural signature to a threshold value associated with the second set of biomarkers; identifying a second medical condition of the subject from the second digital neural fingerprint; and identifying a data pattern in the first and second digital neural fingerprints and using the pattern in the first and second digital neural fingerprints to identify the second medical condition in the first digital neural fingerprint. The system and method can correlate biomarker data from different sources to determine a patient's medical condition, including medical conditions not evident from the core data set. They can be used to identify an individual's cognitive or health condition long before biological symptoms are evident through traditional diagnostic methods.

Description

[Technical field]

[0001] The present invention relates to a system and method for analyzing data to determine a human medical condition. The preferred embodiment provides a system and method capable of correlating datasets (particularly biomarker data) from different sources and determining therefrom a patient's medical condition (not only the specific medical condition associated with a first or core dataset, but also other medical conditions not immediately evident from the core dataset). The system and method may be used to identify an individual's cognitive or health condition, often long before biological symptoms are evident through conventional diagnostic methods. [Background technology]

[0002] As is well documented in the art, the applicant has developed a system and method for analyzing a human condition that generates a Digital Neuro Signature (DNS) based on a set of biomarkers obtained from one or more sensors configured to observe the human over the course of one or more activities. In a preferred practical embodiment of the applicant's technology, an optimal DNS includes 784 active digital biomarkers. The number of digital biomarkers used is not critical to the teachings herein, which may be implemented with any selected or desired set and number of biomarkers. A larger number of biomarkers may optimize the classification of a human's health or cognitive state, consistent with the applicant's currently implemented system.

[0003] The applicant has developed a system for visualizing human biomarkers as a matrix, an example of which is shown in Figure 1. This visualization system is hereafter referred to as Digital Neuro Fingerprint (DNF)®.

[0004] The matrix in this example includes a color-based encoding of the pixels forming the matrix. The color code is used to represent the determined numerical value of the trait / biomarker. In the illustrated example, green is used to represent approximately the average value of the trait / biomarker (based on an existing data set of healthy controls), while red represents the limits of the curve in either direction (below or above the average). The different shades of green and red correspond to the identified numerical values ​​and how close they are to the average.

[0005] The 784 active digital biomarkers in this example are generated by an Inertial Measurement Unit (IMU) of the type that may be found in mobile phones and other wearable devices such as smart watches and smart bracelets (one example is FitBit®). Techniques for obtaining the active digital biomarkers are disclosed in detail in applicant's previous patent applications and in many widely available, more recent publications in the art. Such techniques are not a limiting factor of the teachings herein.

[0006] The DNF provided by the matrix of Figure 1 is a very useful tool in analyzing human disease states. It can provide a visual matrix in which patterns can be easily identified and correlated between subjects, generally known to be healthy or known to have a particular disease or illness. The system and method can be used, for example, to identify cognitive impairments such as Alzheimer's disease, but can be used for many other applications beyond cognitive function.

[0007] It has been found that the system can identify the onset of cognitive impairment at a critical stage before conventional methods can do so. Early detection of diseases such as Alzheimer's can allow for much more effective early treatment and, in the absence of treatment, can allow for a significant slowing of the progression of the disease.

[0008] The technology developed by the applicant for diagnosing the early onset of diseases such as Alzheimer's disease and other cognitive impairments has been widely reported. Additionally, in the summer of 2021, the applicant was awarded FDA Breakthrough Designation for the development of the world's first precision neurological device for predicting Alzheimer's disease.

[0009] The system developed by the applicant can capture multi-dimensional digital biomarkers and is not limited to latency-based or accuracy-based measurements. Several objectively measured characteristics can be integrated into a single task. This integration can result in a more generalizable "real world situation" than in traditional clinical trial settings, thus increasing the ecological validity of the observations. The large amount of data collected by the applicant's system and the novel combination of multiple variables addressing multiple cognitive domains as well as sensor data results in higher sensitivity, especially when variability measures are considered. The digital biomarker platform produces large amounts of high-resolution data that may include cognitive processing and signaling cellular processing; voice-based data indicative of emotional states and micro-errors that reveal where, when and how disease manifestations affect daily functioning. These data have the potential to be further exploited for disease progression modeling, more accurate prediction of transformation events or drug effect modeling leading to large-scale non-invasive life-long monitoring of brain health.

[0010] The teachings herein build upon existing technology which is well documented by the applicant.

[0011] Summary of the Invention The present invention seeks to provide a system and method that can extend the above-described techniques in a manner that preferably allows for the identification of other diseases and illnesses that are not currently ascertainable. In a preferred embodiment, this is accomplished by collecting data from other sources and correlating the data from the other sources with the existing or core digital neuro-fingerprint to generate a new digital neuro-fingerprint that can identify data patterns and sets in the original digital neuro-fingerprint and associated signatures that are indicative of these other diseases or illnesses and that cannot be identified from the core fingerprint / signature alone. In another aspect, data from other sources (typically other sensors) is used to enhance the digital neuro-fingerprint to provide better identification of an individual's medical condition and / or to create a new digital neuro-fingerprint that can identify a first medical condition based on other sensor sources.

[0012] According to one aspect of the present invention, there is provided a method of assessing a medical condition in an individual, the method comprising: acquiring a first set of digital biomarkers from one or more sensors associated with the subject; generating a first digital neural signature for the subject from the first set of digital biomarkers; generating a first digital neural fingerprint from the first digital neural signature, the first digital neural fingerprint being a comparison of biomarkers in the first digital neural signature to a threshold value associated with the first set of biomarkers; identifying a first medical condition in the subject from the first digital neural fingerprint; acquiring a second set of biomarkers from the one or more sensors and generating a second digital neural signature therefrom; generating a second digital neural fingerprint from the second digital neural signature, the second digital neural fingerprint being a comparison of biomarkers in the second digital neural signature to a threshold value associated with the second set of biomarkers; identifying a second medical condition in the subject from the second digital neurofingerprint; identifying any data patterns in the first and second digital neural fingerprints; and identifying the second medical condition in the first digital neuro-fingerprint using any patterns identified in the first and second digital neuro-fingerprints.

[0013] Advantageously, the first and second sets of biomarkers are obtained on different subjects.

[0014] The first and second sets of biomarkers may be associated with different disease states, such as different stages of a disease or condition, for example a pre-clinical stage of a disease or condition on the one hand and a pro-symptomatic or clinical stage of a disease or condition on the other hand. The first and second sets of biomarkers may be associated with different diseases or conditions.

[0015] In one embodiment, any patterns identified in the first and second digital neuro-fingerprints are used to enhance the determination of the first medical condition in the first digital neuro-fingerprint.

[0016] Preferably, the threshold value represents the mean value of that biomarker in a group of subjects believed to have a given condition, which may be a healthy condition.

[0017] The method preferably comprises the steps of: generating a prototypical first digital neural fingerprint from a set of said first digital neural fingerprints obtained from a set of subjects; generating a prototypical second digital neural fingerprint from the set of second digital neural fingerprints; A step of identifying any data patterns is performed on the prototypical first and second digital neural fingerprints.

[0018] Advantageously, the method comprises the steps of: generating a new first digital neural fingerprint from any data pattern matches determined in the identifying step; The new first digital neural fingerprint is indicative of the second medical condition of the subject.

