System for generating simulated animal data and models - Patents.com

Abstract

A method for generating and distributing simulated animal data includes receiving a set of real animal data derived at least in part from one or more sensors that receive, store, and transmit information about one or more target individuals. The simulated animal data is generated from at least a portion of the real animal data or one or more derivatives thereof. Finally, the simulated animal data is provided to a computing device. Characteristically, one or more parameters or variables of the one or more target individuals may be modified.

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Application No. 62 / 897,064, filed September 6, 2019, and U.S. Provisional Application No. 63 / 027,491, filed May 20, 2020, the disclosures of which are incorporated by reference in their entireties herein.

[0002] In at least one aspect, the present invention relates to a system and method for generating simulated animal data from actual animal data. [Background technology]

[0003]

[0003] The continuing advances in the availability of information via the Internet have significantly changed the way business is conducted. Concurrent with this information explosion, sensor technology, and in particular biosensor technology, has also advanced. In particular, miniature biosensors that measure electrocardiogram signals, blood flow, body temperature, sweat levels, and respiratory rate are now available. The ability to transmit data from such sensors wirelessly and via the Internet has opened up potential new uses for data set collection.

[0004]

[0004] With the advancement of sensor technology, new animal data sets are being created. However, users who require animal data sets that characterize specific characteristics of target objects, sensors, activities, conditions, and other variables or parameters may face obstacles in collecting the data, as data acquisition can be costly and time-consuming, making collection difficult. In many cases, the data sets do not exist. At the same time, there is an increasing demand for such target animal data sets in fields such as healthcare, insurance, wellness monitoring, fitness, virtual sports, gaming, sports betting, etc., as the data can be used in various simulations and models to engage users and evaluate outcomes related to one or more future events. No system or method exists that provides the desired animal data sets for incorporation into such simulations.

[0005]

[0005] Therefore, there is a need to create artificial data from real animal data that can be customized and adjusted based on user preferences. Summary of the Invention

[0006]

[0006] In at least one aspect, the present invention provides a method for generating and distributing simulated animal data. The method includes receiving one or more sets of real animal data, at least partially acquired from one or more sensors that receive, store or transmit information about one or more target individuals. The simulated animal data is generated from at least a portion of the real animal data or one or more derivatives thereof. Finally, the simulated animal data is provided to a computing device. Characteristically, one or more parameters or variables of the one or more target individuals may be altered.

[0007]

[0007] In another aspect, a system is provided for generating and providing simulated animal data by performing the methods herein. The system, including a computing device, is operable to perform the steps of: receiving one or more sets of real animal data, at least partially obtained from one or more sensors that receive, store or transmit information about one or more target individuals; generating simulated animal data from at least a portion of the real animal data or one or more derivatives thereof; and providing at least a portion of the simulated animal data to the computing device. Featured, one or more parameters or variables of the one or more target individuals may be modified.

[0008]

[0008] In another aspect, simulated animal data derived from actual animal data obtained at least in part from one or more sensors is used to create, enhance or modify one or more insights, calculated assets or predictive indicators.

[0009]

[0009] In another aspect, at least a portion of the simulated animal data is used in one or more simulation systems to engage one or more users, the simulation systems being at least one of a game-based system, an augmented reality system, a virtual reality system, a mixed reality system, or an extended reality system.

[0010] In another aspect, simulated animal data derived at least in part from the actual animal data obtained from one or more sensors is used as one or more inputs in one or more further simulations to generate simulated data, at least a portion of which is used to create, modify or enhance one or more insights, calculated assets or predictive indicators.

[0011] In another aspect, simulated animal data derived from real animal data acquired at least in part from one or more sensors is used as one or more inputs in one or more further simulations to generate simulation data, at least a portion of which is used to engage a user in a simulation system, the simulation system being at least one of a game-based system, an augmented reality system, a virtual reality system, a mixed reality system, or an extended reality system.

[0012]

[0012] In another aspect, the simulated data derived from actual animal data obtained at least in part from one or more sensors is used, directly or indirectly: (1) as one or more markets in which wagers are made or accepted; (2) to create, modify, enhance, acquire, offer or distribute one or more products; (3) to evaluate, calculate, derive, modify, enhance or communicate one or more predictions, probabilities or possibilities; (4) to formulate one or more strategies; (5) to execute one or more actions; (6) to mitigate or prevent one or more risks; (7) to recommend one or more actions; (8) as one or more signals or measurements utilized in one or more simulations, calculations or analyses; (9) as part of one or more simulations, the output of which directly or indirectly involves one or more users; (10) as one or more core components or supplements to one or more consumption media; (11) in one or more promotions; or (12) any combination thereof.

[0013]

[0013] In another aspect, the simulation data derived from actual animal data obtained at least in part from one or more sensors is used directly or indirectly in one or more sports betting, insurance, health, fitness, biological performance or entertainment applications.

[0014] In another aspect, artificial data is generated to replace one or more outliers or missing values ​​generated from one or more sensors. [Brief description of the drawings]

[0015] For a further understanding of the nature, objects and advantages of the present disclosure, reference should be had to the following detailed description read in conjunction with the following drawings, in which like reference numerals indicate like elements and in which:

[0016] [Figure 1] FIG. 1 is a schematic diagram of a system for generating simulated animal data from actual animal data. [Diagram 2]

[0017] 1 is a plot of collected heart rate data and a polynomial fit to the data. [Figure 3A]

[0018] 1 shows a graph of beats per minute (BPM) values ​​captured from a target subject in an athletic event. [Figure 3B]

[0019] 3B shows the autocorrelation function of the data in FIG. 3A. [Figure 4]

[0020] 1 shows an overview of a neural network that may be used in a neural network architecture for generating simulation data. [Diagram 5]

[0021] 1 shows details of a recurrent neural network that may be used to generate simulated animal data. [Figure 6]

[0022] FIG. 1 shows a schematic diagram of a long short-term memory (LSTM) network that may be used to generate simulated animal data. [Figure 7A]

[0023] 1 shows a plot of artificial heart rate data generated from in-sample actual animal data that occurs when predicting observations that include at least a portion of the animal data sample. [Figure 7B]

[0024] 7B shows an example of artificial heart rate data generated from out-of-sample real animal data corresponding to FIG. 7A. [Figure 8]

[0025] We show how to generate simulated animal data from generative adversarial networks (GANs). [Figure 9]

[0026] 13 is a plot of artificial heart rate data generated from real animal data using a generative adversarial network. [Figure 10]

[0027] 1 illustrates an example of a gaming system (eg, a video gaming system) in which a user can purchase derivatives of animal data. [Figure 11]

[0028] 13 illustrates out-of-sample testing of raw data measurements from a biosensor with predicted raw data values ​​generated using an LSTM neural network. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017]

[0029] Reference will now be made in detail to the presently preferred configurations, embodiments and methods of the invention, which constitute the best modes for carrying out the invention currently known to the inventors. The drawings are not necessarily to scale. It should be understood, however, that the disclosed embodiments are merely exemplary of the invention, which may be embodied in various alternative forms. Thus, the specific details disclosed herein should not be construed as limiting, but merely as a representative basis for any aspect of the invention and / or for teaching one of ordinary skill in the art how to employ the invention in various ways.

[0018]

[0030] Of course, it is to be understood that the invention is not limited to the specific embodiments and methods described below, as specific components and / or conditions may vary. Further, the terminology used herein is used only for the purpose of describing particular embodiments of the invention, and is not intended to be limiting in any way.

[0019]

[0031] It should also be noted that, as used in the specification and the appended claims, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. For example, reference to an element in the singular is intended to include the plural elements.

[0020]

[0032] The phrase "data is" is meant to include both "datum is" and "data are," as well as all other possible meanings, and is not intended to be limiting in any way.

[0021]

[0033] The term "comprising" is synonymous with "including," "having," "containing," or "characterized by." These terms are inclusive and open-ended and do not exclude additional, unrecited elements or method steps.

[0022]

[0034] The phrase "consisting of" excludes any element, step, or ingredient not specified in the claim. When this phrase appears in a clause in the body of a claim rather than immediately following the preamble, it limits only the elements recited in that clause; other elements are not excluded from the claim as a whole.

[0023]

[0035] The phrase "consisting essentially of" limits the scope of a claim to the addition of the specified materials or steps that do not materially affect the basic and novel characteristics of the claimed subject matter.

[0024]

[0036] With respect to the terms "comprising," "consisting of," and "consisting essentially of," when one of these three terms is used in this specification, the presently disclosed and claimed subject matter may include use of either of the other two terms.

[0025]

[0037] The term "one or more" means "at least one," and the term "at least one" means "one or more." The terms "one or more" and "at least one" include the subsets "plurality" and "multiple." In particular, "one or more" includes "two or more."

[0026]

[0038] Throughout this application, when publications are referenced, the disclosures of these publications in their entireties are hereby incorporated by reference into this application in order to more fully describe the state of the art to which this invention pertains.

[0027]

[0039] Although the terms "probability" and "odds" are mathematically distinct (e.g., probability may be defined as the number of occurrences of a particular event expressed as a proportion of all events that can occur, while odds may be defined as the number of occurrences of a particular event expressed as a proportion of the number of non-occurrences of that event), both express the likelihood of an event occurring. They are used interchangeably to avoid redundancy, and reference to one term should be interpreted as meaning reference to both.

[0028]

[0040] With respect to the terms "bet" and "wager," both terms refer to the act of taking a risk (e.g., money, non-monetary consideration) on the outcome of a future event. Risk includes both financial risks (e.g., monetary) and non-financial risks (e.g., health, life). Risks may be taken against another party or parties (e.g., an insurance company determining whether to provide insurance) or against oneself (e.g., an individual determining whether to obtain insurance) based on the outcome or the likelihood of the outcome of a future event. Examples include gambling (e.g., sports betting), insurance, and the like. Where one of these two terms is used herein, the presently disclosed and claimed subject matter may interchangeably use either of the other two terms.

[0029]

[0041] The term "server" refers to any computer or computing device (including but not limited to a desktop computer, a notebook computer, a laptop computer, a mainframe, a mobile phone, a smart watch / glasses, an augmented reality headset, a virtual reality headset, etc.), a distributed system, a blade, a gateway, a switch, a processing device, or combination thereof adapted to perform the methods and functions described herein.

[0030]

[0042] When computing devices are described as performing actions or method steps, it is understood that one or more computing devices are typically operable to perform the actions or method steps by executing one or more lines of source code. The actions or method steps may be encoded in non-transitory memory (e.g., a hard drive, optical drive, flash drive, etc.).

[0031]

[0043] The term "computing device" generally refers to any device capable of performing at least one function, including communicating with another computing device. In particular, a computing device includes a central processing unit capable of executing program steps and a memory for storing data and program code.

[0032]

[0044] The term "electronic communication" means that an electrical signal is transmitted directly or indirectly from an originating electronic device to a receiving electronic device. Indirect electronic communication may include processing of an electrical signal, including but not limited to filtering the signal, amplifying the signal, modifying the signal, modulating the signal, attenuating the signal, adding one signal to another, subtracting one signal from another, etc. Electronic communication may be accomplished using wired components, wireless connection components, or a combination thereof.

[0033]

[0045] The processes, methods or algorithms disclosed herein may be deliverable / implemented by a computer, controller, or other computing device, which may include existing programmable electronic control units or dedicated electronic control units. Similarly, the processes, methods or algorithms may be stored as data and instructions executable by a computer, controller, or other computing device in many forms, including, but not limited to, information persistently stored on non-writable storage media, such as ROM devices, and information mutably stored on writable storage media, such as floppy disks, magnetic tapes, CDs, RAM devices, other magnetic and optical media, shared or dedicated cloud computing resources. The processes, methods or algorithms may be implemented in executable software objects. Alternatively, the processes, methods or algorithms may be embodied in whole or in part using suitable hardware components, such as application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), state machines, controllers, other hardware components or devices, or a combination of hardware, software and firmware components.

[0034]

[0046] The terms "subject" and "individual" are synonymous and refer to humans or other animals, including birds, reptiles, amphibians, and fish, as well as all mammals, including primates (particularly higher primates), horses, sheep, dogs, rodents, pigs, cats, rabbits, and cows. One or more subjects may be, for example, humans participating in athletic training or competitions, horses racing at a racetrack, humans playing video games, humans monitoring personal health, humans providing data to a third party, humans participating in a research or clinical study, or humans attending a fitness class. A subject or individual may also be a derivative of a human or other animal (e.g., a laboratory-grown organism derived at least in part from a human or other animal), one or more individual components, elements, or processes of a human or another animal that make up a human or other animal (e.g., cells, proteins, biological fluids, amino acid sequences, tissues, hair, limbs), one or more digital representations that share at least one characteristic with a human or animal (e.g., a dataset representing a human that shares at least one characteristic with a representation of a human in digital form, such as, for example, gender, age, biological function, but is not generated from a human existing in the physical world; a simulated individual), or one or more artificial products that share one or more characteristics with a human or other animal (e.g., a laboratory-grown human brain cell that produces electrical signals similar to those of a human brain cell). In particular, a subject or individual may be one or more programmable computing devices, such as machines (e.g., robots, autonomous vehicles, mechanical arms) or networks of machines that share at least one biological function with humans or other animals, any of which may derive one or more types of biological data, which may be, at least in part, artificial in nature (e.g., data from artificial intelligence derived activities that mimic biological brain activity, biomechanical movement data from programmable machines).

[0035]

[0047] The term "animal data" refers to any data obtainable from a subject or generated directly or indirectly by a subject that can be converted into a format that can be transmitted to a server or other computing device. Typically, animal data is transmitted electronically over a wired or wireless connection. Animal data includes data from a subject, including any signals or measurements obtained from one or more sensors or sensing devices / systems, and in particular, biosensors. Animal data may include descriptive data about the subject, auditory data about the subject, visual capture data about the subject, neurologically generated data (e.g., brain signals from neurons), assessment data about the subject (e.g., skills of the subject), manually inputtable data about the subject (e.g., medical history, social habits, emotions of the subject), data that includes at least a portion of actual animal data or one or more derivatives thereof, and the like. In particular, the term "animal data" includes any derivative of animal data. Even more particularly, animal data includes any metadata collected by or associated with the animal data. Even more particularly, animal data includes at least a portion of simulated data. In yet another particular, animal data includes simulated data.

[0036]

[0048] In some variations, the term "actual animal data" is used interchangeably with the term "animal data." In other variations, the term "actual animal data" refers to animal data that is obtained at least in part from one or more sensors that receive, store and / or transmit information about one or more target individuals or groups of target individuals.

[0037]

[0049] The term "artificial data" refers to artificially created data derived from, based on, or generated at least in part using actual animal data or one or more derivatives thereof. Artificial data may be created by running one or more simulations utilizing one or more artificial intelligence techniques or statistical models, and may include one or more signals or measurements from one or more non-animal data sources as one or more inputs. Artificial data may include any artificially created data that shares at least one biological function with a human or another animal (e.g., artificially created visual data, artificially created locomotion data). Artificial data includes "synthetic data," which may be any produced data applicable to a particular situation that is not obtained by direct measurement. Synthetic data may be created by statistically modeling original data and then using those models to generate new data values ​​that reproduce at least one of the statistical properties of the original data. In particular, the term "artificial data" includes any derivative of artificial data. For purposes of the presently disclosed and claimed subject matter, the terms "simulated data" and "synthetic data" are synonymous and are used interchangeably with "artificial data," and reference to any one of these terms should not be construed as limiting, but rather as encompassing all possible meanings of all terms. In particular, the term "artificial data" includes the term "artificial animal data."