[0019] In a preferred embodiment, the digital neuro-fingerprint is formed from a plurality of elements, each associated with a biomarker, each element having a value indicative of the deviation of the measured biomarker from said threshold. Each value of the digital neuro-fingerprint advantageously lies within a given range. In a preferred embodiment, the value of each element is an optical value, which optical value is preferably a colour that varies depending on the deviation of the measured biomarker from the associated threshold. Identifying patterns in the first and second digital neuro-fingerprints is preferably by optical pattern recognition.

[0020] It will be appreciated that color pattern recognition is just one exemplary implementation: other schemes may be used, including, for example, numerical or other marks, grayscale scales, black and white patterns, and the like.

[0021] In one embodiment, the first and second sets of biomarkers are obtained from different sensors or different sets of sensors.

[0022] The method may include generating a combined digital neural fingerprint from the first and second digital neural fingerprints, and utilizing the combined digital neural fingerprint in determining a first and / or second medical condition of the subject.

[0023] The method may include generating a new digital neurological fingerprint from the first and second digital neurological fingerprints, and utilizing the new digital neurological fingerprint in determining the first and / or second medical condition of the subject.

[0024] Advantageously, the first and / or second sets of biomarkers are obtained from one or more of a smartphone, a tablet computer, a smart watch, a smart bracelet, a pair of smart glasses, and a camera.

[0025] In a preferred embodiment, the methods may be used in the diagnosis of a disease or category of diseases, for example in the diagnosis of one or more of Parkinson's disease, Alzheimer's disease, ALS.

[0026] Preferably, a match is identified when the number of elements of the first and second digital neuro-fingerprints exceeds a set percentage. For example, a match may be identified when the number of elements of the first and second digital neuro-fingerprints exceeds 80% of the total number of elements in at least one digital neuro-fingerprint. It should be understood that the threshold can be selected by one skilled in the art and may deviate from the examples given. A lower threshold may find more pattern matches between data sets but is potentially less reliable, while a higher threshold is likely to find fewer matches but is more reliable.

[0027] Preferably, the step of identifying a pattern generates and uses Shapley values.

[0028] According to another aspect of the present invention, there is provided a system for determining a medical condition of an individual, comprising: one or more sensors associated with the subject, the sensors operable to acquire one or more biomarkers of the subject; a processing unit for processing the biomarkers, the processing unit comprising: a first register containing a first set of digital biomarkers from the subject obtained from one or more sensors associated with the subject; a first digital neural signature generator configured to generate a first digital neural signature for the subject from the first set of digital biomarkers; a first digital neural fingerprint generator configured to generate a first digital neural fingerprint from the first digital neural signature, the first digital neural fingerprint being a comparison of biomarkers in the first digital neural signature to a threshold value associated with the first set of biomarkers; an identification processor configured to identify a first medical condition in the subject from the first digital neurological fingerprint; an input configured to receive a second set of biomarkers from the one or more sensors; a second digital neural signature generator configured to generate a second digital neural signature from the second set of biomarkers; a second digital neural fingerprint generator configured to generate a second digital neural fingerprint from the second digital neural signature, the second digital neural fingerprint being a comparison of biomarkers in the second digital neural signature to a threshold value associated with the second set of biomarkers; a data pattern identifier configured to identify any data patterns in the first and second digital neuro-fingerprints; The processing unit is configured to utilize any identified patterns in the first and second digital neuro-fingerprints to identify the second medical condition in the first digital neuro-fingerprint.

[0029] The first and second sets of biomarkers may be obtained from different subjects. The first and second sets of biomarkers may be associated with different disease conditions. They may be associated with different stages of a disease or condition, for example, the first set of biomarkers may be associated with a pre-clinical stage of a disease or condition and the second set of biomarkers may be associated with a pro-symptomatic or clinical stage of a disease or condition.

[0030] Preferably, the processing unit is configured to provide an enhanced determination of the first medical condition from any patterns identified in the first and second digital neuro-fingerprints.

[0031] Advantageously, said threshold value represents the mean value of that biomarker in a group of subjects believed to have a given condition.

[0032] The system includes a prototypical first digital neural fingerprint generator configured to generate a prototypical first digital neural fingerprint from a set of first digital neural fingerprints obtained from a set of subjects, and a prototypical second digital neural fingerprint generator configured to generate a prototypical second digital neural fingerprint from a set of second digital neural fingerprints, and the processing unit is configured to identify a data pattern from the first and second prototypical digital neural fingerprints.

[0033] The system may include a new first digital neuro-fingerprint generator configured to generate a new first digital neuro-fingerprint from any data pattern matches determined in the identification step, the new first digital neuro-fingerprint being indicative of a second medical condition of the subject.

[0034] In a preferred embodiment, the first and second digital neuro-fingerprint generators are configured to form the first and second digital neuro-fingerprints as arrays of elements, each of which is associated with a biomarker, and in the array, each element has a value representing the deviation of the measured biomarker from a threshold value.

[0035] In a practical embodiment, the value of each element is an optical value, each element being a pixel of a display, and the system includes an optical sensor for determining the optical value, so the processing unit may include an optical pattern recognition unit.

[0036] The system may include a combined digital neural fingerprint generator configured to generate a combined digital neural fingerprint from the first and second digital neural fingerprints, and the processing unit configured to utilize the combined digital neural fingerprint in determining the first and / or second medical condition of the subject.

[0037] The system may include a new digital neural fingerprint generator configured to generate a new digital neural fingerprint from the first and second digital neural fingerprints, and the processing unit configured to utilize the new digital neural fingerprint in determining the first and / or second pathology of the subject.

[0038] Advantageously, the system includes one or more of a smartphone, a tablet computer, a smart watch, a smart bracelet, a pair of smart glasses, and a camera configured to acquire said first and / or second set of biomarkers.

[0039] The system advantageously includes a diagnostic unit configured to diagnose a disease or category of disorders, such as one or more of Parkinson's disease, Alzheimer's disease, ALS.

[0040] According to another aspect of the present invention, there is provided a method for assessing a medical condition in an individual, comprising: acquiring a first set of digital biomarkers from a first set of sensors associated with the subject; generating a digital neural signature for the subject from the first set of digital biomarkers; generating a first digital neural fingerprint from the digital neural signature, the first digital neural fingerprint being a comparison of biomarkers in the first digital neural signature to a threshold value associated with the first set of biomarkers; identifying a first medical condition in the subject from the first digital neural fingerprint; acquiring a second set of biomarkers from a second set of sensors different from the first set of sensors and generating a second digital neural signature therefrom; generating a second digital neural fingerprint from the second digital neural signature, the second digital neural fingerprint being a comparison of biomarkers in the second digital neural signature to a threshold value associated with the second set of biomarkers; identifying any data patterns in the first and second digital neural fingerprints; Utilizing any patterns identified in the first and second digital neural fingerprints to enhance determination of the first medical condition in the subject.

[0041] The first and second sets of biomarkers may be obtained from different subjects.

[0042] Advantageously, the first and second sets of biomarkers are associated with different disease states, for example different stages of a disease or condition and / or different diseases or conditions.

[0043] The threshold value may represent the mean value of that biomarker in a group of subjects considered to have a given condition, eg, a healthy condition.