[0038]

[0050] The term "insight" refers to one or more descriptions assignable to a target individual that utilize at least a portion of the animal data to describe the target individual's situation or condition. Examples include descriptions or other characterizations of stress level (e.g., high stress, low stress), energy level, fatigue level, etc. Insights may be quantified by one or more numbers or numbers and may be expressed as a probability or similar likelihood-based indicator. Insights may be quantified, communicated, or characterized by one or other predetermined metrics or performance indices (e.g., codes, graphs, charts, plots, colors or other visual representations, plots, measurements, numerical representations, descriptions, text; physical responses such as vibrations, auditory responses, visual responses, kinesthetic responses, or verbal descriptions). Insights may also include one or more visual representations of one or more target subject's situation or condition (e.g., an avatar or realistic depiction of a target subject visualizing future weight loss goals for the avatar or depiction of the target subject). In particular, an insight is a personalized score or other index for one or more target individuals or groups of target individuals that utilizes at least a portion of the simulation data to (1) evaluate, assess, prevent or mitigate risk based on the animal data, (2) evaluate, assess and optimize performance (e.g., biological performance) based on the animal data, or a combination thereof. The personalized index score may be utilized by one or more target subjects from whom the animal data or one or more derivatives thereof are derived, and by one or more third parties (e.g., insurance organizations, health care providers or experts, sports performance coaches, medical billing agencies, fitness trainers, etc.). More particularly, an insight is derived from two or more types of animal data. Still more particularly, an insight includes one or more signals or measurements from one or more non-animal data sources as one or more inputs in one or more calculations, computations, derivations, incorporation, simulations, extractions, extrapolations, modifications, extensions, enhancements, creation, estimations, inferences, predictions, determinations, processing, communications, etc. More particularly, an insight is comprised of a plurality of insights. More specifically, insights are assigned to a number of target individuals and to one or more groups of target individuals.

[0039]

[0051] The term "computed asset" refers to one or more numbers, values, metrics, measurements, insights, graphs, charts, or plots derived from at least a portion of the animal data or one or more derivatives thereof (which may include simulated data). As used herein, one or more sensors initially provide electronic signals. A computed asset is extracted or derived, at least in part, from the one or more electronic signals or one or more derivatives thereof. A computed asset describes or quantifies an interpretable characteristic of one or more target individuals or groups of target individuals. For example, electrocardiogram measurements may be derived from an analog front-end signal (e.g., an electronic signal from a sensor), heart rate data (e.g., beats per minute) may be derived from an electrocardiogram or PPG sensor, body temperature data may be derived from a temperature sensor, sweat data may be derived or extracted from a sweat sensor, glucose information may be derived from a biofluid sensor, DNA and RNA sequencing information may be derived from sensors that obtain genomic and genetic data, brain activity data may be derived from a neural sensor, hydration data may be derived from an oral saliva or sweat analysis sensor, location data may be derived from a GPS or RFID-based sensor, biomechanical data may be derived from an optical or translational sensor, and respiration rate data may be derived from a respiration sensor. In particular, computed assets include one or more signals or measurements from one or more non-animal data sources as one or more inputs in one or more calculations, computations, derivations, incorporation, simulations, extractions, extrapolations, modifications, enhancements, creation, estimations, inferences, inferences, determinations, processing, communication, etc. More particularly, the calculated asset is derived from two or more types of animal data. Even more particularly, the calculated asset is made up of a plurality of calculated assets.

[0040]

[0052] The term "predictive indicator" refers to a metric or other indicator (e.g., one or more colors, codes, numbers, values, graphs, charts, plots, measurements, numerical representations, descriptions, text, physical responses, auditory responses, visual responses, kinesthetic responses) from which one or more forecasts, predictions, probabilities, assessments, possibilities, predictions or recommendations regarding one or more outcomes of one or more future events involving one or more target individuals or one or more groups of target individuals may be calculated, calculated, derived, extracted, extrapolated, simulated, created, modified, assigned, enhanced, estimated, evaluated, inferred, established, determined, transformed, inferred, observed, communicated or implemented. In particular, a predictive indicator is a calculated asset derived from at least a portion of the animal data or one or more derivatives thereof. More particularly, the predictive indicator includes one or more signals or measurements from one or more non-animal data sources as one or more calculations, calculations, derivations, extractions, extrapolations, simulations, creations, modifications, assignments, enhancements, estimates, evaluations, inferences, establishments, determinations, conversions, inferences, observations or communication of the one or more predictions, forecasts, probabilities, possibilities, evaluations, estimates or recommendations. More particularly, the predictive indicator includes at least a portion of the simulation data as one or more inputs in the calculations, calculations, derivations, extractions, extrapolations, simulations, creations, modifications, assignments, enhancements, estimates, evaluations, inferences, establishments, determinations, conversions, inferences, observations or communication of the one or more predictions, forecasts, probabilities, possibilities, evaluations, estimates or recommendations. More particularly, the predictive indicator is derived from two or more types of animal data. More particularly, the predictive indicator is comprised of a plurality of predictive indicators.

[0041]

[0053] With reference to FIG. 1, a computer-implemented method and system for generating simulation data is provided. The simulation system 10 includes a computing device 12 that receives animal data 14. Typically, methods and systems for generating such animal data 14 deploy one or more sensors 18 that collect real animal data from one or more target individuals 16. In some variations, the animal data represents data about the target individuals (e.g., their bodies) that is derived, at least in part, from one or more sensors 18, particularly biosensors. In many useful applications, the target individuals are humans (e.g., athletes, soldiers, medical patients, research subjects, fitness class participants, video gamers) and the animal data is human data. The animal data may be derived from a target individual or multiple target individuals (e.g., including a target group of multiple target individuals, multiple target groups of multiple target individuals). The animal data may be obtained from a single sensor for each target individual or from multiple sensors for each target individual. In some cases, a single sensor may capture data from multiple target individuals, a target group of multiple target individuals, or multiple target groups of multiple target individuals (e.g., an optical-based camera sensor that can identify and measure distances traveled by a target group of target individuals). Each source sensor may provide a single type of animal data or multiple types of animal data. In one variation, the sensor 18 may include multiple sensing elements for measuring one or more parameters (e.g., heart rate and accelerometer data) within a single sensor. In particular, the one or more sensors 18 include at least one biological sensor. The one or more sensors 18 may collect data from target individuals engaged in a variety of activities, including vigorous activities that may change one or more biological signals or measurements of the target individuals, such as blood pressure, heart rate, or biological fluid levels. The activities may also include sedentary activities, such as sleeping or sitting, that have less variability in changes in biological signals or measurements. In one variation, the simulation system 10 may also receive (e.g., collect) animal data not obtained from a sensor (e.g., manually entered animal data; sensor-collected animal data sets that include artificial data values ​​not generated from a sensor).

[0042]

[0054] Continuing to refer to FIG. 1 , the one or more sensors 18 may transmit the animal data 14 to the computing device 12 either directly or wirelessly via a cloud 22 or via a wired connection 24. The cloud 22 may be the Internet, a public cloud, a private cloud, or a hybrid cloud. In particular, the computing device 12 communicates with the one or more sensors 18 through a local server (e.g., localized or networked server / storage, a localized storage device, a distributed network of computing devices) or other computing device 19 that intervenes in the transmission of the animal data 14 to the computing device 12 (e.g., collects and transmits data to the computing device 12, or collects and transmits data to a cloud accessible by the computing device 12). For example, the intermediate computing device may be a smartphone or other computing device. The animal data input to the system may be raw data obtained from the one or more sensors or transformed (e.g., manipulated, processed) data. In particular, transformed data includes data that has been cleaned, edited, modified, and / or manipulated in one or more ways (e.g., data with metadata attached, data transformed into one or more measurements related to heart rate, blood pressure, sweat rate, etc.). More particularly, the act of transforming data includes one or more of: calculating, computing, deriving, incorporating, simulating, extracting, adding, subtracting, extrapolating, modifying, enhancing, creating, estimating, inferring, guessing, determining, converting, processing, communicating, etc. For example, in connection with measuring heart rate, a biosensor may be configured to measure electrical signals from a target subject's body, transform (e.g., convert) the analog-based measurements into digital measurements, and transmit the digital measurements. In another example, a computing device may receive digital measurements from a sensor and convert the digital measurements into one or more heart rate values.Further details regarding systems for measuring heart rate and other biological data are disclosed in U.S. Patent Application No. 16 / 246,923, filed January 14, 2019, and International Application No. PCT / US20 / 13461, filed January 14, 2020, the disclosures of which are incorporated herein by reference in their entireties. More specifically, the act of transforming data includes one or more actions of normalizing, time stamping, aggregating, tagging, storing, manipulating, denoising, enhancing, organizing, visualizing, analyzing, anonymizing, synthesizing, summarizing, replicating, productizing, or synchronizing animal data. More specifically, one or more transformations occur by utilizing (e.g., incorporating) one or more signals or measurements from non-animal data.

[0043]

[0055] With further reference to FIG. 1 , the computing device 12 utilizes at least a portion of the actual animal data or one or more derivatives thereof and executes the steps of a simulation program with the data converted into a format to be input into the simulation, or transmits the data to another one or more computing devices 30 (e.g., a computing device associated with or within the network of the computing device 12, or a third party computing device) to execute the simulation. In this regard, the computing device 12 and the one or more computing devices 30 may be operable to execute a simulation. The executed simulation involves one or more simulated target individuals, and one or more parameters or variables of the simulated target individuals may be altered, randomized, and / or modified. In one variation, the one or more parameters or variables of the one or more target individuals may include any inputs related to or relating to the one or more target individuals (including both internal and external characteristics of the one or more target individuals) and inputs that affect (e.g., affect, change, modify, adjust) or may affect one or more outputs of the one or more simulations based on their inclusion in the simulation. In particular, the one or more parameters or variables modified to generate the simulation data are comprised of non-animal data. In one form of simulation, the simulation provides a medium for user engagement with one or more inputs and outputs that are limited to a computing device. In these cases, the simulation may be integrated with other components (e.g., hardware, software) that interact with one or more users.For example, a simulation system that performs a simulation and incorporates at least a portion of the actual animal data or one or more derivatives thereof may be a game-based system (e.g., a video game system, a virtual gambling system, a fitness game system, etc.), an augmented reality system, a virtual reality system, a mixed reality system, an extended reality system, or other form of interactive simulation. In another form of simulation, a simulation is a method for implementing a model over a period of time to predict one or more future events. The simulation data may be derived from one or more simulation events, concepts, objects, or systems. The simulation data may be generated using one or more statistical models or artificial intelligence techniques. Characteristically, multiple simulations may occur utilizing the same one or more inputs, and a simulation may consist of multiple simulations. In particular, multiple simulation systems may be operable to work together. For example, the simulation data may be generated by a computing device and provided to another computing device that operates a simulation program into which the simulation data is input. In more detail, one or more simulations may include one or more data sets from non-animal data as one or more inputs.

[0044]

[0056] Upon execution of the simulation program by computing device 12 and / or one or more computing devices 30, simulation data 28 is generated and provided to one or more computing devices. Characteristically, the generated simulation data may be artificial animal data (e.g., artificial heart rate data, artificial respiratory rate data, artificial glucose data, etc.). For example, the simulated animal data may indicate a fatigue level of a simulated target individual at any given time in a simulated sporting event using one or more variables or parameters (e.g., distance run, environmental data) adjusted within the simulation, one or more of which may be signals or measurements from non-animal data (e.g., time). As another example, the simulated animal data, such as a measurement of a simulated heart rate, may represent a future biological activity of a simulated target individual within a simulated sporting event. Advantageously, such information may be utilized as part of one or more predictions, probabilities, or likelihoods regarding the simulated animal data. As another example, the simulated animal data may indicate or predict how one or more simulated target subjects will respond to a particular drug in a simulated pharmaceutical study using one or more drugs and one or more characteristics of one or more target individuals that are one or more variables in the simulation. In many useful variations, one or more simulation target objects of a simulation represent (e.g., resemble) one or more real-world target objects or groups of target objects that share one or more biological and / or non-biological characteristics associated with one or more real-world target objects or groups of target objects, thereby allowing one or more simulation target objects or groups of target objects to represent one or more real-world target objects or groups of target objects in the simulation. Simulation data can include actual animal data (e.g., actual heart rate data of a subject incorporated into a simulation system such as a video game system) that has been converted into a format for input into the simulation.In particular, at least a portion of the simulation data may be used to create, enhance or modify one or more insights, calculated assets or predictive indicators.

[0045]

[0057] In particular, at least a portion of the simulated animal data 28 or one or more derivatives thereof are used as one or more inputs in one or more further simulations. The one or more further simulations may be tailored to utilize the previously generated simulated animal data to predict one or more future events. For example, the simulated animal data 28 may be used in a sporting event simulation to predict one or more outcomes (e.g., by having a generated artificial "fatigue level" of a target subject for an event such as a professional sports match, one or more outcomes--win / lose, whether the target subject will experience a biological event such as heat stroke, etc.) The simulated animal data 28 may be used in one or more further simulations to simulate other animal data (e.g., a subject's simulated heart rate data may be used as an input to generate another simulated biological output such as simulated hydration or glucose information). A variety of simulated biological functions and activities may benefit from generating and incorporating simulated animal data, including simulations of physical activities (e.g., sporting events, fitness activities), health monitoring (e.g., insurance, military, home monitoring / telemedicine applications), biological analysis (e.g., DNA sequencing), biological responses (e.g., cellular or biofluid responses to a particular type of drug), and the like. In particular, the simulations are based on one or more target individuals involved in at least one of a fitness activity, a sporting event (e.g., professional sports competition), a health evaluation (e.g., remote patient monitoring, in-hospital patient evaluation, general wellness platform providing feedback from one or more sensors), or an insurance evaluation (e.g., including receiving insurance quotes, obtaining insurance, adjusting premiums). More particularly, at least a portion of one or more simulated data sets may be used to create, modify, or enhance one or more insights, calculated assets, or predictive indicators. The simulated animal data 28 may be used within a simulation involving one or more users.In one variant, the simulated animal data 28 may be generated based on one or more animal data sets from a plurality of subjects representing one or more defined groups. For example, the system may generate simulated average heart rate data for a defined group of individuals characterizing a particular biological characteristic in a defined scene / situation environment (e.g., engaged in a particular activity for a particular period of time). The identity of one or more target subjects or target groups of target subjects may or may not be known. In another variant, the simulated data may be used as a baseline data set to represent a particular subject group (having one or more defined characteristics) in one or more further simulations. Advantageously, one or more simulations may be performed in real time or near real time with adjustment of one or more parameters or variables. Near real time in this context means that the transmission is not intentionally delayed, except for the necessary processing by the sensors and computing devices. In particular, the simulated data derived from at least a portion of the actual animal data or one or more derivatives thereof may be used, directly or indirectly: (1) as a market in which one or more wagers are made or accepted; (2) to create, modify, enhance, acquire, offer or distribute one or more products; (3) to evaluate, calculate, derive, modify, enhance or communicate one or more predictions, probabilities or possibilities; (4) to formulate one or more strategies; (5) to implement one or more actions; (6) to mitigate or prevent one or more risks; (7) to recommend one or more actions; (8) as one or more signals or measurements utilized in one or more simulations, calculations or analyses; (9) as part of one or more simulations, the output of which directly or indirectly engages one or more users; (10) as one or more components or supplements to one or more consumption media; (11) in one or more promotions; or (12) any combination thereof.