[0044] The method includes generating a prototypical first digital neural fingerprint from a set of the first digital neural fingerprints obtained from a set of subjects, and generating a prototypical second digital neural fingerprint from a set of the second digital neural fingerprints, and a step of identifying any data patterns is performed in the prototypical first and second digital neural fingerprints.

[0045] The method includes generating a new first digital neuro-fingerprint from any data pattern matches determined in the identifying step, the new first digital neuro-fingerprint being indicative of the first medical condition of the subject.

[0046] In certain embodiments, the digital neuro-fingerprint is formed from a number of elements, each associated with a biomarker, and each element has a value indicative of the deviation of the measured biomarker from the threshold value.

[0047] The method preferably includes identifying patterns in said first and second digital neuro-fingerprints by optical pattern recognition.

[0048] The method includes generating a combined digital neural fingerprint from the first and second digital neural fingerprints, and utilizing the combined digital neural fingerprint in determining a first and / or second medical condition of the subject.

[0049] The method may include generating a new digital neurological fingerprint from the first and second digital neurological fingerprints, and utilizing the new digital neurological fingerprint in determining the first and / or second medical condition of the subject.

[0050] According to another aspect of the present invention, there is provided a system for determining a medical condition of an individual, comprising: a first set of sensors associated with the subject, the first set of sensors being operable to acquire one or more biomarkers of the subject; a processing unit for processing the biomarkers, the processing unit comprising: a first register containing a first set of digital biomarkers from the subject acquired from the first set of sensors associated with the subject; a first digital neural signature generator configured to generate a first digital neural signature for the subject from the first set of digital biomarkers; a first digital neural fingerprint generator configured to generate a first digital neural fingerprint from the first digital neural signature, the first digital neural fingerprint being a comparison of biomarkers in the first digital neural signature to a threshold value associated with the first set of biomarkers; an identification processor configured to identify a first medical condition in the subject from the first digital neurological fingerprint; an input configured to receive a second set of biomarkers from a second set of sensors different from the first set of sensors; a second digital neural signature generator configured to generate a second digital neural signature from the second set of biomarkers; a second digital neural fingerprint generator configured to generate a second digital neural fingerprint from the second digital neural signature, the second digital neural fingerprint being a comparison of biomarkers in the second digital neural signature to a threshold value associated with the second set of biomarkers; a data pattern identifier configured to identify any data patterns in the first and second digital neuro-fingerprints; The processing unit is configured to utilize any patterns identified in the first and second digital neuro-fingerprints to enhance a determination of the first medical condition in the subject.

[0051] While various aspects of the invention disclosed herein are set forth above and in the claims, it should be understood that these aspects can be combined together into single methods and systems and that elements thereof can also be combined with each other in a similar manner.

[0052] A preferred embodiment of the present invention provides a system and method for creating a new Digital Neuro Fingerprint (DNF) using a data set from a base Digital Neuro Fingerprint (DNF) augmented with data obtained from other sources. The additional data may be collected from other devices, including inertial measurement units, but may also be collected from any other data source using other sensors as described below.

[0053] The system and method are configured to correlate data from such other sources into a common characterization framework (in this example, the color coding scheme shown in FIG. 1, or more preferably, common to that used in the base DNF generated from the base digital neural signature). The additional data can be used to generate combined digital neural fingerprints or new combinations of digital neural fingerprints that can be used not only in identifying previously known pathologies, but also in identifying potential new pathologies in patients not immediately derivable from the core DNF.

[0054] In a preferred embodiment, the system and method uses optical pattern recognition to locate data patterns useful for identifying a health or cognitive condition.

[0055] In practical embodiments, the system and method provide: (a) Creating a new DNF within a dataset that does not initially collect data using the device(s) for collection of the base set of biomarkers, using other devices including IMUs or other sensors. Examples include smart watches or bracelets, such as FitBit® and Apple Watch®; (b) mathematical processing of these DNFs using optical pattern recognition devices or methods; (c) Creation of a combined DNF or combination of DNFs using a core set of active digital biomarkers and / or novel digital biomarkers utilizing sensors other than IMUs.

[0056] A major advantage of the systems and methods disclosed herein is that they allow for the rapid repurposing of DNFs obtained from one source, such as the devices that are the subject of Applicant's previously filed patent applications, to new disease areas based on DNF similarities. Once such similarities are found in combined DNFs or new combinations of DNFs, clinical trials for specific validation in new disease areas can require significantly fewer patients and reduce study time, thus making clinical trials more efficient and agile.

[0057] According to another aspect of the invention, there is provided a use of a determination of an identified medical condition in a subject obtained by a method or by a system as described in any of the preceding paragraphs to prescribe a medication or other treatment to treat the identified medical condition.

[0058] According to another aspect of the invention, there is provided a method of treating a subject based on a determination of an identified medical condition of the subject obtained by a method or system as described in any of the preceding paragraphs to administer a medication or other treatment to treat the identified medical condition.

[0059] In the uses or methods, the suggested treatment may be a pharmaceutical intervention and the information relates to characteristics of the particular drug to be administered to the individual.

[0060] In the use or method, the drug may be a cholinesterase inhibitor (such as donepezil, rivastigmine, galantamine), memantine (optionally in combination with a cholinesterase inhibitor), a monoclonal antibody (such as aducanumab (Aduherm)), BAN2401, gantenerumab (optionally in combination with solanezumab), solanezumab (optionally in combination with gantenerumab), a sigma-1 receptor agonist (optionally also an M2 autoreceptor antagonist such as ANAVEX2 (bulcamesine), AVP-786 or AXS-05, or or NMDA receptor antagonists), SV2A modulators (such as AGB101 (low dose levetiracetam)), mast cell stabilizers (such as ALZT-OP1 (cromolyn + ibuprofen)), anti-inflammatory agents (such as ALZT-OP1 (cromolyn + ibuprofen)), RAGE antagonists (such as azeliragon), glutamate modulators (such as BHV4157 (troriluzole)), D2 receptor partial agonists (such as brexpiprazole), serotonin dopamine modulators (such as brexpiprazole), amyloid Vaccines (such as CAD106), bacterial protease inhibitors (such as COR388), selective serotonin reuptake inhibitors (such as escitalopram), antioxidants (such as ginkgo), plant extracts (such as ginkgo), alpha-2 adrenergic agonists (such as guanfacine), omega-3 fatty acids (such as ethyl eicosapentaenoate (IPE), a purified form of eicosapentaenoic acid), angiotensin II receptor blockers (such as losartan), calcium channel blockers (such as amiodipine), cholesterol agents (such as atorvastatin). ), combinations of angiotensin II receptor antagonists (such as losartan), calcium channel blockers (such as amiodipine), and cholesterol agents (such as atorvastatin) with or without exercise, tyrosine kinase inhibitors (such as masitinib), insulin sensitizers (such as metformin), dopamine reuptake inhibitors (such as methylphenidate), alpha-1 antagonists (such as mirtazapine), acetylcholinesterase inhibitors (such as octohydroaminoacridine succinate), ketogenic stimulants (such as tricaprylin),Caprylic triglyceride (such as tricaprylin), tau protein aggregation inhibitors (such as AADvac1 or TRx0237 (LMTX)), positive allosteric modulators of the GABA-A receptor (zolpidem and zoplicone), or BPDO-1603, or any combination thereof administered together or separately.

[0061] In the use or method, aducanumab (Aduherm) is suitably administered to the subject.

[0062] In the uses or methods, the suggested treatment may be a pharmaceutical intervention and the information relates to the frequency and / or dosage of the pharmaceutical intervention or a particular drug to be administered to the individual.

[0063] In the use or method, the individual may have been previously diagnosed with mild cognitive impairment, and the information provided by the information output relating to a pharmaceutical intervention or other treatment is related to whether a previously prescribed treatment was effective for the individual.

[0064] Other advantages and aspects of the present invention will become apparent to those skilled in the art from the following detailed description.

[0065] Embodiments of the invention are hereinafter described, by way of example only, with reference to the accompanying drawings, in which: [Brief description of the drawings]

[0066] [Figure 1] FIG. 1 is an example of a digital neural fingerprint generated from a digital neural signature itself generated from a set of biomarkers obtained from an individual. [Diagram 2] 1 is a flow diagram of one embodiment of a system and method for determining pattern matches in digital neuro-fingerprints from multiple different sets of subjects. [Diagram 3] FIG. 13 illustrates an example of a pattern match. [Figure 4] FIG. 13 shows an example of pattern matching using Shapley values. [Diagram 5] 1 is an example of a pattern matching matrix. [Figure 6] An example of a combined digital neural fingerprint. [Figure 7] 1 is a flow diagram of one embodiment of a system and method for generating a new combined digital neural fingerprint in accordance with the teachings herein. [Figure 8] 1 is a schematic diagram of one embodiment of a system in accordance with the teachings herein. [Figure 9] 1 is a schematic diagram of one embodiment of a processing unit in accordance with the teachings herein.

[0067] Description of the Preferred Embodiments Activity Biomarkers Biomarkers aid in the diagnosis of heart disease. Biomarkers or biological markers (biological indicators) refer to parameters that can be measured to reliably and accurately indicate the presence and severity of a disease state. Biomarkers may include a wealth of measurable indicators ranging from an elevated white blood cell count to indicate an infection, to the presence of beta amyloid plaques in the brain to indicate Alzheimer's disease.

[0068] With the rise of digital health data collection, researchers and providers alike are leveraging the potential of digital biomarkers. Applicant's systems can collect clinically meaningful data via digital devices that Applicant has discovered can provide new and more robust ways to monitor and diagnose an individual's condition. They enable the collection and analysis of physiological and behavioral data that can be used for predictive diagnosis of disease.

[0069] Digital biomarkers can provide and facilitate earlier diagnosis for earlier access to healthcare and treatment when diseases are more treatable, providing better health outcomes.

[0070] Digital biomarkers are quantifiable, generally objective physiological and behavioral data that can be collected and measured via digital devices. Exemplary devices include portable, wearable, implantable, or ingestible. Digital biomarkers yield robust data sets that can be used to learn more about the nuances of specific diseases and gain valuable health insights.

[0071] Passive data from sensors integrated into a wearable device, such as a smartwatch, is generated when a user simply wears the device. The collected data is then referred to as passive digital biomarker data. Similarly, digital biomarker data can be generated and captured from smart devices, such as smartphones and tablets, when a user interacts with the device in response to active prompts. Integrated or separate sensors, including cameras, microphones, touch screen sensors, accelerometers, and gyroscopes, can be used to collect active digital biomarker data.

[0072] While wearables collect more obvious data like heart rate, heart rate variability, and oxygen saturation levels from photoplethysmography sensors, smartphones and tablets collect less intuitive, yet very powerful, data. Some examples of digital biomarker data that can be collected from smart devices include: (a) a microphone (which can be used to detect biomarkers of speech, such as fluency, mood, and emotion); (b) a camera (capable of detecting eye movements, pupil dilation, and facial expressions); (c) touch screen sensors (capable of identifying the fine motor skills required for tapping, swiping, and typing); (d) Inertial sensors (including accelerometers and gyroscopes capable of detecting human motion and posture and enabling measurements of gait metrics) Includes.

[0073] Applicant has found that the numerous and types of biomarkers used can provide a comprehensive view of an individual's condition. They can include motor markers, including, for example, speed of movement, range of movement, force applied to an activity (e.g., pressing a button or pressure-sensitive pad), tremors, and reaction time, and they may also include cognitive indicators, such as time to complete a task, accuracy of a movement or reaction, and accuracy and time to complete a series of exercises.

[0074] Digital biomarkers can also enable longitudinal data collection at individual and population levels. Most tools used to assess brain health lack the infrastructure for longitudinal analysis and generally only provide a means for cross-sectional data collection and analysis. Longitudinal data provides the ability to analyze brain health on an individualized basis to gain insight into how an individual's brain health is changing over time. Using longitudinal data that tracks the evolution of digital biomarkers over time, a processing unit (preferably an artificial intelligence) can make predictions from the data to determine if, when, and how an individual is developing a particular disease or condition, such as Alzheimer's disease.

[0075] As more and more individuals take advantage of newer health-related technologies, the amount of available health data is growing at an astonishing rate. When this amount of data is combined with powerful analytical tools, it can potentially be leveraged to track trends and patterns related to many diseases.

[0076] The Applicant has developed a sophisticated neuroplatform capable of measuring and analyzing a wide range of cognitive and functional digital biomarkers, which provides a comprehensive analysis of neurocognitive function and more general function at an individualized level. The platform has been developed to collect and analyze approximately 800 active digital biomarkers, which in a preferred embodiment allows for specific, accurate and generalizable data for cross-sectional and longitudinal analysis of an individual's health.

[0077] Further information relating to such datasets and the applicant's systems and methods for generating digital neural signatures (DNS) are disclosed in U.S. patent application Ser. No. 63 / 211,953, filed June 17, 2021, which is incorporated by reference in its entirety into this specification.

[0078] Creating a subject's core DNF The system and method are configured to generate a data set from commonly available sensors. This may be provided by wearable sensors, for example. These may include, for example, one or more of the following: PIR motion sensors, body-worn sensors (usually incorporating an IMU), pressure sensors, video surveillance, and audio recordings. The primary embodiment preferentially uses (i) body-worn sensors, and (ii) video surveillance. The signals from the sensors are compared to a predefined set of biomarkers, and from this comparison, a core digital neural fingerprint is created based on whether the output of the sensor(s) matches a predefined condition (pathology) associated with the biomarkers. When multiple sensors are used, the preferred embodiment treats the condition as satisfied if the output of only one sensor indicates the condition. In other embodiments, the system and method may be configured to request the output of multiple sensors or the output of each sensor matching the condition, or some other correlation of the outputs to be matched.

[0079] The predetermined condition may be a response obtained from a healthy individual, while in other embodiments it may be a response obtained from an individual with a specified condition (e.g., cognitive impairment, as just one example). Whatever it may be, any reliable response may be used as the benchmark.