[0046]

[0058] In one variation on application (1), a market may be a particular type or category of bet or wager on a particular event (e.g., a sporting event, a health or medical event, a simulated event). Markets may be created and offered or utilized for any event. Often, one or more betting accepting organizations offer multiple betting markets on each event, with odds listed for each market. The particular type or category may include a proposition bet ("prop bet"), a spread bet, a line bet, a future bet, a parlay bet, a round robin bet, a handicap bet, an over / under bet, a full coverage bet, or a teaser bet. Additionally, the acceptance of a bet may be, for example, acceptance by an insurance system (e.g., an insurance company) of a bet utilizing one or more outputs (e.g., a bet type utilizing a predictive indicator derived from the simulation data), acceptance by an insurance system (e.g., an insurance company) of a payment from an individual that correlates with a risk the insurance company will take based on one or more outputs (e.g., an insurance policy offered to an individual that may or may not be bet by the company based on the likelihood that the individual will experience any given biological event predicted by a predictive indicator derived from the simulation data), acceptance by an insurance system of one or more treatments related to a particular diagnosis for a given individual - and the payments and timelines associated with the one or more treatments - recommended by a medical provider based on the simulated effectiveness of the treatments utilizing at least a portion of the individual's animal data and the generated simulation data, etc.

[0047]

[0059] In one variation on application (2), the one or more products may be one or more goods or services designed to be distributed or sold. The product may be any product in any industry or business that may be created, modified, enhanced, offered, or distributed, so long as the product uses, directly or indirectly, at least a portion of the simulation data. For example, the product may be a market in which one or more wagers are placed or accepted. In particular, at least a portion of the simulation data or one or more derivatives thereof are used to create, modify, enhance, offer, obtain, accept, or distribute at least one of a proposition bet, a spread bet, a line bet, a futures bet, a parlay bet, a round robin bet, a handicap bet, an over / under bet, a full cover bet, or a teaser bet. This includes the simulation data or one or more derivatives thereof that lead to (or result in) the creation of a product. For example, a product may be the simulation data itself (e.g., purchase of one or more outputs of a simulation), an insurance offer, a health application that displays one or more simulation outputs, a set of algorithms designed to provide specific simulation insights about a subject, a sports betting application, a consumer product that utilizes the simulation data (e.g., beverages, such as isotonic drinks, that utilize the simulation data to personalize ingredients based on the subject's biological information, foods, etc.). For clarity, "enhancement" may include "being part of" a product if the enhancement provides added value. Furthermore, in many cases, "creating" may include "deriving" and vice versa. Similarly, "creating" may include "generating" and vice versa. Furthermore, "modifying" may include "correcting," "compensating," "adjusting," "changing," and "improving." Furthermore, "proposing" may include "providing." Finally, an "acquirer" of a product may be, for example, a consumer, an organization, another system, other endpoints that can consume or receive the product, etc.

[0048]

[0060] In one variation on application (3), one or more predictions, probabilities, or possibilities may be associated with a future outcome or event by combining one or more predictions, probabilities, or possibilities. For example, a probability may be calculated to determine the likelihood that any given athlete's heart rate will exceed 200 bpm in any given basketball game using various types of data including the athlete's current heart rate, average heart rate, maximum heart rate, past heart rates under similar conditions, biological fluid levels, sEMG data, minutes on court, total distance run, simulated biological data, environmental data, other state / situational information, etc. Using this probability, another probability may be calculated to determine the likelihood that the athlete will make a shot outside n feet a greater than n% percentage of the time if the athlete's heart rate exceeds 200 bpm. Additionally, "communicating" may include visualization of one or more predictions, probabilities, or possibilities (e.g., displaying probabilities via an application, displaying output-based probabilities for the target individual in an augmented or virtual reality system), verbal communication of one or more predictions, probabilities, or possibilities (e.g., a voice-activated virtual assistant informing the target individual of the likelihood that an event may occur based on simulated biological data, or the likelihood that an event will occur. An example may be the likelihood of becoming hypoglycemic if a particular action is not performed, the likelihood of having a stroke in the next n days based on collected biological data, or the likelihood of a biologically related event occurring based on simulated data). Finally, modifying a prediction, probability, or possibility may include a correction of a previously determined prediction, probability, or possibility of an event.

[0049]

[0061] In one variation on application (4), the strategy may include any strategy that directly or indirectly uses at least a portion of the simulation data. For example, the strategy may be an action plan for determining whether to insure an individual, whether to place a bet, whether to take a particular action on the simulation data, etc. The strategy may include a complete trading / betting strategy that is based entirely on the simulation and the simulation data to predict potential outcomes and thresholds where predefined rules come into play. Furthermore, one or more simulation data outputs or one or more derivatives thereof may be utilized in one or more further calculations, calculations, derivations, extractions, extrapolations, simulations, creations, modifications, enhancements, estimates, evaluations, inferences, establishments, determinations, transformations, inferences, observations, or communications regarding the formulation of one or more strategies. In this regard, the term "formulation" may include one or more modifications, enhancements, etc.

[0050]

[0062] In one variation on application (5), the action may be any action that directly or indirectly relates to at least a portion of the simulation data, including an action derived from (or resulting from) the simulation data. For example, the action may be to verify or authenticate the health status of an individual, to insure an individual (e.g., the probability that the target subject will have a heart attack within the next 24 months is x, so the bet is y), to accept or reject a healthcare provider's personalized treatment plan for a medical event or need of a subject (e.g., based on one or more simulations, the probability that the healthcare provider's recommended treatment will cure the target subject is n, so the insurance company agrees to pay for w weeks of treatment at a price p based on the simulation data), an action regarding the biology of the target individual (e.g., a passenger in an autonomous vehicle may have biological measurements that trigger one or more simulations to occur via a computing device, the output of which may alert the autonomous vehicle to drive to the nearest hospital), to place a bet (e.g., an athlete's energy level derived from one or more simulations is x percent, so the user places a bet), to perform a particular action (e.g., a system communicating an action to perform a particular action such as "place a bet," "run for 20 minutes today," "eat n calories today," etc.), to take no action at all, etc.

[0051]

[0063] In one variation on application (6), mitigating or preventing a risk can include actions, inactions, strategies, recommendations, recategorization of risk, changes to a risk profile, etc. related to reducing or preventing the risk. It can also include taking additional risk.

[0052]

[0064] In one variation on application (7), recommending one or more actions includes both recommendations inferred directly or indirectly by the simulation data (e.g., a predictive indicator derived from the simulation data providing a probability of an event occurring may infer an action to be taken) and recommendations directly stated based on one or more outputs (e.g., a recommendation for an action to be taken based on a predictive indicator derived from one or more simulations providing a probability or prediction of an event occurring). In particular, a recommendation may consist of multiple recommendations.

[0053]

[0065] In one variation of application (8), the signals or measurements may be of any format and may include any type of information (eg, including as one or more data sets).

[0054]

[0066] In one variation on application (9), simulation includes both the generation of one or more computer models and the imitation of one or more situations or processes. Simulation has a wide range of engagement applications, including simulations that are utilized to generate one or more outputs, any use of the outputs may be considered direct or indirect engagement, and may include one or more outputs within one or more simulations (e.g., video games or other game-based systems, augmented reality or virtual reality systems) that may involve one or more users.

[0055]

[0067] In one variation on the application 10, the one or more media of user consumption may be any media that allows a user to directly or indirectly consume one or more outputs from one or more simulations. Media may include, for example, a health monitoring application (e.g., a remote monitoring platform) that communicates a cardiac status check via one or more outputs, a telerehabilitation or telemedicine platform that allows a remote medical or rehabilitation professional to examine a patient via an integrated video display while communicating one or more outputs to the platform during an activity (e.g., remote exercise, virtual doctor visit), an insurance application that communicates insurance adjustments based at least in part on a simulation data output, a sports betting platform utilizing a simulation data output, etc. Also included may be media broadcasts that incorporate simulation data (e.g., providing predictions regarding the outcome of a sporting event), sports streaming content platforms (e.g., a video platform) that integrate simulation data as a supplement to a live sporting event being viewed (e.g., allowing a user to bet while viewing the live content), etc. Also included may be non-display media (e.g., a key fob or a scannable object) that provide information regarding the health status of one or more individuals to one or more other systems.

[0056]

[0068] In one variation on application (11), the one or more promotions can be any promotion that provides support to increase acceptance and / or acquisition (e.g., sale, distribution) of one or more products, including one or more advertisements, offers using simulation data (e.g., an offer to a target subject to obtain insurance that may lower bets by utilizing the target subject's animal data to enable one or more simulations to be run), discounting mechanisms using simulation data (e.g., n simulations predict that player X will lose a match against player Y; thus, the betting system updates the resulting odds in real time or near real time based on new information collected by the simulation system and new simulations being run, providing the user / bettor with better odds for player X to win the match), etc.

[0057]

[0069] In one variation on application (12), "combinations thereof" may include any combination of the aforementioned applications, including all of the aforementioned applications or a subset of the aforementioned applications.

[0058]

[0070] More specifically, computing device 12 or computing device 30 may be operable, directly or indirectly: (1) to offer or accept one or more bets; (2) to create, enhance, modify, acquire, offer or dispense one or more products; (3) to evaluate, calculate, derive, modify, enhance or communicate one or more predictions, probabilities or possibilities; (4) to formulate one or more strategies; (5) to perform one or more actions; (6) to mitigate or prevent one or more risks; (7) to recommend one or more actions; (8) to engage one or more users; or (9) to perform any combination thereof.

[0059]

[0071] As mentioned above, the one or more sensors 18 may include one or more biological sensors (biosensors). A biosensor collects a biosignal, which in the context of the present embodiment is any signal or characteristic in or derived from an animal that can be continuously or intermittently measured, monitored, observed, calculated, computed or interpreted, including both electrical and non-electrical signals and artificially generated information. A biosensor may collect biological data (including measurements and signals), such as physiological data, biometric data, chemical data, biomechanical data, genetic data, genomic data, location data or other biological data, from one or more target individuals. For example, some biosensors may collect eye tracking data (e.g., pupillary response, movement, pupil diameter, EOG related data), blood flow and / or blood volume data (e.g., PPG data, pulse wave transit time, pulse wave arrival time), biofluid data (e.g., analyses derived from blood, urine, saliva, sweat, cerebrospinal fluid), body composition data (e.g., weight-based data including bioelectrical impedance analysis, weight, body mass index, body fat data, bone mass data, protein data, basal metabolic rate, fat-free body mass, subcutaneous fat data, visceral fat data, body water data, metabolic age, skeletal muscle data, muscle mass data), pulse data, oxygenation data (e.g., SpO2), core body temperature data, galvanic skin response data, skin temperature data, sweat data (e.g., rate, composition), blood pressure data (e.g., systolic, diastolic, MAP), glucose data, etc. (e.g., fluid balance I / O), hydration data (e.g., fluid balance I / O), heart rate based data (e.g., heart rate, average HR, HR range, heart rate variability, HRV time domain, HRV frequency domain, autonomic tone, ECG related data including PR, QRS, QT, RR intervals, echocardiographic data, thoracic electrical bioimpedance data, transthoracic electrical bioimpedance data), neurological and other neurological related data (e.g., EEG related data), gene related data, genomic related data, skeletal data, muscle data (e.g., surface EMG, EMG related data including amplitude), respiratory data (e.g., respiratory rate, breathing pattern, inhalation / exhalation ratio, tidal volume, respiratory measurement data).Some biosensors may detect biological data such as biomechanical data, which may include, for example, angular velocity, joint path, motion or kinematic load, gait profile, step count, or position or acceleration in various directions that characterize the movement of a subject. Some biosensors may collect biological data such as location and position data (e.g., GPS, ultra-wideband RFID-based data, posture data), facial recognition data, audio data, kinesthetic data (e.g., physical pressure captured from sensors placed on the soles of shoes), or auditory data about one or more target individuals. Some biosensors are image or video based and may collect, provide and / or analyze video or other visual data (e.g., still or moving images including video, MRI, computed tomography scans, ultrasound, echocardiograms, x-rays) on which biological data may be detected, measured, monitored, observed, extrapolated, calculated or computed (e.g., biomechanical movement or position based information derived from video data, fractures detected based on x-rays, or stress or disease of a subject observed based on video or image based visual analysis of a subject). Some biosensors may derive information from biological fluids such as blood (e.g., veins, capillaries), saliva, urine, sweat, etc., including triglyceride levels, red blood cell count, white blood cell count, adrenocorticotropic hormone levels, hemoglobin levels, platelet counts, ABO / Rh blood type, blood urea nitrogen levels, calcium levels, carbon dioxide levels, chloride levels, creatinine levels, glucose levels, hemoglobin A1c levels, lactate levels, sodium levels, potassium levels, bilirubin levels, alkaline phosphatase (ALP) levels, alanine transaminase (ALT) levels and aspartate aminotransferase (AST) levels, albumin levels, total protein levels, prostate specific antigen (PSA) levels, microalbuminuria levels, immunoglobulin A levels, folate levels, cortisol levels, amylase levels, lipase levels, gastrin levels, bicarbonate levels, iron levels, magnesium levels, uric acid levels, folate levels, vitamin B-12 levels, etc. In one variation, some biosensors may collect biochemical data including acetylcholine data, dopamine data, norepinephrine data, serotonin data, GABA data, glutamate data, hormone data, and the like.In addition to biological data regarding one or more target subjects, some biosensors may measure non-biological data, such as ambient temperature data, humidity data, altitude data, and barometric pressure data. In particular, one or more sensors provide biological data that includes one or more calculations, calculations, predictions, probabilities, possibilities, estimates, estimates, inferences, judgments, inferences, observations, or predictions that are derived at least in part from the biosensor data. More particularly, one or more biosensors may provide two or more types of data, at least one of which is biological data (e.g., heart rate data and VO2 data, muscle activity data and accelerometer data, VO2 data and altitude data).