[0080] Use cases / embodiments In one embodiment according to the teachings herein, a public domain dataset may be provided for a particular disease or illness (e.g., Parkinson's disease) that includes both body-worn sensor data points (e.g., smartwatch or FitBit®-like bracelet) and video surveillance modalities (e.g., home cameras). In this embodiment, researchers in the field may segment the data stream from the body-worn sensor into IMU outputs that correspond to specific activities associated with "acts" of daily life captured by the video surveillance system. These acts most commonly have a set duration, e.g., 30-90 seconds or more. One example is the IMU sensor output obtained while an individual is walking to find their car keys or walking into their living room to look for an object such as the TV remote. One example of such a dataset is the CART Home® by ORCATECH Laboratories and many others. Namely, https: / / www.ohsu.edu / collaborative-aging-research-using-technology / cart-home

[0081] Once the IMU outputs have been separated, they can be mapped by someone with knowledge of the digital neural signature to a set of digital biomarkers (e.g., the Altoida 784 DNS) of the digital neural signature. The result of this process is the generation of a set of (in this example, a set of 784) distinct numerical inputs that correspond to the digital neural signature and have durations similar to the "action" time windows of the relevant actions, such as the time it takes to walk to find your keys, the time it takes to find the remote control, etc. The sum of these time windows is typically anything between 2 and 10 minutes.

[0082] The digital neural signature is then mapped to a reference matrix (the matrix in FIG. 1 is an example) to generate a digital neural fingerprint at the end of each session. In practical terms, such a process can be expected to take a time in the range of 10 minutes. The process, once set up, can be automated to accomplish the mapping quickly.

[0083] The different numerical inputs are preferably associated with particular behaviors or other measurable characteristics.

[0084] In another embodiment, the public domain of sensors available or used includes only body-worn sensor data points and does not include video surveillance modalities. For example, in this embodiment, uncorroborated information from the subject's activities during the previous day is used to separate the IMU output into time frames corresponding to "acts" of daily living activities. Once these activities are completed, the same process as above is followed. In this particular example, the resulting digital neural fingerprint may contain more noise than a dataset having both body-worn sensor data points and video surveillance data points. However, that noise can be taken into account in subsequent processing or by relying on a larger dataset.

[0085] New DNF(multiple) neural network processing The preferred method in this example generates “prototypical DNFs” for different disease categories in different disease datasets (e.g., Parkinson's, Alzheimer's, ALS, etc.) and compares them together using an optical matching algorithm.

[0086] Use cases / embodiments In one embodiment, the Parkinson's Disease (PD) public domain dataset satisfies the above-mentioned conditions for the generation of one or more digital neural fingerprints and includes, in this example, the following patient categories: a) Pre-symptomatic PD (n=100) b) Prodromal PD (n=50) c) Clinical treatment PD (n=500), etc.

[0087] The first step in the process is to create a digital neurofingerprint for each of these disease categories: 100 digital neurofingerprints of pre-symptomatic subjects 50 digital neurofingerprints of prodromal subjects · 500 digital neurofingerprints of clinically treated PD subjects.

[0088] The following algorithm is then applied to generate one 'prototypical' digital neuro-fingerprint for each of these categories: one for pre-symptomatic, one for prodromal and one for clinically treated PD.

[0089] For each of the digital neurofingerprint squares in each category (e.g., pre-symptomatic subjects), the algorithm starts with the top left square and counts how many were red, green, light green, light red, etc., out of the 100 cases of pre-symptomatic subjects. Majority wins. For example, if there were 70 red biomarkers, 20 light red biomarkers, 7 light green biomarkers, and 3 green biomarkers within the 100 cases of pre-symptomatic PD subjects, then the "prototypical" digital neurofingerprint for the pre-symptomatic PD subject category would have the top left square set as red (the first one in this process). The same process is done for the remaining 783 squares until a new "prototypical digital neurofingerprint" is generated for the pre-symptomatic PD subject category.

[0090] The same process is performed for each of the other categories in this data set, thereby generating three "prototypical" digital neural fingerprints. The numerical values ​​that cause the color coding are not important for this particular algorithm; the process simply ranks the color codings with respect to generating a "prototypical" digital neural fingerprint. The end result may be a less than accurate representation of the average digital biomarker value for that category, but this is acceptable for this embodiment.

[0091] The same process is repeated with respect to a second dataset for this comparison, which may be, for example, an Alzheimer's Disease (AD) dataset from the applicant's existing database, having the following categories: a) Pre-symptomatic AD (n=20) b) Prodromal AD (n=500) c) Clinical treatment AD (n=50), etc.

[0092] A total of three prototypical digital neural fingerprints are generated for the Alzheimer's disease dataset using the same algorithm as above.

[0093] The final step in the process is optical pattern recognition based on the prototypical digital neural fingerprints. An artificial neural network or another pattern recognition algorithm is used to compare the six prototypical digital neural fingerprints generated in the above example for similarity. An acceptable match in this embodiment is >80%. Other embodiments may set the match threshold to a higher or lower percentage, as desired.

[0094] Once such matches are found, it may be demonstrated, for example, that a prototypical digital neural fingerprint of a pre-symptomatic PD subject has >80% match with a prototypical digital neural fingerprint of a prosymptomatic AD subject, and >90% match with a prototypical digital neural fingerprint of a pre-symptomatic AD subject, etc.

[0095] The most matching digital neural fingerprints are then selected and reverse engineered to create a clinical profile based on the clinical information present in the dataset. For example, a prototypical digital neural fingerprint of a pre-clinical PD subject based on color coding may be most frequently associated with sleep disorders in 59% of subjects, depressive symptoms in 76% of subjects, and visual hallucinations in 74% of subjects. Furthermore, a prototypical pre-clinical AD digital neural fingerprint may be most frequently associated with micromotor disorders in 61% of subjects, mood disorders in 46% of subjects, and sensorimotor dysfunction in 64% of subjects.

[0096] FIG. 2 shows an exemplary flow diagram for implementing this process.

[0097] With reference to the flow diagram, in step 100, digital neuro-fingerprints of pre-clinical PD subjects 1-100 are collected from which a prototypical digital neuro-fingerprint of pre-clinical PD is generated in step 102. Similarly, digital neuro-fingerprints of pro-symptomatic PD subjects are acquired and processed in step 104 to generate a prototypical digital neuro-fingerprint of pro-symptomatic PD in step 106. As described above, these are combined with the digital neuro-fingerprint of Alzheimer's Disease in the following steps. In step 108, digital neuro-fingerprints of pre-clinical AD subjects (1-20 in this example) are acquired from which a prototypical digital neuro-fingerprint of pre-clinical AD is generated in step 110. Similarly, digital neuro-fingerprints of pro-symptomatic AD subjects (500 in this example) are acquired in step 112 to generate a prototypical digital neuro-fingerprint of pro-symptomatic AD in step 114.

[0098] In step 116, in this example, using a processing unit, preferably an artificial neural network, patterns within the prototypical digital neural fingerprint are identified and from these patterns new digital neural fingerprints can be created for previously unidentified Parkinson's and Alzheimer's diseases.

[0099] All of this revealed information will help select the appropriate neuropsychological battery for validation in clinical trials, which would otherwise require having more comprehensive neuropsychological batteries for data collection. It can also speed up the results of such clinical trials for validation, because investigators can collect newly created digital neural fingerprints from clinical trials and use the same optical recognition algorithms described above to see how well they match with the "prototypical digital neural fingerprints" of the population.

[0100] A prototypical digital neural fingerprint-to-clinical profile reverse engineering method utilizes feature contributions to predictions using Shapley values ​​(see, for example, https: / / github.com / slundberg / shap) as Feature Interaction Score (FIS) and estimates the value of data points based on their Shapley impact on the model output. With reference to Figures 3 and 4, once a pattern of squares corresponding to FIS has been identified, the process creates a list of the most significant colored digital neural fingerprint squares i (1-784) and assigns them a value Φ i and their actual average value is taken. The most important squares are the ones that are found to have the greatest contribution to the match and are most relevant to the associated biomarker. For example, for a biomarker related to the analysis of a subject while performing an activity, the most important squares are the ones that are most relevant to that activity (e.g., eye movements, subject behavior (like which direction the subject faces)). Other biomarkers, such as blood pressure or sleep patterns, are considered less relevant.

[0101] From this, in the process, the system generates the digital neural fingerprint Φ i-m(or more) We compute the nonlinear patterns between the sigma-based neural fingerprints and the clinical data from the original dataset using XGBoost and SHAP “interaction effects.” The final result is a digital neural fingerprint, Φ i-m(or more) Generate a matrix of the most relevant clinical features that can reveal interesting interactions with the disease. An example is shown in Figure 5.

[0102] Combined DNFs or creating new DNF combinations In some cases, the dataset includes new types of data (e.g., grip force data) from Internet of Things (IoT) wearable sensors that are not part of the set of (e.g., 784) DNS digital biomarkers. Similarly, the DNS digital biomarker library may be updated with new types of sensor data that are not part of the IMU. In these cases, the following steps can be followed:

[0103] (i) Use Cases / Implementation Examples If the Parkinson's disease public domain dataset includes new types of data, such as grip force data, from medical devices and from body-worn sensor data points (e.g., from FitBit®) or one video surveillance modality (e.g., from a smart home camera), one skilled in the art can segment the new types of data streams from grip force to either the body-worn IMU output or the video system output as a first step in creating a digital neural fingerprint. Similarly, if a video system is not used for annotation, anecdotal annotations can be used. The new types of sensor values ​​are then utilized to create a new, independent digital neural fingerprint (separate from the core digital neural fingerprint created by the set of (e.g., 784) digital neural signature digital biomarkers).

[0104] As an illustrative example, a new digital neural fingerprint may have 120 digital neural fingerprint digital biomarkers and is generated based on the same color-coding rules. In other words, the new digital neural fingerprint is generated using an identifier framework or system to enable comparison with other digital neural fingerprints, whether generated from the same sensor device or from other types of sensors. A possible visualization of the resulting digital neural fingerprint including both the "core digital neural fingerprint" (e.g., 784 digital biomarkers) and the "novel (new type) digital neural fingerprint" (120 digital biomarkers) is shown in FIG. 6.

[0105] Essentially, a new combined digital neural fingerprint is created that combines the “core digital neural fingerprint” and the “novel digital neural fingerprint.” This combination shape can be done by starting from the bottom left and adding new shapes to it (e.g., “novel DNF2,” “novel DNF3,” etc.).

[0106] After creation of the combined digital neurofingerprint, new "prototypical combined DNFs" can be created for each of the patient disease categories in the dataset, e.g., a) pre-symptomatic PD (n=100), b) prodromal PD (n=50), c) clinical treatment PD (n=500), etc. The final step in this process is optical pattern recognition based on the "prototypical combined DNFs".

[0107] When comparing the "prototypical combined DNF", the following situations may occur: a) the prototypical combined digital neural fingerprint is >80% matched, b) it is >80% matched with only the core digital neural fingerprint, and c) it is >80% matched with only the "novel" digital neural fingerprint. If the prototypical combined digital neural fingerprint matches situation (a) above, the match is reverse engineered such that a clinical profile is created based on the clinical information present in the dataset. If only the "core DNF" or the "novel DNF" match, the following procedure can be followed: the optical recognition search is repeated focusing on the non-matching digital neural fingerprints (e.g., core or novel) until a >80% match is found. These two separate matches (the match on the "core DNF" and the match on the "novel DNF") are then used to create a new "prototypical combined DNF" that is the result of a deeper search and is not the result of the initial annotation based on the creation described above.

[0108] FIG. 7 shows an exemplary flow diagram for implementing this process.

[0109] 7, in this example, in step 200, the pre-clinical Parkinson's core and novel digital neural fingerprints are combined with combined digital neural fingerprints from multiple subjects 1-100 (step 202) to generate a prototypical pre-clinical Parkinson's combined digital neural fingerprint. In parallel, in step 204, the prodromal Parkinson's core and novel digital neural fingerprints are combined with combined digital neural fingerprints from multiple prodromal PD subjects 1-50 (step 206) to generate a prototypical prodromal Parkinson's combined digital neural fingerprint (step 208).

[0110] Further, in step 212, the pre-clinical Alzheimer's Disease core and novel digital neural fingerprints are combined with combined digital neural fingerprints from multiple pre-clinical Alzheimer's Disease subjects 1-20 (step 214) to generate a prototypical pre-clinical Alzheimer's Disease combined digital neural fingerprint in step 220.

[0111] Further, in step 222, the prodromal Alzheimer's core and novel digital neural fingerprints are combined with the combined digital neural fingerprints of the plurality of prodromal Alzheimer's subjects 1-50 (step 226) to generate a prodromal Alzheimer's prototypical combined digital neural fingerprint (in step 230). In step 240, the matched patterns are present in a processing unit, preferably an artificial intelligence, as described above, to generate a new prototypical combined digital neural fingerprint that is the result of deeper exploration.

[0112] Referring to step 250, if the combined digital neural fingerprint is determined to be inconsistent in step 240, a new combined digital neural fingerprint is generated by the processing unit (or artificial neural network) based on the prototypical combined digital neural fingerprint of pre-clinical Parkinson's Disease in step 252 generated from the core digital neural fingerprint of the pre-clinical Parkinson's Disease subject and the novel digital neural fingerprint of the prodromal Parkinson's Disease subject (steps 254 and 256, respectively), which is compared to the prototypical combined digital neural fingerprint of pre-clinical Alzheimer's Disease generated in step 260 from the core digital neural fingerprint from the prodromal Alzheimer's Disease subject and the novel digital neural fingerprint from the prodromal Alzheimer's Disease subject (steps 262 and 264, respectively).

[0113] In another example, one of the data sets does not have any "novel DNFs" to be matched. In such a case, matching is done using only the "core DNFs" and the match is reverse engineered such that the clinical profile is enriched with new types of (novel) data (e.g., grip strength). This creates a new clinical profile that may have unique clinical features that inform, for example, a novel subtype of a known disease. While a previous prototypical digital neuro-fingerprint case of pre-symptomatic PD in this example was most frequently associated with sleep disturbances in 59% of patients, depressive symptoms in 76% of patients, and visual hallucinations in 74% of patients, the novel digital neuro-fingerprint may also reveal reduced grip strength in 34% of patients.