[0060]

[0072] More specifically, at least one sensor 18 and / or one or more appendages thereof may be attached to, contact, transmit one or more electronic communications relating to or derived from one or more target objects, including one or more target objects' skin, eyes, biological organs, muscles, hair, veins, biological fluids, blood vessels, tissues, or skeletal systems, be embedded in, reside within, or implanted in, ingested by, or integrated to include at least a portion of one or more target objects, such as a saliva sensor attached to a tooth, a set of teeth, or a device that contacts one or more teeth, a sensor that extracts DNA information derived from a target object's biological fluids or hair, a wearable sensor (e.g., on the human body), a sensor on a phone that tracks location information of a target individual, a sensor attached to or implanted in a target object's brain that may detect brain signals from neurons, a sensor ingested by a target object to track one or more biological functions, a sensor attached to or integrated with a machine (e.g., a robot) that shares at least one characteristic with an animal (e.g., a robotic arm capable of performing one or more tasks similar to a human's capabilities; a robot capable of processing information similar to a human's capabilities), etc. Advantageously, the machine itself may be comprised of one or more sensors and may be classified as both a sensor and a target. More specifically, one or more sensors 18 are integrated into or part of or attached or embedded in a fiber, fabric, garment, material, fixture, object or device that contacts or communicates with a target individual, either directly or through one or more intermediaries or intervening items.Examples include sensors attached to the skin via adhesive, sensors integrated into a watch or headset, sensors integrated or embedded into a shirt or jersey, sensors integrated into a steering wheel, sensors integrated into a video game controller, sensors integrated into a basketball that is in contact with the hand of a target subject, sensors integrated into a hockey stick or hockey puck that is in intermittent contact with an intermediary (e.g., a hockey stick) held by a target subject, sensors integrated into or embedded into one or more handles or grips of a fitness machine (e.g., a treadmill, a bicycle, a bench press), sensors integrated into a robot (e.g., a robotic arm) controlled by a target individual, sensors integrated into or embedded into a shoe that may contact the target individual through an adhesive tape wrapped around the sock of an intermediary and the ankle of the target individual, and the like. More specifically, one or more sensors may be woven, embedded, integrated, or attached to a floor or ground (e.g., artificial turf, grass, basketball court, soccer field, manufacturing / assembly line floor), a seat / chair, helmet, bed, an object that is in contact with a target subject (e.g., a subject in contact with a seat sensor through the medium of clothing) directly or through one or more intermediaries, etc. More specifically, the sensor and / or one or more of its appendages may contact one or more particles or objects originating from the subject's body (e.g., tissue from an organ, hair from the subject) from which the one or more sensors derive or provide information that can be converted into biological data. More specifically, one or more sensors may be optically based (e.g., camera based) and provide an output from which biological data can be detected, measured, monitored, observed, extracted, extrapolated, inferred, deducted, estimated, determined, calculated, or computed. More specifically, one or more of the sensors may be light-based and may use infrared technology (eg, a temperature or heat sensor) to calculate the temperature of the individual or the relative heat of different parts of the individual.

[0061]

[0073] In one variant, the simulated animal data is generated by randomly sampling at least a portion of a set of real animal data. In another variant, the real data is transformed into simulated data by adding a small random number to each value of the real data set. Small in this context means that the random number has a value within a predefined percentage of the number to which it is added. In particular, the predefined value of the priority is 1, 10, 20, 30, 40 or 50 percent of the value to which it is added. Even more particularly, the average of the small random number is zero. In another variant, an offset value is added to each value of the real animal data. Even more particularly, the offset value of the priority is 0.1, 0.5, 1, 2, 3, 5 or 10 percent of the value to which it is added. For this purpose, the random numbers used for random sampling can be uniformly distributed or normally distributed (e.g. Gaussian random numbers).

[0062]

[0074] In another variation, one or more simulations may be created on the fly based on past data and learning. In this regard, the simulated animal data may be converted by many methods into a form that can be input into a simulation (e.g., a video game, a simulated sporting event, a simulated event to predict or forecast one or more biological events for purposes such as adjusting health insurance premiums). In one detail, the real animal data is numerically modeled by fitting the real animal data to a function using one or more independent variables or one or more adjustable parameters optimized to provide a fit. In this context, such a fitting function is referred to as a model. In such a data model, one or more independent variables or parameters are input by the simulation to provide a simulated data output. In this regard, time (t) is a useful independent variable that can be used to output a simulated biological output (e.g., a physiological output) as a function of the time that the simulated individual participates in the simulated event. In particular, biological parameters may be associated with the virtual participants of the simulation as a function of time.

[0063]

[0075] In another variation, the biological parameters for the actual animal data previously acquired from one or more target subjects may be approximated by a probability distribution. Examples of probability distributions include, but are not limited to, Bernoulli distribution, uniform distribution, binomial distribution, normal distribution (i.e., Gaussian distribution), Poisson distribution, exponential distribution, Lorentz distribution, etc. Typically, these probability distributions may be randomly sampled to assign one or more biological parameters (e.g., physiological parameters) to one or more simulation participants in the simulation. For example, the biological parameters for the actual animal data previously acquired from one or more target subjects may be approximated by a Gaussian distribution with mean and standard deviation as adjustable parameters. The Gaussian distribution may then be randomly sampled to provide values ​​for the simulation. Alternatively, the actual animal data may be fitted to any function (e.g., linear, polynomial, exponential, Lorentzian, piecewise linear, or spline between the actual data points, etc.) that is then applied by the simulation. In particular, previously acquired real animal data may have one or more externally associated parameters, such as temperature, humidity, altitude, time, and other non-biological data, that may be applied as independent variables or parameters in one or more simulations. More particularly, one or more biological parameters (e.g., heart rate, diastolic blood pressure, systolic blood pressure, sweat rate, distance traveled, etc.) for a particular target individual may be functionally modeled (e.g., fitted to a polynomial) as a function of time while engaged in an activity. In this latter example, the simulation may use the modeled function to provide values ​​for the target individual as the simulation progresses over time. In this regard, the simulation data may be used to assess biological events (e.g., fatigue levels) of participants in the simulation. For example, an increasing heart rate, diastolic blood pressure, systolic blood pressure, sweat rate, cumulative over a period of time that a player has may be used as a measure of fatigue. FIG. 2 shows a plot of collected heart rate data and a polynomial fit (polynomial order 60) to the data.

[0064]

[0076] In one variation, the artificial data set may be generated randomly or otherwise in response to one or more starting parameters set by the user. This may be useful when the real animal data desired by the user cannot be acquired, captured or created in the required time frame or manner. If the user has requirements that may preclude acquisition of real animal data, the simulation system 10 may create artificial animal data derived from at least a portion of the real animal data that fits the parameters established by the user, or one or more derivatives thereof, that may be made available for consumption. In this regard, the one or more parameters selected by the data acquirer determine the range of relevant real animal data that may be utilized as one or more inputs from which the artificial data is generated and / or to ensure that the generated artificial output meets the acquirer's desired requirements. For example, a pharmaceutical company or research institute may want to obtain 10,000 2-hour ECG datasets from at least 10,000 unique males ages 25-34, weighing 175-185 pounds, with habits including social smoking (15-20 cigarettes per week), at least one alcoholic drink 2-3 days per week, during sleep, with a particular blood type indicating a biofluid-derived level, and with a family history of diabetes and stroke, from sensor C, at a sampling rate setting of x. Since the simulation system may have, for example, 500 datasets from 500 unique males that meet the requester's minimum requirements, the simulation system may create another 9,500 datasets for the 9,500 unique simulated males to meet the pharmaceutical company's or research institute's requirements. To create the requested datasets, the simulation system may randomly generate artificial datasets (e.g., artificial ECG datasets) using the required parameters and based on the 500 sets of real animal data. The new artificial data set or data sets may be created by applying one or more artificial intelligence techniques to analyze previously captured data sets that match some or all of the characteristics required by the acquirer.One or more artificial intelligence techniques (e.g., one or more trained neural networks, machine learning models) may recognize patterns in the actual data set, be trained with the collected data to understand the biology and related profiles of the animals (e.g., humans), be further trained with the collected data to understand the impact of one or more parameters or variables on the animal biology and related profiles, and generate artificial data that takes into account one or more parameters or variables selected by the acquirer to match or meet the acquirer's minimum requirements. In particular, different data sets from similar individuals, or similar data sets from different individuals, may also be utilized by one or more artificial intelligence models for both model training and data generation purposes. More particularly, a user selects one or more parameters or variables for one or more simulations utilizing at least a portion of the animal data, one or more simulations are generated, and one or more users obtain at least a portion of the simulation data or one or more derivatives thereof for consideration (e.g., payment, other non-monetary value). For example, in the context of sports betting, the simulation system is operable to provide bettors, bookmakers, or other interested parties with the opportunity to obtain (e.g., purchase) one or more simulations that utilize at least a portion of the collected animal data (e.g., collected athlete sensor data) to predict one or more outcomes. Advantageously, such simulations may occur in real-time or near real-time. More particularly, at least a portion of the non-animal data is utilized as one or more parameters or variables in the one or more simulations. Further details relating to monetization systems for animal data having particular application for generating and monetizing simulation data derived from one or more animals are disclosed in U.S. Application No. 62 / 834,131, filed April 15, 2019, U.S. Application No. 62 / 912,210, filed October 8, 2019, and International Application No. PCT / US20 / 28355, filed April 15, 2020, the disclosures of which are incorporated herein by reference in their entireties.In one detail, the data model described above may be used to create the simulated data. More specifically, the simulated data may be created by application of one or more artificial intelligence techniques (e.g., machine learning, deep learning) that may, for example, utilize one or more neural networks to analyze one or more previously captured or created datasets that match at least one of the characteristics required by the acquirer, the details of which are described herein. In this regard, the artificial intelligence-based engine recognizes one or more patterns or upper and lower limits of what is possible in various scenarios of one or more real animal datasets, and creates artificial data that matches or meets the minimum requirements of a user (e.g., a betting entity, a bettor, a pharmaceutical or health care provider seeking to acquire a large amount of data with specific characteristics, an insurance company, etc.). The one or more datasets may be created based on a single individual, a group of one or more individuals with one or more similar characteristics, a random selection of one or more individuals in a defined group of one or more characteristics, a random selection of one or more characteristics in a defined group of one or more individuals, a defined selection of one or more individuals in a defined group of one or more characteristics, or a defined selection of one or more characteristics in one or more defined groups. In particular, the group may include multiple groups. Based on the user's requirements, the simulation system may isolate a single variable / parameter or multiple variables / parameters for reproducibility in creating one or more artificial data sets to keep the data related and random.

[0065]

[0077] In one variation, one or more neural networks are utilized to generate the simulated animal data. Generally, the neural network generates the simulated animal data after being trained with real animal data. Animal data (e.g., ECG signals, heart rate, biological fluid measurements) are collected from one or more sensors from one or more individuals, usually as time series observations. A sequence predictive machine learning algorithm may be applied to predict possible animal data values ​​based on the collected data. The collected animal data values ​​are passed to one or more models during the training phase of the neural network. The neural network utilized to model this non-linear data set trains itself based on established principles of one or more neural networks. At least two separate methodologies are described herein for generating artificial animal data from real animal data based on the utilization of one or more trained neural networks. However, the present invention is not limited to the type of methodology or neural network utilized to generate artificial animal data from real animal data. In the first method, a long short-term memory (LSTM) is used to generate the simulated animal data. Long short-term memory (LSTM) is a type of neural network that does not suffer from the disadvantages (i.e. exploding / vanishing gradient) of recurrent neural networks (RNNs). In a second method, a generative adversarial network (GAN) is used to generate simulated animal data. A generative adversarial network (GAN) is a deep neural network architecture composed of two neural networks, pitting one against the other (adversarial). A generator utilizes the GAN to generate one or more novel data values, which may include one or more novel data sets, while a discriminator evaluates the one or more novel values ​​based on one or more user-defined criteria to prove, verify, or authenticate the newly created values.

[0066]

[0078] Before defining or designing a model and associating one or more neural networks, the first step is to evaluate the data and determine what relevant characteristics are exhibited in the data. There are many relevant animal data characteristics that can be input to train one or more neural networks. For example, in the case of ECG-based data, there are several characteristics that may be relevant, including time series, non-linear functions, autoregressive behavior, and thresholds. The thresholds encompass commonly accepted values ​​or principles (e.g., it may be established that men over 90 should contact their physician if their heart rate exceeds 200 beats per minute, or that the age-based maximum heart rate for a 33-year-old man is n beats per minute). FIG. 3A shows a graph of heart rate per minute (BPM) captured from a professional athlete, while FIG. 3B shows the autocorrelation function of the data in FIG. 3A.

[0067]

[0079] Figure 4 shows an overview of neural networks that can be used in neural network architectures to generate simulation data. Neural networks have been proven to be universal function approximators (i.e., they can model any nonlinear function). A neural network passes an input, such as an image, through multiple layers of digital neurons. Each layer reveals additional features of the input. The architecture of the network - how many neurons and layers it has, and how they are connected - determines the type of tasks it is good at. When data is fed into the network, each artificial neuron that fires sends a signal to a specific neuron in the next layer, and these neurons may fire if multiple signals are received. This process reveals abstract information about the input. Shallow networks have several layers, but many neurons per layer. These types of networks are intensive from a computational perspective. Deep networks have many layers and relatively few neurons per layer. Deep networks can achieve a high level of abstraction using a relatively small number of neurons. Each neuron is activated based on the following rules: TIFF0007656942000001.tif7170f is the activation function; W is the weight matrix; x is the input vector; b is the bias; and Y is the output vector.

[0068]

[0080] As is known in neural network technology, the weight matrix is ​​updated by a process called backpropagation, in which the gradient of the error between the predicted and expected outputs for the weights is used to update the weights of each neuron based on a learning rate in the direction in which the gradient decreases.

[0069]

[0081] FIG. 5 shows details of a recurrent neural network that can be used to generate animal data. Recurrent neural networks (RNNs) are a classification of neural networks in which the connections between nodes form a directed graph along a time series. This allows the neural network to exhibit temporal dynamic behavior. Unlike forward propagation neural networks, RNNs may process a sequence of inputs using their internal state (memory). They are designed to handle sequence prediction problems. RNNs may keep track of any long-term dependencies of the input sequence. As shown in FIG. 5, a recurrent neural network 40 includes a base network 42 that is repeated multiple times i. Typically, i is chosen during the training phase of the RNN model to optimally preserve the required amount of history without adding computational complexity. This is usually achieved by training and testing the model with several possible values ​​that are chosen based on patterns observed in the data, autocorrelation statistics, heuristics, and the modeler's domain-specific knowledge, among others. In this figure, the inputs are labeled X, weighted by a weight matrix U, and provided to a hidden layer S. W is the output from hidden neuron layer S with an appropriate activation function f. In the case of a single hidden layer, W = f(UX + b), where b is the bias, at least for the first hidden layer if there are multiple hidden layers. In particular, a hidden layer component may contain multiple hidden neuron layers. The output W from hidden layer j is j is provided to the corresponding hidden layer j at the next time step. Examples of suitable activation functions include, but are not limited to, sigmoid function, Tan function, ReLU, leaky ReLU, and other activation functions known to those skilled in the art. In this network, an output O is generated from a hidden neuron layer S, e.g., O=f(VW+b 2 After the RNN is trained, simulated data is generated (e.g., randomly generated) by providing input to the first cell. The output from this cell is then provided as input to the next cell, and the process is repeated for each subsequent cell to generate a complete set of data.

[0070]

[0082] The problem with standard RNNs is computational (or practical) in nature; when training a standard RNN using backpropagation, the backpropagated gradients may vanish (i.e., tend to zero) or explode (i.e., tend to infinity) due to the calculations involved in the process using finite precision numbers. RNNs using LSTM units solve the vanishing gradient problem because the LSTM unit allows the gradients to flow unchanged. The general architecture consists of a cell (the memory part of the LSTM unit) and three regulators of the information flow inside the LSTM unit (usually called gates): an input gate, an output gate, and a forget gate. Figure 6 shows a schematic diagram of an LSTM that can be used to generate simulated animal data. The LSTM 50 includes a recurrent cell 52. The recurrent cell includes a forget gate layer 54, an input gate layer 56, an output gate layer 58, and a tangent gate layer 60. The output of these layers is given by the following equation: TIFF0007656942000002.tif25170i is the output of the input gate layer; f is the output of the forget gate layer; o is the output of the output gate layer; where t is the current time step; t-1 is the previous time step; t+1 is the next time step; g is the output of the tongue gate layer; W is the weight matrix; x t is the input vector (or value) at t time step; h t is the hidden state vector at time step t-1; σ is the sigmoid activation function; and tanh is the tan activation function. Memory cell value c t The formula for is: To generate simulated data after the LSTM is trained, the simulated data is generated by providing inputs (e.g., randomly generated) to a first LSTM cell, and then the hidden states generated from this cell are provided to a trained neuron layer (e.g., trained with the LSTM cell) to generate input values ​​for the next cell. This process is repeated to generate a complete set of simulated data. Table 1 shows an example of pseudocode for generating simulated animal data using the LSTM method, where one or more parameters of the method may be adjustable. TIFF0007656942000004.tif171170

[0071]

[0083] When applying the RNN method (including LSTM variables), animal data from multiple events (e.g., multiple sporting events; multiple biological monitoring sessions in the individual's daily activities, which may include sleep, exercise, work, etc.) are used as samples to train the neural network. The animal data measurements are time-stamped and occur at a predefined period (e.g., approximately every second). First, the model is trained using N such observations (the length of the LSTM sequence), which may be a few (e.g., 20), hundreds, thousands, millions, etc. The network is trained for N epochs (e.g., 100) using MSB (mean squared error) as the error metric and the ADAM optimizer for the implementation of backpropagation (weight updates). For reference, ADAM is an optimization algorithm that can be used instead of traditional stochastic gradient descent to iteratively update the network weights based on the training data. After training the data-specific model (and making the model intelligent), the model is then applied to make predictions for the animal data. In this example using ECG-based data, the model predicts the heart rate data that the model subsequently generates. The predicted animal data (e.g., heart rate) generated by the model is first tested in-sample and then out-sample. Sample refers to the data sample that is being used to fit the model. Once the user has obtained the sample and fit the model to the sample, the user may use the model for prediction. In-sample prediction utilizes a subset of the available data to predict values ​​outside the estimation period and compare them to corresponding known or actual results. By using in-sample prediction, all the artificial animal data generated by the neural network has been previously seen by the model. Thus, if the user is predicting an observation that was part of the data sample, it is an in-sample prediction. In out-of-sample prediction, the data generated by the neural network has not been previously seen by the model. Thus, if the user is predicting an observation that was not part of the data sample, it is an out-of-sample prediction.

[0072]

[0084] 7A and 7B show plots illustrating simulated animal data generated using the LSTM method. The simulated animal data in FIG. 7A and FIG. 7B is simulated heart rate data. Both figures include artificially created animal data (i.e., labeled "predicted heart rate") generated based on real animal data (i.e., labeled "original heart rate"). FIG. 7A shows in-sample, which occurs when a prediction of an observation includes at least a portion of an animal data sample. As described above, by using in-sample prediction, all artificial animal data generated by the neural network has been previously confirmed by the model. For example, if a user wants to generate artificial heart rate data for athlete X based on actual heart rate characteristics (or at least a portion of the actual heart rate data) and incorporate it as part of a video game, the system first trains a model using athlete X's previously captured actual heart rate data, and then uses the just-trained model to generate artificial heart rate data (in-sample) for athlete X. The generated artificial heart rate data is composed of previously confirmed values. FIG. 7B shows out-sample. In out-of-sample prediction, the model is trained using at least a portion of real animal data. However, unlike in-sample, the artificial animal data generated by the neural network has never been seen by the model before. This generated artificial animal data is entirely new data and is based on at least a portion of real animal data. Out-of-sample prediction includes the generation of new animal data for target subjects (1) based on one or more characteristics of the subject, for which the model has never seen data from that particular subject, and (2) for target subjects for which the model has seen their animal data, but by introducing one or more variations (e.g., changes, adjustments) to one or more parameters or variables that create a new artificial dataset. Characteristically, out-of-sample datasets can be used for prediction use cases.For example, if Athlete X is playing in a real sporting event (e.g., match, game) and a user wants to predict what Athlete X's heart rate will be for the next 5 minutes, the user can use Athlete X's previously collected heart rate datasets from previous matches / games and auxiliary datasets associated with each heart rate dataset that provide context to one or more heart rate datasets, and the system utilizes the collected heart rate data and the auxiliary data to train a model to artificially create "next 5 minutes" heart rate data for Athlete X. In another example, if the model has only seen Athlete X's heart rate data when the field temperature is 90 degrees Fahrenheit, but the system is requesting the model to generate Athlete X's heart rate data when the field temperature is 110 degrees Fahrenheit, out-of-sample predictions are used to generate artificial heart rate data for Athlete X based on an adjustable temperature input (e.g., 110 degrees).

[0073]

[0085] FIG. 8 illustrates a method for generating simulated animal data from a generative adversarial network (GAN). A GAN is a deep neural net architecture consisting of two neural nets, one against the other (adversarial). The GAN 60 includes a generator component 62 that generates one or more novel data values ​​64, which may include one or more novel data sets, while a discriminator component 68 evaluates the one or more novel values ​​based on one or more user-defined criteria to prove, verify, or authenticate the one or more novel values. For example, the discriminator component 68 determines whether each example of data it reviews belongs to the actual training data set. A discrimination algorithm tries to classify input data (i.e., predict the label to which the data belongs given the characteristics of the observations of the data). Mathematically, a label is called y and a feature is called x. A discriminator tries to predict the probability that the data belongs to a label given y, i.e., p(y|x), or the features. A generative algorithm tries to obtain the features, i.e., x. They capture the probability of a particular feature given p(x|y) or a label.

[0074]

[0086] In this method, one or more animal datasets are used as samples to train a discriminator. The discriminator is provided with alternative datasets (e.g., fake data or bad data) so that it can register differences between one dataset versus another (e.g., true data versus fake data, good data versus bad data). For example, an application may use a GAN to train a neural network to distinguish whether an object is a particular type of food. In another example, a GAN may be used to determine whether athlete Y's sweating pattern in a game played at 80% humidity and 95 degrees ambient temperature is trained using data related to athlete Y's sweating pattern at 80% humidity and 95 degrees ambient temperature. A user may determine one or more characteristics (e.g., quantity, quality) of one or more alternative datasets (e.g., fake data, bad data) that they wish to provide to train the neural network. Characteristically, as the neural network acquires more data, it may have the ability to determine the appropriate characteristics required to train itself. Generally, the higher the data quality of the system, the better the network will be. The discriminator performs a process of evaluating good vs. bad data. The discriminator creates a feedback loop so that it can learn the characteristics of good and bad data and evaluate why good data is good and bad data is bad. This allows the discriminator to evaluate whether the generated animal data meets a threshold (in this example, an ECG-based measurement) established as true or not by the trained model. The generator receives one or more inputs (e.g., random numbers, a constrained set of numbers) and generates a single value (e.g., a candidate ECG measurement) that is evaluated by the discriminator. The discriminator then feeds the results back to the generator, creating a learning feedback loop.For example, if the generator generates three consecutive heart rate values ​​at 1x per second: 43 beats per minute (bpm), 45 beats per minute, and 300 beats per minute, the discriminator will look at this pattern and recognize that this heart rate pattern is invalid if the neural network is trained to recognize that the heart rate cannot increase from 45 to 300 bpm in one second. When this occurs, the generator will regenerate new values ​​until the discriminator "approves" the value generated by the generator. In particular, the algorithm is adapted to a time series regression style problem for streaming animal data, although the basic ideas of GANs are applicable to creating artificial animal data or predicting animal data values.

[0075]

[0087] Figure 8 shows a plot illustrating heart rate data generated using the GAN method. Table 2 shows an example of pseudocode for implementing the GAN method for generating simulated animal data, one or more parameters of which may be adjustable. TIFF0007656942000005.tif201170TIFF0007656942000006.tif69170

[0076]

[0088] In particular, the one or more trained neural networks utilized to generate the simulated animal data may be a forward propagation, perceptron, deep forward propagation, radial basis network, gated recurrent unit, autoencoder (AE), variational AE, denoising AE, sparse AE, Markov chain, Hopfield network, Boltzmann machine, restricted BM, deep belief network, deep convolutional network, deconvolutional network, deep convolutional inverse graph network, liquid state machine, extreme learning machine, echo state network, deep residual network, Kohonen network, support vector machine, neural Turing machine, group method of data processing, probabilistic, time-domain, or time-domain. The neural network may be one or more of the following types: inter-delay, convolutional, deep stacking network, general regression neural network, self-organizing map, learning vector quantization, simple recursive, reservoir computing, echo state, bidirectional, hierarchical, probabilistic, genetic scale, modular, machine committee, associative, physical, instantly trained, spike, regulatory feedback, neocognitron, composite hierarchical deep model, deep predictive coding network, multi-layer kernel machine, dynamic, cascade, neuro-fuzzy, constituent pattern generation, memory network, one-shot associative memory, hierarchical temporal memory, holographic associative memory, semantic hash, pointer network, or encoder-decoder network. In one variation, multiple neural networks are utilized with at least a portion of the same animal data or one or more of its derivatives to generate the simulated data.

[0077]

[0089] In each of the neural network methods, the data utilized by the model may include one or more adjustable parameters or variables that can create a more targeted artificial data set based on the user's preferences. For example, in the context of professional sports, parameters or variables that a user may want to incorporate when creating a data set of artificial heart rate data for a target subject (e.g., in a basketball game) may include inputs such as body temperature, environmental temperature, distance traveled, biofluid measurements, hydration levels, muscle fatigue, respiratory rate, etc.Additionally, parameters or variables may include traditional statistics (e.g., points, rebounds, assists, time played), in-game data (e.g., whether a player is on or off the court, whether a player is playing offense or defense, whether a player has the basketball or not, a player's position on the court at any given time, specific on-court movements at any given time, which players are guarding defensive players, who is guarding offensive players), historical data (e.g., historical heart rate data, historical body temperature / mileage / biofluid measurements / hydration levels / muscle fatigue / breathing rate data, biological data sets of players for any given team, who is guarding a player at any given time, who is guarding a player in any given game, biological measurements of players guarding any player, the dynamics of a player's performance at any given time, and the dynamics of a player's performance at any given time). The variables may also include data that provides context to the biological data, including biological measurements of players guarded by a player, playing time, biological measurements of players playing against any given offense or defense, playing time, on-court position and movement for any given game, traditional statistics, other in-game data), comparison data with similar and dissimilar players in similar and dissimilar situations (e.g., statistics of other players when guarding or being guarded by a particular player, playing against a particular team), injury data, recovery data (e.g., sleep data, rehabilitation data), training data (e.g., how a player performed in training in the days or weeks leading up to a game), nutritional data, player self-evaluation data (e.g., how they feel physically, mentally or emotionally), etc. Other variables may also include age, weight, height, date of birth, race, nationality, habits, activity, genomic information, genetic information, medical history, family history, medication history, etc. It should be understood that such parameters / variables are merely exemplary and are not exhaustive.

[0078]

[0090] Characteristically, animal data may provide information about why one or more outcomes occurred and what the future may hold for one or more outcomes. For example, predictions are often made in limited context and based on past performance (e.g., biological performance, task performance) without knowing which animal data drove the past outcome. In the context of sports betting, analysis may be focused around past statistical performance (e.g., individual statistics, team statistics), situational context of performance (e.g., venue, conditions, playing time, past results against opponents), and derived trends (e.g., Player C's batting average with the bases loaded against Team X's right-handers, .274). Often, the missing context for the captured data is how (often biologically) one or more target individuals or groups of target individuals arose to drive (e.g., influence) past outcomes. In this regard, animal data may provide the missing context and simulation data may provide information about what may or will occur next based on the animal data and other variables or parameters. More specifically, simulation data may provide context for future outcomes. For example, the simulation system may utilize information captured regarding historical statistical performance, situational context of the performance, and derived trend information, and correlate this information with the animal data to determine what caused these results and establish a baseline for any given individual or group of individuals. In one variation, the simulation system may utilize past animal data (e.g., heart rate, hydration data, biomechanical data, position data) of one or more target individuals, situational context information regarding the animal data (e.g., was the player stressed when the result occurred? Was the player dehydrated or tense when the result occurred?), and one or more trends within past animal data states (e.g., Player A misses outside shots from n feet y% of the time when fatigue level is less than z%), and correlate this information with non-biological information regarding performance.Once a baseline is established of what caused past outcomes, the simulation system may utilize the baseline data to interpret collected animal data measurements to better understand why any given outcome occurred. Advantageously, data collection and analysis may occur in real-time or near real-time. In one variation, the simulation system may run one or more simulations based on the baseline data and collected animal data (e.g., historical data, real-time or near real-time animal data) to generate simulated data for predicting future animal data for one or more target individuals (e.g., future heart rate data of an athlete in the fourth quarter of a given sports competition). In a variation, the simulated animal data (e.g., future simulated heart rate data) generated by the simulation may be used as one or more inputs in one or more further simulations to predict outcomes (e.g., an athlete will make / miss their next shot, win / lose a match, etc., based on the athlete's future heart rate measurements). In particular, one or more artificial intelligence techniques may be utilized to correlate the data sets and identify known biologically relevant problems from one or more target individuals or groups of target individuals, as well as to identify hidden patterns within one or more data sets and identify biologically relevant problems based on the collected data. This may include the study of entirely novel patterns in data that have not previously been correlated with known problems, or the study of novel patterns within one or more data sets that may identify novel problems. For example, collecting animal data, situational context information about the animal data, and trends based on real-time or near real-time animal data allows for comparison of real-time or near real-time data to historical data corresponding to the situation, while allowing for evaluation of both micro-trends (e.g., within seconds or minutes) and macro-trends (e.g., over the course of a match).Advantageously, one or more simulations may occur in real time or near real time and / or utilizing historical animal data information and may predict one or more occurrences and / or predict one or more outcomes for one or more future animal data measurements of one or more target individuals or groups of target individuals.