[0114] In the case of "prototypical combined DNF", a novel digital neural fingerprint of grip strength reduction in PD may match a similar novel digital neural fingerprint of grip strength reduction in 25% of prodromal AD patients. This is a novel subtype of prodromal PD that can be verified by clinical trials. All of the above examples show that the system and process can reveal information that helps select the appropriate neuropsychological battery for clinical trial validation (otherwise, it is necessary to have a more comprehensive neuropsychological battery for data collection). It can also speed up the outcome of clinical trials in terms of validation, because the investigator (researcher) can collect the newly created digital neural fingerprints from the clinical trial and use the same optical recognition algorithm described herein to see how well they match with the "prototypical combined DNF" of the population.

[0115] Exemplary Devices / Systems Examples of suitable apparatus for implementing the taught systems and methods will be readily apparent to those skilled in the art from the above disclosure, but for completeness, an example is nevertheless described below in connection with FIG.

[0116] 8 illustrates one embodiment of an apparatus 300 for use in embodying the teachings herein. The apparatus 300 in this example includes a mobile device 302 provided with first and second cameras 304, 306 (typically one on the rear (away from the user) and the other on the front (facing the user and on the same side as the display)). The mobile device 302 also typically includes an output unit 310, a position sensor 312 (such as, for example, a GPS module and an accelerometer), a microphone 320, a user input unit 322, and one or more processing units 330, 340, 360.

[0117] The mobile device 302 is preferably a handheld portable device such as a smartphone. However, the mobile device 302 may be any other user portable device. For example, it may be a wearable such as a smart watch or bracelet, smart glasses (eyeglasses) or the like. The mobile device 302 may be a single device or may be embodied in multiple devices such as a smartphone in conjunction with a smart watch or bracelet, or even glasses. FIG. 8 shows such a smart device 420 as an external accessory configured to communicate with the mobile device 302.

[0118] The output unit 310 may include a display 316 and, in some implementations, a projector, such as an eye projector in a pair of smart glasses. The output unit 310 may also include an audio unit 318, such as a speaker and / or an audio output port for earphones or headphones.

[0119] As provided in / for the above teachings, an internal device 400 (generally a processing unit, advantageously an artificial neural network) may be provided for performing computational tasks available over a communication line to the mobile device 302, including but not limited to computation of data from a number of different subjects. The processing unit 400 is typically coupled to the mobile device 302 to exchange data over a network, such as over the Internet, a wireless network, or via a GSM network. In some implementations, the processing unit 400 may include a central processing computer. It should be appreciated that in some embodiments, all processing is performed within the mobile device 302.

[0120] The device may also include an external optical sensor 430, such as a smart home camera or other camera configured to capture images of the subject and communicate them to the mobile unit 302 or to the external processing unit 400, or both, as described above. As will be appreciated, the external optical unit 430 may include a set of cameras or the like, and multiple images of the subject may be captured, either sequentially or simultaneously.

[0121] An example of a processing unit according to the teachings herein is shown in FIG.

[0122] The system includes a number of sensors 502-506 associated with a subject, each sensor operable to acquire one or more biomarkers of the subject, as described above. The system also includes a processing unit 500 for processing the biomarkers and coupled to the sensors using a suitable input / output unit 510. The processing unit 500: a first register 512 containing a first set of digital biomarkers from the subject obtained from one or more sensors associated with the subject; a first digital neural signature generator 514 configured to generate a first digital neural signature for the subject from the first set of digital biomarkers; a first digital neural fingerprint generator 516 configured to generate a first digital neural fingerprint from the first digital neural signature, the first digital neural fingerprint being a comparison of biomarkers in the first digital neural signature to a threshold value associated with a first set of biomarkers; and an identification processor 518 configured to identify a first medical condition of the subject from the first digital neurological fingerprint.

[0123] The input unit 510 of the processing unit 500 is also configured to receive a second set of biomarkers from one or more sensors 502-506, which may be the same sensor, the same type of sensor, or a different sensor(s) type.

[0124] The processing unit 500 also a second digital neural signature generator 530 configured to generate a second digital neural signature from the second set of biomarkers; a second digital neural fingerprint generator 532 configured to generate a second digital neural fingerprint from the second digital neural signature, the second digital neural fingerprint being a comparison of biomarkers in the second digital neural signature to a threshold value associated with a second set of biomarkers; and a data pattern identifier 540 configured to identify any data patterns in the first and second digital neuro-fingerprints.

[0125] The processing unit 500 is configured to utilize any identified patterns in the first and second digital fingerprints to identify a second medical condition in the first digital neurological fingerprint.

[0126] The processing unit 500 may also include a prototypical first digital neuro-fingerprint generator 520 configured to generate a prototypical first digital neuro-fingerprint from a set of first digital neuro-fingerprints obtained from a set of subjects, and a prototypical second digital neuro-fingerprint generator 522 configured to generate a prototypical second digital neuro-fingerprint from a set of second digital neuro-fingerprints. In such an embodiment, the processing unit 500 is configured to identify some data pattern from the first and second prototypical digital neuro-fingerprints.

[0127] In some embodiments, the processing unit 500 includes a new first digital neuro-fingerprint generator 550 configured to generate a new first digital neuro-fingerprint from any data pattern matches determined in the identification step, the new first digital neuro-fingerprint being indicative of the second medical condition of the subject.

[0128] As mentioned above, the first and second digital neuro-fingerprint generators 516, 532 are preferably configured to form the first and second digital neuro-fingerprints as an array of elements, each of which is associated with a biomarker, in which each element has a value representative of the deviation of the measured biomarker from a threshold, preferably as a pixel of a display. To this end, the processing unit 500 may include an optical pattern recognition unit 560.

[0129] Processing unit 500 may also include a combined digital neural fingerprint generator 570 configured to generate a combined digital neural fingerprint from the first and second digital neural fingerprints. To this end, processing unit 500 is configured to utilize the combined digital neural fingerprint in determining the first and / or second medical condition of the subject.

[0130] The teachings herein allow for the generation of new digital neuro-fingerprints based on the preferred algorithms as described above, or based on combinations of existing biomarkers and / or new sets of biomarkers in an attempt to identify new disease or illness signatures, or new signatures of previously identified diseases or illnesses, or a combination of both, within a group of subjects. This can allow for the identification of subjects potentially susceptible to a disease or illness from a core set of biomarkers not previously identified or considered. Identification of such subjects can significantly reduce the number of subjects required for clinical trials, and can also potentially aid in earlier diagnosis of a disease or illness, well before previously considered symptoms appear in the patient himself.

[0131] Early diagnosis of cognitive impairment potentially leading to Alzheimer's disease could provide for earlier treatment and potentially significantly improved medical outcomes compared to existing methods and treatments.