[0079]

[0091] When utilizing one or more of the aforementioned methods, the collection period of a previously collected or current actual data set may be extended with simulated data. For example, a simulation system 10 having access to a given amount of in-play data for athlete A (e.g., 10 hours, 100 hours, 1000 hours or more) as well as one or more other types of data associated with athlete A and the matches played by athlete A (e.g., in the context of a sport such as tennis, altitude, temperature on the court, humidity, heart rate, distance traveled, swing speed, energy level, respiration rate, muscle activity, hydration level, data derived from biological fluids, shot power, length of point, court location, opponent, opponent's performance in specific environmental conditions, win percentage against opponent, % win percentage against opponent in similar environmental conditions, current match statistics, past match statistics based on performance trends in the match) may use one or more artificial intelligence based models to augment a given data set to recreate data from matches that a given athlete did not play or has not yet played (e.g., Player A played a 2 hour, 3 set match with a captured heart rate, but the user would like to know Player A's heart rate data for the 4th set before the event occurred. Thus, the simulation system runs one or more simulations to create simulated animal data). More specifically, one or more neural networks may be trained with one or more biological and non-biological data sets associated with athlete A to understand athlete A's biological function and how one or more variables may affect any given biological function. The neural network may be further trained to understand what outcome (or outcomes) occurred based on the one or more biological functions and the influence of the one or more variables.For example, one or more biological functions of athlete A in any given scenario including the current scenario, one or more variables that may affect one or more biological functions of athlete A in any given scenario including the current scenario, one or more outcomes that have occurred previously in any given scenario including the current scenario based on one or more biological functions exhibited by athlete A and / or one or more variables present, one or more biological functions of athletes similar and dissimilar to athlete A in any given scenario including scenarios similar to the current scenario, one or more other variables that may affect one or more biological functions of athlete A in any given scenario including scenarios similar to the current scenario, one or more variables that may affect one or more biological functions of other athletes similar and dissimilar to athlete A in any given scenario including scenarios similar to the current scenario, and one or more variables that may affect one or more biological functions of athletes similar and dissimilar to athlete A in any given scenario including scenarios similar to the current scenario. When trained to understand information such as one or more outcomes that have occurred previously in any given scenario, including scenarios similar to the current scenario based on one or more biological functions associated with the above variables, the acquirer of the data may request to run one or more simulations to augment the currently collected data set with artificially generated data (e.g., Athlete A has just played for two hours with various biological data including location-based data captured. The acquirer needs Athlete A's location-based data for the third hour of the match, under the same match conditions, three hours before the match is occurring, so the system may run one or more simulations to create data based on the previously collected data), or may predict the outcome that will occur in any given activity (e.g., looking only at Athlete A's data, the likelihood of Athlete A winning the match, or the likelihood of winning the fourth set, or other outcomes for Athlete B).In one variation, one or more neural networks may be trained with multiple animals (e.g., athletes), possibly competing in teams, groups, or against each other, and one or more neural networks may be trained with one or more data sets from each animal to more accurately generate simulation data from which predictive indicators can be derived for predicting one or more outcomes (e.g., whether athlete A will win a match against athlete B). In this example, one or more simulations may be run to initially generate artificial animal data based on the actual animal data for each athlete, and then at least a portion of the generated artificial animal data may be utilized in one or more further simulations to determine the likelihood of and / or predict any given outcome.

[0080]

[0092] In one variation, the simulated animal data generated by any of the methods described herein may be converted into a lookup table to be used by the simulation. In another variation, one or more inputs may be provided by a user or may be artificially created by an artificial intelligence based model, depending on one or more user requirements or what the artificial intelligence based model selects.

[0081]

[0093] In another variation, the simulation system 10 provides one or more simulated data sets as a replacement for the animal-generated data sets. In particular, one or more computing devices 12 and / or 30 generate one or more insights, calculated assets, or predictive indicators from at least a portion of the simulated data or one or more derivatives thereof. Advantageously, the simulated data may be used in animal data prediction systems with a particular focus on gambling applications, as well as probability assessment systems relating to medical, telemedicine, insurance, fitness, health / wellness monitoring, and the like. More specifically, the generated simulated animal data may be used, directly or indirectly: (1) as a market in which one or more wagers are made or accepted; (2) to create, modify, enhance, acquire, offer or distribute one or more products; (3) to evaluate, calculate, derive, modify, enhance or communicate one or more predictions, probabilities or possibilities; (4) to formulate one or more strategies; (5) to take one or more actions; (6) to mitigate or prevent one or more risks; (7) as one or more signals or measurements utilized in one or more simulations, calculations or analyses; (8) as part of one or more simulations, the output of which directly or indirectly engages with one or more users; (9) to recommend one or more actions; (10) as one or more core components or supplements to one or more consumption media; (11) in one or more promotions; or (12) any combination thereof. In particular, the simulation system performs one or more simulations utilizing at least a portion of the animal data, and one or more simulations occur and generate simulation data from which one or more products or services are created, modified, enhanced, obtained, offered or distributed by the simulation system or another computing device.For example, a simulation system may be utilized to place one or more bets that bettors obtain (purchase) (e.g., a simulation system may function as a bookmaker and propose animal-data-based prediction products generated from one or more simulations that reveal favorable odds and allow users to obtain products and place bets within the same platform) or may function as a sports betting platform (e.g., a bookmaker) that proposes one or more betting incentive products based on the generated simulation data (e.g., predictions of outcomes of sporting events based at least in part on animal data). In another example, a simulation system may accept one or more bets utilizing at least a portion of the generated simulation data (e.g., a simulation system functioning as a bookmaker may propose or accept one or more bets based on virtual horse races operated by the simulation system; a simulation system functioning as a bookmaker may propose or accept one or more bets on real-world events based on adjusted odds utilizing results of one or more simulations). In another example, a functioning simulation system may adjust the real-time or near real-time odds proposed for any given bet based on one or more results of one or more simulations. Additional details regarding animal data prediction systems utilizing simulated data are disclosed in U.S. Patent Application No. 62 / 833,970, filed April 15, 2019; U.S. Patent Application No. 62 / 912,822, filed October 9, 2019; and International Application No. PCT / US20 / 28313, filed April 15, 2020, the disclosures of which are incorporated herein by reference in their entireties. Such simulated data sets may be derived from animal data and other data that may be utilized as one or more inputs. In particular, the one or more inputs include user behavior (e.g., in connection with sports betting, this may include one or more previous bets or interactions with data; in connection with other scenarios such as insurance, this may include any behavior that may be recorded in a form that can be input into the simulation).Advantageously, the ability to randomize at least one parameter or variable to change or modify one or more parameters or variables in one or more simulations to provide one or more outcomes to potential users may occur in real-time or near real-time. Examples of such parameters include adjustable parameters used to fit real animal data to the functions described above. In the context of sports betting, the ability to run another simulation utilizing a data set based on real animal data in real-time or near real-time may generate entirely new data sets that the simulation system or a third party system may directly or indirectly: (1) offer or accept one or more wagers; (2) create, enhance, modify, obtain, offer or dispense one or more products; (3) evaluate, calculate, derive, modify, enhance or communicate one or more predictions, probabilities or possibilities; (4) formulate one or more strategies; (5) perform one or more actions; (6) mitigate or prevent one or more risks; (7) recommend one or more actions; (8) engage one or more users; or (9) any combination thereof.

[0082]

[0094] In particular, the simulation data incorporating at least a portion of the animal data may be utilized to enhance one or more insights, calculated assets, or predictive indicators. For example, the simulation system 10 may derive a probability or create a prediction regarding a particular outcome occurring based on historical data collected by the system (historical data includes any current data set, e.g., data collected in real-time or near real-time data). By running one or more simulations using the simulation data incorporating at least a portion of the animal data, the system may create, modify, or extend a predictive indicator to determine which outcomes are likely to occur based on various versions of the simulation data. Advantageously, the one or more simulations may occur in real-time or near real-time to provide real-time or near real-time output. Various versions of the simulation data may have different adjustable parameters, as described above, determined by fitting actual animal data. For example, in connection with a tennis match, one or more simulations may be run based on traditional statistics to determine whether a player will win or lose a match against another opponent. This may include head-to-head win / loss ratios, previous win / loss records, rankings, the player's performance in tournaments over the past few years, the player's performance on the court surface (e.g., grass, hard court, clay), and the player's previous match lengths. Analysis may be more detailed within a match, including a tennis player's current match status (e.g., Player A is losing the 4th game of the 2nd set, 6-4, 3-2), historical data (e.g., all of Player A's match outcomes when Player A is losing the 4th game of the 2nd set, 6-4, 3-2, the first serve percentage in the 2nd set after playing n minutes, the percentage of unforced errors on the backhand side after hitting three consecutive topspin backhands).Entirely novel artificial data sets may be created by incorporating animal-derived sensor data (e.g., calculations of location data such as distance traveled, physiological characteristics, biofluid data, biomechanical movements) and other sensor data (e.g., humidity, elevation and temperature of current conditions; humidity, elevation and temperature of previous match conditions). For example, data may be generated for a particular scenario, where the data predicts how a player will perform when the player's heart rate exceeds 190 bpm in the fourth game of the second set, or how a player will perform when in the fourth game of the second set: (1) the player's heart rate exceeds 190, (2) the player has run more than 2.1 miles in the match, and (3) the temperature on the court is greater than 95 degrees. The granularity of the outcome analysis may be finer, allowing a player to determine whether or not to win or lose a particular set, game, or point against another opponent based on the simulation data. The animal data provides context as to why an outcome occurs in any given scenario, and one or more simulations generate artificial data that allows one or more predictions to be based on the added context. In this example, one or more simulations may occur to predict Player A's biological outputs during a match (e.g., future heart rate and respiration rate data), which may be utilized in one or more further simulations to fine-tune insights, calculated assets, or predictive indicators.

[0083]

[0095] In particular, the simulation system is operable to perform one or more simulations utilizing at least a portion of the biological data of one or more subjects to generate simulation data to predict another parameter, condition, or requirement required to achieve optimal performance by one or more subjects. Optimal performance (e.g., a given task, biological function) may include both physical and neurological performance. For example, the mental state of a subject in the "zone" (e.g., flow state) may be predicted by utilizing one or more types of animal data derived from one or more sensors capturing information from one or more target individuals (e.g., one or more combinations of biological data that may include, but are not limited to, heart rate data, ECG data, RR interval data, heart rate variability data (LF / HF ratio), pupil diameter data, respiration rate data, EEG data, EMG data, functional MRI data, movement data, glucose data, FFA metabolism data, movement pattern data, hormone data such as glucocorticoid or HGH data, other biochemical data such as norepinephrine, cortisol levels, and / or dopamine levels, etc.). Such biological information may be combined with one or more variables or parameters (e.g., situational context, environmental data, time, output information such as subject's emotions, subject's skill, the task the subject is performing, sleep, task difficulty for the subject's skill, goal clarity, risk, level of control perceived by the individual, etc.), one or more of which may be determined by questionnaire or other medium that allows for communication of information by one or more individuals, to determine the parameters of the optimal performance state. For example, by examining one or more biological parameters, related variables, and one or more outcomes associated with such parameters and variables, correlations may be made between biological occurrences, variables, and outcomes. Given an individual's baseline of both the biological data and the conditions under which the optimal performance state is achieved, the system may generate simulation data derived from the biological data to predict future optimal performance states within any given set of variables / parameters in order to make one or more adjustments to keep the individual in their optimal performance state.Such optimization may occur for any subject-based performance, such as sports, medical, fitness / wellness, military, general business (e.g., employee wellness), etc. In one variation, the system may utilize one or more artificial intelligence techniques to determine optimal one or more variables / parameters at which a desired biological state (and corresponding animal data measurements) is achieved. For example, once a subject's individual baseline for an optimal performance state - including the subject's animal data measurements - has been determined, the system may generate simulation data from the collected animal data to predict the optimal set of variables / parameters required for one or more target subjects to achieve their optimal performance state.

[0084]

[0096] In another example, an elderly care facility may run one or more simulations utilizing at least a portion of the animal data collected from the target individuals to determine the likely health outcomes of the target individuals and thus the amount of future care required for the target individuals. Based on the amount of care that may be required in the future, the facility may create, modify or enhance pricing for each individual based on the individual's profile (e.g., individualized pricing). In such a scenario, one or more types of artificial animal datasets may first be generated for the target individuals (e.g., a dataset comprised of future ECG measurements for the target subjects for the next n years), from which one or more biological events may be predicted. Furthermore, based on the simulation data, the facility may adjust its staff levels and skills to reflect the expected workload and requirements for caring for one or more target individuals. In some variations, the generated artificial animal data may be utilized in one or more further simulations to create and / or fine-tune predictive indicators (e.g., with future ECG measurements for the target subjects generated by the simulation system, the simulation system may run one or more simulations to determine the likelihood of a heart attack or stroke in the next m months). In another example, an automobile or aircraft manufacturer may want to run simulations to fine-tune predictive indicators to provide one or more responses for target objects within the vehicle or aircraft to mitigate or prevent risk. More specifically, an automobile manufacturer may want to determine whether someone exhibiting a particular biological characteristic (e.g., physiological or biomechanical characteristic) while operating the vehicle may be at risk of causing an accident.The vehicle may utilize the animal data, which may include one or more derivatives thereof, to take one or more actions based on predictive indicators derived from one or more simulations to mitigate or prevent risk (e.g., stop, pull over to the side of the road, drive itself to a hospital) (e.g., if a person is determined to be having a heart attack based on collected sensor data, the vehicle may drive itself to a hospital; if while driving behind the wheel, the vehicle may stop itself if the likelihood of a person having a heart attack with a certain age, weight range, height range, cardiac condition, increased heart rate, elevated blood pressure, elevated stress level, irregular biomechanical movements, etc., is determined to be greater than a predefined threshold; or a more absolute prediction via predictive indicators that a subject will have a heart attack with a set of these predefined characteristics and parameters). In another example, an airline may monitor real-time biological characteristics of one or more pilots during flight via one or more source sensors and take one or more actions based on the probability of an event occurring with respect to at least a portion of the animal data (e.g., notify the airline, take control away from the pilot, put the plane on autopilot, enable remote control of the plane to the airline or airline manufacturer).

[0085]

[0097] In another example, an insurance company may have a hypothesis regarding the outcome of one or more individuals who share one or more characteristics (e.g., height, weight, medical condition). The outcome may be, for example, the likelihood that a person will contract a disease or viral infection in the next n months, the likelihood that a given injury will achieve a given recovery rate, the likelihood that an individual may experience a medical episode (e.g., stroke, heart attack), etc. To test the hypothesis, the insurance company may utilize at least a portion of the animal data to run one or more simulations via the simulation system 10 to calculate the probability of one or more events occurring and determine the expected outcome for those individuals. Based on the one or more simulations, odds of a given outcome may be created, adjusted, or enhanced. In an insurance example, premiums may be adjusted or rates may be set based on the likelihood of an outcome occurring. In one variation, the animal data and one or more derivatives thereof, including the simulation data, may enable the insurance company to better understand the biological state and potential future events of any given target subject or group of target subjects. The animal data, including the simulation data, may be used for a wide range of opportunities in the insurance industry, including those related to the creation, modification, enhancement, acquisition, offering or distribution of one or more products or product categories, and the personalization (including personalized pricing) of such products. By utilizing at least a portion of the simulation data, insurance companies may be able to identify personalized and grouped risks with much greater accuracy and granularity, allowing for the creation or modification of more micro (e.g., activity or group of activity specific) products, and for greater accessibility (e.g., real-time or near real-time availability of such insurance products). Additionally, the animal data, including the simulation data, may enable the creation of one or more risk profiles for any given target individual, group of target individuals, or multiple groups of target individuals, allowing for greater flexibility in segmenting and pricing products based on any given risk associated with any given individual or group.Simulation data can also enable insurance models to be more continuous (rather than fixed), suggest products based on subject biological measurements and corresponding derivatives (e.g., predicted future biological measurements), and customize products based on the specific needs of one or more subjects or groups of subjects.

[0086]

[0098] In the case of a system relating to insurance, the subject may have one or more sensors that collect animal data. The one or more devices may generate one or more signals or measurements that allow the insurance company to monitor the biological measurements of the subject. Advantageously, simulated data may be generated from at least a portion of the collected animal data to provide one or more insights or predictive indicators regarding future events (or likelihood of events) that allow the insurance company to take one or more actions (e.g., adjust premiums). The generated simulated data may be utilized to create, modify or enhance predictive indicators that can provide the insurance company with a risk profile of the subject (e.g., their personal risk profile). Such simulated data may include one or more signals or measurements from one or more non-animal data sources as one or more inputs in one or more simulations (e.g., activities involving humans). Alternatively, one or more simulations incorporating simulated animal data (e.g., measurements of future animal data) of the target subject may be run to generate a predictive indicator output regarding the subject's personal risk (e.g., the likelihood that the target subject will experience any given medical event). The personal risk may be represented in various ways, such as a number or multiple numbers (e.g., a score or other indicator) that the insurance company may use to evaluate any given subject.

[0087]

[0099] Running one or more simulations may provide the insurance company with one or more predictions, forecasts, probabilities, possibilities, estimates, or recommendations regarding one or more outcomes of one or more future events involving one or more target subjects or one or more groups of target subjects. For example, one or more sensors may collect animal data information including heart rate data, ECG measurements, oxygenation data, blood pressure data, weight data, body composition data, pulse data, biofluid data (e.g., data measurements derived from blood or urine), genetic data, and the like. One or more measurements derived from one or more sensors may be anomalous compared to similar and previous measurements from the target individual or other individuals who share one or more comparable characteristics with the target individual (e.g., a baseline group similar in age, weight, height, social habits, medical history, and other physiological characteristics to which the target individual is compared). Ancillary information may be derived from the sensor data (e.g., certain habits or lifestyle choices) or may be included as part of the metadata (e.g., activity) based on one or more measurements. This may be indicative of one or more health issues (e.g., illness, disease, infection, obesity issues, genetic mutations or derivatives), lifestyle issues (e.g., drug use, tobacco use, alcohol abuse), activity issues (e.g., lack of exercise), etc. Upon collecting the data, one or more simulations may be run to generate simulated data. The simulated data may be in the form of a predictive index (e.g., likelihood of any given medical event, illness, disability, death, recovery rate from any given illness, viral infection, bacterial infection, injury, etc.) or may be representative of future animal data under a predefined set of conditions that may be used in its own or further simulations or models to generate predictive indexes.Based on the collected animal data and the output of one or more simulations, the insurance company may take one or more actions that may include adjusting one or more premiums for any given policy (e.g., which may occur in real-time and whenever real-time or near real-time information is collected and / or generated), educating subjects on how any given health outcome can be improved to reduce future costs for all parties (e.g., including one or more specific recommendations regarding exercise, nutrition, habit changes), etc. The insurance company may provide additional incentives based on the animal data (e.g., bonuses or penalties for achieving or not achieving any given animal data-related health goal or sensor-related data goal, alternative pricing options based on the frequency of animal data shared, etc.). Advantageously, the simulation system's ability to collect animal data and / or provide simulation data, either or both of which may occur in real-time or near real-time, allows for real-time or near real-time insurance claims.

[0088]

[0100] In one variation, the animal data and its derivatives (e.g., simulated animal data) may enable a more efficient and automated underwriting process. Claims may be based, at least in part, on captured animal data and derived simulation data from one or more sensors (e.g., one or more sensors may capture irregular biological activity from which information about biological events that have occurred or are likely to occur in the future may be obtained). One or more simulations may be performed to determine the expected effect of one or more other biological processes or elements in the target subject. In such health monitoring scenarios, which may also be applicable to health and wellness applications and other biological tracking use cases, one or more communication media may be utilized (e.g., one or more unmanned aerial vehicles, such as drones, high altitude pseudolites, etc.; an on-body transmission hub that serves as a sensor collection and communication hub) to monitor sensor data and communicate directly with one or more sensors, continuously or intermittently, to evaluate the collected data via one or more simulations, one or more subjects, or groups of subjects. Such communication media may include one or more sensors to provide additional information about one or more target objects (e.g., a drone deployed at an accident scene with an integrated camera that records video of the target object and the accident scene while collecting data from sensors placed on the target object). Continuous or intermittent data collection from the sensors allows one or more simulations to provide and interpret important information regarding any claim (e.g., what happened to the object, what is likely to happen based on this event, the likely cost of the claim, what is needed to cover the claim, what is needed to cover any potential future claims based on current biological data).

[0089]

[0101] In particular, at least a portion of the simulation data or one or more derivatives thereof is used to create or modify one or more health services, identifications, classifications, rates, reimbursements, or combinations thereof. Even more particularly, one or more health classifications, treatments, procedures, identifications, rates, reimbursements, or services are created, modified, or assigned, directly or indirectly, based on at least a portion of the simulation data or one or more derivatives thereof. In one variation, at least a portion of the simulation data or one or more derivatives thereof is used to create, modify, or assign one or more health treatments, services, treatments, codes, identifications, classifications, rates, reimbursements, or combinations thereof.

[0090]

[0102] In another variation, one or more simulations may be used to create or modify procedures, services, identifications, classifications, or rates (e.g., costs) associated with one or more codes for the International Statistical Classification of Diseases and Related Health Problems (ICD), the International Classification of Functioning, Disability and Health (ICF), the International Classification of Medical Procedures (ICHI), Current Procedural Terminology (CPT) for insurance and health applications (e.g., remote patient monitoring, chronic care management, transitional care management, etc.), the Healthcare Common Criteria Coding System (HCPCS), and other similar systems. For example, in one variation, a patient must have a medical need in order to receive any given medical treatment. A medical need may be any diagnosis associated with one or more individuals. In these situations, the diagnosis may receive an ICD code. ICD codes are diagnostic codes used to group and identify diseases, disorders, symptoms, adverse effects of drugs or chemicals, injuries, etc. The CPT code defines one or more treatments aligned with the diagnosis. Before receiving a treatment based on the diagnosis, one or more simulations may be performed to evaluate one or more outcomes (e.g., to determine a probability regarding the effectiveness of the treatment). Insurance companies and healthcare providers may enter into one or more contracts to determine the number of simulations to be performed, associated data to be used, rates for reimbursement of simulations for CPT codes, creation of new CPT codes based on the simulation data and the values ​​it provides, etc. Rates may be affected by one or more parameters including the number of simulations to be performed (e.g., 1 simulation vs. 10,000), the number of target subjects, the quality of the sensor data, etc. In particular, the output of one or more simulations utilizing at least a portion of the animal data utilizes the type of treatment recommended or prescribed, the length of treatment, animal data goals (e.g., including targets, goals, thresholds) during treatment (e.g., targets or objectives for one or more measurements of a target individual), etc.

[0091]

[0103] In particular, the simulation system is a remote health monitoring system. Characteristically, such a system may enable virtual medical check-ins between a patient and his / her medical professional (e.g., a system that may include audio, video, and sensor data from the patient). Simulation systems may provide entirely novel value to the medical ecosystem if one or more reimbursements may be provided for medical services, especially telemedicine services (e.g., time spent on initial setup, patient education, collecting and interpreting health status data generated remotely by the patient). By allowing the patient to access their animal data to the system to perform one or more simulations, the simulation system may provide the medical professional with context for any given condition derived from the animal data and potential future outcomes based on the one or more conditions. Such information may save time and costs (in terms of current and future medical care). In particular, at least a portion of the simulation data may be used to create or modify one or more costs or reimbursements associated with one or more medical services or procedures. For example, a reimbursement code, such as HCPCS code G2012 in the United States, which provides reimbursement to organizations for virtual check-ins, may modify or adjust its cost structure based on running one or more simulations and the output of one or more simulations (e.g., reimbursement for providing a predictive indicator for one type of animal data, such as heart rate, may be different than providing multiple separate predictive indicators for multiple types of animal data or a predictive indicator that takes into account multiple types of animal data).In this example, entirely new cost structures, codes, procedures, services, identities, classifications, and / or rates may be created based on the simulation data.

[0092]

[0104] More specifically, the simulation data may be utilized as part of one or more human connection applications or platforms (e.g., social interaction applications such as dating applications, Facebook, Instagram, or virtual-based social platforms) to create, adjust, or modify one or more predictions, probabilities, or possibilities regarding one or more human connections. The human connections may be romantic connections, physical connections, love connections, friendship connections, business connections, and the like. For example, a dating application may run one or more simulations utilizing captured animal data from two or more target individuals who have interacted with each other's profiles (e.g., photos, videos, profile information, or other forms of engagement) to predict whether the two or more target individuals are potential romantic partners. The simulation system may evaluate one or more biological measurements (e.g., physiological indicators captured from sensors such as elevated heart rate, neurochemical data, pupil response, or pupil diameter) from the two or more target individuals upon visual engagement and detect one or more variations in one or more animal data measurements indicative of attraction to connect. The strength of the connection may be determined by establishing a baseline for each target individual coupled with the degree of variation of each of the one or more types of animal data compared to one or more other visual engagements. The simulation system may generate simulation data from which insights (e.g., scores) may be made based on the strength of the connection between two or more people based on the variance of one or more animal data measurements. In particular, a user may purchase one or more simulations to determine the strength of the connection. Even more particularly, a user (e.g., a female subject) who grants access to the data of one or more simulations may receive a portion of the consideration for allowing the simulation system to access their data (e.g., if one or more male subjects express an interest in knowing if there is a biologically based connection).

[0093]

[0105] More specifically, the simulation data incorporating at least a portion of the animal data may be utilized to create one or more products (e.g., prop bets or sports betting markets) for one or more simulated events. For example, in connection with a sports competition, if the system collects heart rate data for Team A when playing against Team B in a real sports competition, and other "Team A vs. Team B" based data (e.g., non-animal database results of previously played Team A vs. Team B competitions), the system may be operable to create one or more new bets utilizing at least a portion of the previously collected data, including at least a portion of the animal data, that was incorporated as part of one or more simulations (e.g., when utilizing at least a portion of the animal data for Team A and / or Team B as part of one or more simulated matches, a bet may be "whether Team A's average maximum heart rate exceeds 170 beats per minute during the duration of the 10,000 simulated matches against Team B" or "whether Team A wins the match more than 80% of the time when playing against Team B in the 10,000 simulated matches"). For purposes of the presently disclosed and claimed subject matter, "previously collected data" may include "current data," including data or data sets currently being collected in a current scenario (e.g., including real-time or near real-time data collected in any current scenario). More specifically, the simulation data incorporating at least a portion of the animal data may be utilized to create one or more new products for one or more virtual / simulated events. For purposes of the presently disclosed and claimed subject matter, virtual events (and subject matter) may be used interchangeably with simulated events (and subject matter), and vice versa, to describe the application of the simulation data. References to virtual events or simulated events are meant to be exemplary and inclusive of all possible systems that may utilize artificial animal data. The products may include prop bets or productized information utilized as part of a wager, or risk mitigation strategies for virtual betting.For example, if the system collects respiration rates of one or more real subjects (e.g., real horses) in one or more real races, the system may generate simulated data (e.g., simulated respiration rates) for one or more simulated races based on the collected real-world respiration rate data that represents (at least for modeling purposes) the respiration rates of real-world horses, allowing the system to create one or more prop bets or betting products for one or more virtual subjects (e.g., virtual horses) that utilize at least a portion of the simulated animal data generated in one or more virtual races (e.g., bets could be: "will a virtual horse's maximum respiration rate in a virtual race exceed index X"; "will virtual horse Y have a higher maximum respiration rate than virtual horse Z"; "will virtual horse Z win the race" when at least a portion of the animal data or one or more derivatives thereof are utilized as part of a virtual race, or when simulated data indexes are utilized to derive from the animal data). In one variation, the simulated data generated by the simulation system may not share the same one or more characteristics from which it is derived. For example, when generating artificial data (e.g., respiratory rate) for a virtual object (e.g., a virtual horse), the simulation may characterize and display the generated respiratory rate as another index (e.g., color, another name such as "fatigue"). More specifically, one or more virtual objects share at least one common characteristic with one or more real objects, and a virtual event shares at least one common characteristic with the event from which the real animal data was collected (e.g., horse Z is running in a real race and virtual horse Z is running in a virtual race, with at least one characteristic of the real horses and events in the system. The characteristic may be, for example, respiratory rate, and the event may be a horse race. Bet: "Will virtual horse Z's maximum respiratory rate in the virtual race exceed index X?"). The characteristics of the objects may include one or more biological characteristics, physical characteristics, profile characteristics (e.g., the same name, jersey number, team name, team color), etc.In another example, a simulation (e.g., video game, virtual world video game) in which one or more subjects can participate may create one or more bets or products (e.g., in-game virtual products to purchase) related to the real animal data of one or more users playing the game (e.g., utilizing the real animal data of the users incorporated as part of the virtual video game, using simulated data incorporating at least a portion of the in-game animal data or other animal data and creating in-game rewards for users who reach goals; allowing users to purchase artificial database virtual products generated at least in part from the animal data; creating bet types or products based on the artificial animal data utilized in the video game). In yet another detail, simulated data is created for a simulation (e.g., virtual event, video game) based at least in part on the animal data that may create one or more novel values. For example, in the above scenario, a user may want to know the probability of horse Z winning a race in a simulated event when the simulated respiration rate exceeds metric X, and how frequently this occurs in any given simulated race. The system may generate the simulation data utilizing a variety of data including at least a portion of the animal data (e.g., real-world respiration rate of horse Z collected from one or more source sensors for all available races; respiration rates of other similar horses racing in similar conditions from one or more source sensors, if available; respiration rates of other similar horses racing in different conditions from one or more source sensors, if available; respiration rates of other different horses racing in similar conditions from one or more source sensors, if available; respiration rates of other different horses racing in different conditions from one or more source sensors, if available; other comparison animal data (e.g., heart rate) collected from one or more source sensors for similar and different horses in similar and different conditions; simulated respiration rate data generated from one or more simulated races; other factors collected in the real world that may be used as input for the simulated races - environmental conditions such as weather or temperature, injuries, biofluid data, position-based data such as speed and acceleration data, training data, etc.).The simulated animal data may be further utilized in one or more simulations to generate one or more probabilities or predictions regarding the virtual representation of Horse Z winning a virtual race. Based on the information, a user may place a bet on the virtual horse race.

[0094]

[0106] More specifically, the data for a simulated subject in a simulation system may be composed of data from multiple real-world animals. For example, in the context of horse racing, if the simulation system is characterizing the heart rate of a virtual horse, the heart rates may be derived from multiple real-world horses that collectively make up the virtual horse.