[0132] Medications that may be prescribed to slow or prevent further deterioration or to treat symptoms include cholinesterase inhibitors (such as donepezil, rivastigmine, galantamine), memantine (optionally in combination with a cholinesterase inhibitor), monoclonal antibodies (such as aducanumab (Aduherm)), BAN2401, gantenerumab (optionally in combination with solanezumab), solanezumab (optionally in combination with gantenerumab), sigma-1 receptor agonists (optionally also ANAVEX2 (bulcamesin), AVP-786 or AXS -05), SV2A modulators (such as AGB101 (low dose levetiracetam)), mast cell stabilizers (such as ALZT-OP1 (cromolyn + ibuprofen)), anti-inflammatory agents (such as ALZT-OP1 (cromolyn + ibuprofen)), RAGE antagonists (such as azeliragon), glutamate modulators (such as BHV4157 (troriluzole)), D2 receptor partial agonists (such as brexpiprazole), serotonin and dopamine modulators (such as brexpiprazole), amyloid vaccines (such as CAD106), bacterial protease inhibitors (such as COR388), selective serotonin reuptake inhibitors (such as escitalopram), antioxidants (such as ginkgo), plant extracts (such as ginkgo), alpha-2 adrenergic agonists (such as guanfacine), omega-3 fatty acids (such as ethyl eicosapentaenoate (IPE), a purified form of eicosapentaenoic acid), angiotensin II receptor blockers (such as losartan), calcium channel blockers (such as amiodipine). , cholesterol agents (such as atorvastatin), combinations of angiotensin II receptor antagonists (such as losartan) and calcium channel blockers (such as amiodipine) and cholesterol agents (such as atorvastatin) with or without exercise, tyrosine kinase inhibitors (such as masitinib), insulin sensitizers (such as metformin), dopamine reuptake inhibitors (such as methylphenidate), alpha-1 antagonists (such as mirtazapine), acetylcholinesterase inhibitors (such as octohydroaminoacridine succinate),These may include ketone stimulants (such as tricaprylin), caprylic triglyceride (such as tricaprylin), tau protein aggregation inhibitors (such as AADvac1 or TRx0237 (LMTX)), positive allosteric modulators of the GABA-A receptor (zolpidem and zoplicone), or BPDO-1603, or any combination thereof administered together or separately.

[0133] Depending on the results, the system may suggest pharmaceutical intervention, modify an already implemented pharmaceutical intervention (e.g., modify dosage or administration regimen), and / or indicate whether treatment should be continued as effective.

[0134] The system also offers the physician the possibility to examine the scores obtained in the individual areas of the test in order to determine the optimal treatment for the individual.

[0135] Thus, the system can be used to diagnose individuals with mild cognitive impairment or AD, or to predict whether an individual with mild cognitive impairment will convert to AD in time. It can also be used to assist physicians in prescribing appropriate interventions and / or to help determine whether an already prescribed intervention is working. Thus, the system may assist physicians by suggesting to start an intervention, to stop an intervention, or to change an intervention, medication, or the like. It may suggest the appropriate frequency and / or dosage of a pharmaceutical intervention or a particular drug to be administered to an individual, and / or may suggest the appropriate route of administration of the pharmaceutical intervention for that individual. This applies to the specific pharmaceutical interventions described above (e.g., in Example 4), and to all other potential formulations, whether or not disclosed herein.

[0136] One notable advantage of the system described herein is that it can assess cognitive ability in a single test, compared to standard neurophysiological assessments currently used in diagnosing AD. As a result, cognitive function measurements can be administered in about 10 minutes, compared to 2 hours for traditional neurophysiological assessments (e.g., MMSE, ADAS-Cog).

[0137] The disclosures of US patent application Ser. No. 63 / 277,456, from which this application claims priority, and in the Abstract accompanying this application, are hereby incorporated by reference.

Claims

1. 1. A method for determining a medical condition in an individual, comprising: acquiring a first set of digital biomarkers from one or more sensors associated with the subject; generating a first digital neural fingerprint from the first set of digital biomarkers, the first digital neural fingerprint being a comparison of the first set of biomarkers with a threshold value associated with the first set of biomarkers; identifying a first medical condition in the subject from the first digital neural fingerprint; acquiring a second set of biomarkers from one or more sensors and generating a second digital neural fingerprint therefrom, the second digital neural fingerprint being a comparison of the second set of biomarkers with a threshold value associated with the second set of biomarkers; identifying a second medical condition in the subject from the second digital neural fingerprint, the second medical condition being different from the first medical condition in the subject; identifying any data patterns in the first and second digital neural fingerprints; and identifying a signature of the second medical condition in the first digital neuro-fingerprint using data patterns identified in the first and second digital neuro-fingerprints.

2. 10. The method of claim 1, wherein the identified data patterns in the first and second digital neuro-fingerprints are used to enhance the determination of the first medical condition in the first digital neuro-fingerprint.

3. The method of claim 1 , wherein the threshold value represents a mean value of that biomarker in a group of subjects believed to have a given condition.

4. The method of claim 3 , wherein the predetermined state is a healthy state.

5. generating a new first digital neural fingerprint from any data pattern matches determined in the identifying step; 10. The method of claim 1, wherein the new first digital neural fingerprint is indicative of the second medical condition of the subject.

6. 10. The method of claim 1, wherein the digital neural fingerprint is formed from a plurality of elements, each element associated with a biomarker, and each element has a value indicative of deviation of a measured biomarker from the threshold value.

7. The method of claim 1 , wherein each value of the digital neural fingerprint is within a given range.

8. The method of claim 7 , wherein the value of each element is an optical value.

9. The method of claim 8 , wherein the optical value is a color that varies depending on the deviation of the measured biomarker from the associated threshold value.

10. 10. The method of claim 9, comprising identifying patterns in the first and second digital neuro-fingerprints by optical pattern recognition.

11. The method of claim 1 , wherein the first and second sets of biomarkers are obtained from different sensors or different sets of sensors.

12. generating a combined digital neural fingerprint from the first and second digital neural fingerprints; 10. The method of claim 1, comprising using the combined digital neural fingerprint in determining a first and / or second medical condition of a subject.

13. generating a new digital neural fingerprint from the first and second digital neural fingerprints; 10. The method of claim 1, comprising using the new digital neural fingerprint in determining a first and / or second medical condition of a subject.

14. 10. The method of claim 1, wherein the first and / or second sets of biomarkers are obtained from one or more of a smartphone, a tablet computer, a smartwatch, a smart bracelet, a pair of smart glasses, and a camera.

15. The method of claim 1, wherein the digital neural fingerprint is formed from a plurality of elements, each associated with a biomarker, each element having a value indicating the deviation of the measured biomarker from the threshold, and a match is identified when a number of elements of the first and second digital neural fingerprints exceed a set percentage.

16. 10. The method of claim 1, wherein a match is identified if the number of elements in the first and second digital neuro-fingerprints exceeds 80% of the total number of elements in at least one digital neuro-fingerprint.

17. The method of claim 1 , wherein the step of identifying a pattern generates and uses a Shapley value.

18. 1. A system for determining a medical condition of an individual, comprising: one or more sensors associated with the subject, operable to acquire one or more biomarkers of the subject; a processing unit for processing the biomarkers, the processing unit comprising: a first register containing a first set of digital biomarkers from the subject obtained from one or more sensors associated with the subject; a first digital neural fingerprint generator configured to generate a first digital neural fingerprint, the first digital neural fingerprint being a comparison of the biomarkers to a threshold value associated with the first set of biomarkers; an identification unit configured to identify a first pathology of the subject from the first digital neural fingerprint; an input configured to receive a second set of biomarkers from one or more sensors; a second digital neural fingerprint generator configured to generate a second digital neural fingerprint, the second digital neural fingerprint being a comparison of the biomarkers to a threshold value associated with the second set of biomarkers; a data pattern identifier configured to identify any data patterns in the first and second digital neural fingerprints; the processing unit is configured to use any identified patterns in the first and second digital neuro-fingerprints to identify the second medical condition in the first digital neuro-fingerprint.

Images