[0095]

[0107] In yet another detail, simulation data incorporating at least one type of animal data may be utilized to more accurately create or adjust odds (e.g., betting lines). For example, if a line is set between player A and player B in a particular real-world match, the computing device may run one or more simulations using a variety of similar match conditions (e.g., on-court environmental data, current score, current statistics, previous win / loss records, previous head-to-head statistics), and one or more animal data inputs (e.g., all animal data of player A vs. player B including current match data, animal data of past player A vs. player B head-to-head matches, all animal data of players A and B under similar environmental conditions, pre-match training data of players A and B, injury data, etc.), and simulation data inputs (e.g., simulated heart rate and related physiological metrics for player A for the remainder of the match, simulated heart and related physiological metrics for player B for the remainder of the match), allowing the system to more accurately determine the probability of the outcome. In one variation, by utilizing one or more artificial intelligence techniques, such as machine learning techniques, the system may analyze previously collected data sets and current data sets to create, modify or enhance one or more probabilities, possibilities or predictions. The one or more data sets may include at least a portion of the simulation data. When a machine learning based system is configured to learn from collected data rather than explicit program instructions, the ability to search for and recognize patterns that may be hidden within the one or more data sets allows the machine learning based system to uncover insights from the collected data that allow predictions to be made. Advantageously, because machine learning based systems learn using data, they often take an iterative approach to not only improve the predictions and accuracy of the model as new data is input into the system, but also to improve the predictions and accuracy of the model derived from feedback provided from previous calculations made by the system (which also allows for the production of reliable and reproducible results).In such a scenario, new animal data, such as new biosensor data, entering the system from any given subject at any given time may allow new simulations to be run and new correlations to be made based on a broader set of data. For example, with data entering the system point by point (or shorter time periods) for a match between Player A and Player B, the simulation system may be able to more accurately predict future animal data measurements and correlate these measurements with other data in the system to more accurately predict outcomes. In one variation, probabilities or predictions may then be derived and utilized to create or adjust one or more odds, which may occur in real time or near real time.

[0096]

[0108] In yet another detail, simulation data is created for a simulation (e.g., a virtual event) based at least in part on the animal data that may create new value for the betting system. For example, a user may want to know the probability of a virtual subject (e.g., virtual horse #3) winning a virtual race in a simulation when its virtual respiration rate exceeds indicator X (the aggregate of real-world horse data from multiple horses may constitute virtual horse #3, and may characterize actual data from races performed by one or more real-world equivalents of horse #3). Probability-based or predictive data sets may be packaged to create one or more novel betting products that users can obtain to develop strategies and / or place bets on virtual horse races. Similarly, an insurance company may want to know the probability of an event occurring to a particular group of target individuals (e.g., target individuals with a particular heart condition, or target individuals with one or more positive / negative social habits that affect the animal data measurements) and create products specifically tailored to such target individuals to lower their bets. In another example, one or more artificial data sets created based on actual animal data from a target subject (and in some variations, actual animal data from other subjects) may be modified using the simulation system 10 to introduce deviations in the data corresponding to any given characteristic of the target subject (e.g., fatigue, rapid heart rate changes). With the ability to vary, change or adjust one or more parameters or variables to generate modified data sets, one or more simulations may be run to ascertain how the target subject would perform based on changes in one or more parameters or variables (e.g., high stress situations, high altitude environments, extreme heat, extreme movement or travel), and the simulation system 10 may establish patterns between body metrics (e.g., heart rate, respiration, etc.), the one or more parameters / variables, and the likelihood of an event occurring (e.g., winning a particular match), allowing the simulation system to calculate one or more probabilities for certain conditional scenarios (e.g., what-if scenarios) based on changes in any given parameters / variables.In one variation, the creation, adjustment or enhancement of one or more products (e.g., wagers, insurance products, analytics packages for health monitoring platforms) and / or odds may be generated in real-time or near real-time based on adjustments to the simulation data as events occur. For example, in the context of sports, the simulation system may run one or more simulations of a current match between players A and B, with new data being input into the system in real-time or near real-time in a real-time or near real-time simulation, thereby providing a value or set of values ​​to which one or more probabilities, likelihoods or predictions may be assigned regarding any given or desired outcome occurring. For example, if the outcome under analysis is "will player A's heart rate reach 200 beats per minute in the current match" or "will player A win the match against player B," the system may run one or more simulations that utilize at least a portion of the collected animal data to generate probabilities or predictions regarding the outcomes that will occur. Examples of simulations that the system can run include: (a) Player B wins the first set in a longer time than expected and Player A starts to feel stressed based on cardiac derived data; the user is interested in ascertaining Player A's potential outcome in the second set based on collected data (current and previously collected animal and non-animal data); (b) the environmental temperature rises 15 degrees from the start of the first set of the match to the end of the first set and high temperatures start to affect Player A's distance covered as Player A's fatigue data indicates that Player A is fatigued; the user is interested in ascertaining how far Player A is expected to run based on his current fatigue level, the expected fatigue level based on the likely match outcome, the current environmental temperature, and the expected environmental temperature during the match. There can be n such simulation scenarios, and furthermore, there can be one or more simulation scenarios that are created on the fly by the system's artificial intelligence engine based on similar matches in the past. Once all these simulations are run, the output is collected and analyzed to provide a probability or prediction of the outcome under investigation.

[0097]

[0109] In another method for creating one or more simulation data sets, previously captured data or previously created simulation data are re-run through one or more simulations to create one or more new data sets. In this example, existing data (e.g., real animal data and simulation data) may be used as a baseline to determine the probability, likelihood, or prediction for a particular outcome through one or more simulations (e.g., a simulation run n times - e.g., 10,000 times - with the same one or more data inputs to determine the likelihood of an event). In yet another method for creating simulation data, one or more new variables or parameters may be applied to existing data to create a new data set. More specifically, existing data with one or more randomized variables are re-run through one or more simulations to create a new data set not previously seen by the system. For example, when a simulation system has a data set of a target individual (e.g., an athlete) and a target event (e.g., a game played by the target individual), the system may be operable to recreate and / or vary one or more parameters or variables (e.g., altitude, on-court temperature, humidity) in the data set and re-run the one or more events through one or more simulations to generate a target simulation data output. For example, in the context of tennis, an acquirer may need one hour of heart rate data for Player A when the temperature is 95 degrees Fahrenheit or higher throughout a two hour match. The system may have one or more sets of heart rate data at various temperatures (e.g., 85°F, 91°F, 78°F) and inputs as described herein for target subjects for Player A in similar conditions, and for Player A in similar conditions along with other similar and different athletes in similar and different conditions. Since no heart rate data has been collected for Player A at 95 degrees or higher, the system may run one or more simulations to create artificial data, and then utilize that data in one or more additional simulations. In particular, the system may be operable to combine different data sets to create or recreate one or more new data sets.For example, a user may require heart rate data for Player A for an hour when temperatures are above 95 degrees throughout a two hour match of a particular tournament, where one or more characteristics, such as altitude or humidity, may affect performance. Although this data has never been collected in its entirety, various data sets may be identified by the simulation system that comprise at least a portion of the requested data and that feature one or more desired parameters / variables (e.g., one or more data sets from Player A that characterize heart rate, one or more data sets from Player A that characterize playing tennis in temperatures above 95 degrees, one or more data sets at the required tournament with the required characteristics, such as altitude). With the simulation system operable to identify these required parameters within and across data sets and trained to understand the impact of the one or more parameters / variables on the collected animal data and associated outcomes, the simulation system may run one or more simulations to create one or more new artificial data sets (which may be, for example, predictive indicators, calculated assets or artificial animal data) based on these different data sets that meet the user's requirements. In one variation, the different sets of data used to create or recreate the one or more new datasets may characterize one or more different subjects who share at least one common characteristic with the target subject (which may include, for example, age range, weight range, height range, gender, similar or different biological characteristics, habits, sensor measurements, etc.). Using the above example, heart rate data may be utilized for player A, but the system may utilize another one or more datasets from players B, C, or D, selected based on their relevance to the desired dataset (e.g., some or all of the players may exhibit similar heart rate patterns as player A; some or all of the players may have similar measurements of biological fluid derivation as player A; some or all may have a very similar playing style as player A; some or all of the players may have a dataset collected by the system that characterizes tennis played in temperatures above 95 degrees).These one or more data sets may serve as one or more inputs within one or more simulations to more accurately generate future biological measurements (e.g., heart rate) of Player A under desired conditions.

[0098]

[0110] Yet another method of simulating data is to create an artificial data set that is general in nature (e.g., a data set that lacks a predefined selection of one or more desired biological characteristics). In one variation, one or more randomized data sets are created, with one or more variables selected by the system rather than by the acquirer. This is particularly useful, for example, when an insurance company is looking for a particular data set (e.g., 1,000,000 smokers) among a random sample (e.g., no defined age or medical history that can be randomly selected by the system), or when a betting company is looking to create one or more new markets (e.g., prop bets) for an event that did not exist (e.g., prop bets centered around a video game simulation outcome). In particular, one or more artificial data sets are created based on a predefined number of individuals selected by a given user of the system. More particularly, one or more artificial data sets are created from a predefined number of individuals randomly selected by the system.

[0099]

[0111] In particular, one or more tags are assigned to the artificial data based on one or more characteristics from the associated animal data. Characteristics may include one or more sources of the animal data, specific individual attributes of one or more individuals or groups of individuals from which the animal data is derived (e.g., name, weight, height, corresponding identification number or reference number), type of sensor used, sensor characteristics, classification, specific sensor configuration, location, activity, data format, type of data, algorithm used, quality of data, date and time the data was collected, associated organization, associated event (e.g., simulation, real world), latency information (e.g., speed at which data is provided), etc. It should be understood that any single characteristic regarding the animal data from which the simulated data is derived (including, for example, any characteristic regarding the data, one or more sensors, and one or more target individuals) may be assigned or associated with one or more tags. Characteristically, one or more tags associated with the animal data may contribute to creating or adjusting the associated value of the artificial data. In particular, one or more neural networks may be trained to assign one or more tags to one or more types of simulated data and simulated data sets.

[0100]

[0112] In further detail, the simulated data may be assigned to one or more classifications. Classifications (e.g., including groups) may be created (e.g., as one or more searchable tags) to simplify the search process for data acquirers and may be based on data collection processes, practices, quality or associations, as well as target individual and simulated target individual characteristics. Classifications may be identifiers for the data. For example, one or more classifications may be assigned to an artificial data set derived from and / or representing ECG data (with one or more classifications including "simulated ECG data", "simulated ECG data of target individual Z", "simulated ECG data of a male aged 25-34", "simulated ECG data from sensor C", "simulated ECG data from n simulations", "simulated ECG data from target individual of team Y", "simulated ECG data from match X", "simulated ECG data from target individual of team Y in match X", etc.). Another classification may be assigned to an artificial data set representing ECG data from a particular sensor according to a particular data collection methodology in a particular setting. In another example, the classification may be made for a dataset representing a target individual who has previously experienced a stroke, or for a simulated dataset representing a simulated target individual that is based at least in part on a real-world target individual.Examples of classifications or tags include metric classifications (e.g., a characteristic of a simulated subject captured by one or more sensors that can be assigned a numerical value, such as heart rate, hydration, etc.), a personal classification of a simulated target individual (e.g., age, weight, height, medical history), an insight classification of a simulated target individual (e.g., "stress," "energy level," a score indicating the likelihood of one or more outcomes occurring), a sensor classification (e.g., sensor type, sensor brand, sampling rate, other sensor settings), a classification of simulation data characteristics (e.g., raw data or processed data), a classification of simulation data quality (e.g., good data and bad data based on defined criteria), a classification of simulation data timeliness (e.g., providing data within milliseconds vs. hours), a classification of simulation data context (e.g., NBA Finals games vs. NBA preseason games), a classification of simulation data ranges (e.g., providing a range of data, such as bilirubin levels between 0.2 and 1.2 mg / dL), a classification associated with a simulation system (e.g., the number of data sets trained by one or more neural networks, the type of neural network utilized to generate the simulation data), etc. In one variation, one or more classifications may be assigned based on what the artificial data represents, including one or more of tissue, sensor type, sensor parameter, data type, data quality, timestamp, location, activity, target individuals, groupings of target individuals, data measurements, etc. Characteristically, one or more classifications associated with the animal data and / or the simulation system may contribute to creating or adjusting values ​​associated with the artificial data. In particular, one or more neural networks may be trained to assign one or more classifications to one or more simulation data sets.

[0101]

[0113] In one variation, some classifications of simulated data may have more value than other classifications. For example, simulated heart rate data from people ages 25-34 may have less value than simulated glucose data from people ages 25-34. Differences in value may result from a variety of reasons, including scarcity of the type of data used in one or more simulations (e.g., on average, real glucose data is harder to collect than real heart rate data, and therefore not as readily available or collectable), the quality of the real data coming from any given sensor to be used in one or more simulations (e.g., one sensor may provide better quality data than another sensor), the individual or individuals from which the real data originates compared to any other given individual (e.g., one individual's data may be more valuable than another individual's data based on one or more unique characteristics of that individual, which may or may not be biological in nature), the type of real data (e.g., data from a group of individuals with a particular biological characteristic from sensor X, given that AFE data may provide an opportunity for additional non-ECG insights to be derived, including surface EMG data), and the quality of the real data coming from any given sensor to be used in one or more simulations (e.g., one sensor may provide better quality data than another sensor), the type of real data coming from a group of individuals with a particular biological characteristic from sensor X, given that AFE data may provide an opportunity for additional non-ECG insights to be derived, including surface EMG data, and the quality of the ... These may include raw AFE data from which ECG data can be derived may be more valuable than derived ECG data alone from the same group of individuals with the same biological characteristic from the same sensor X), the derived use case for the data (e.g., glucose data may also be used to derive hydration information, which is more valuable because it is a more difficult data type to collect than heart rate based data), the amount or volume of data (e.g., one year of daily heart rate data from 100 people ages 45-54 may be more valuable than one month of daily heart rate data from the same 100 people ages 45-54), the context in which the data was collected (e.g., respiration rate data collected from a target individual in a culminating sports competition compared to a training session, or respiration rate data collected from a target individual with a life-threatening infectious respiratory disease versus when the target individual does not have a respiratory related illness), etc.

[0102]

[0114] In further detail, one or more classifications have one or more corresponding values ​​that are created, assigned, modified, and / or enhanced by the simulation system. It should be understood that one or more classifications may have a predefined value, an evolving value, or a dynamic value, or a combination thereof. For example, a classification for a given type of simulation data may increase in value as more relevant data is added to the simulation system, as more data in the classification becomes available, or as demand for simulation data from a data set associated with one or more particular classifications increases. Conversely, the value of the simulation data may decrease as time passes from when the data was created (e.g., the value of the simulation data for creating predictive indicators regarding outcomes may be significantly more valuable before an outcome occurs rather than after the outcome has already occurred), as the data becomes less relevant (e.g., as new sensors that capture more precise and specific information become available), or as demand for data from that particular classification decreases. Depending on the use case, multiple values ​​may be created, assigned, modified and / or enhanced for the same classification (e.g., in the case of sports betting, a classification in one market may be more or less valuable than the same classification in another market). In more detail, one or more classifications may dynamically change and one or more new categories may be created or modified based on one or more requirements or the input of new information or sources into the system. For example, a new type of sensor may be developed, a sensor may be updated with new firmware that provides the sensor with new settings and capabilities, or one or more new data types (e.g., data types derived from biological fluids) may be introduced into the system. In more detail, one or more neural networks may be trained to create, assign, modify and / or enhance one or more monetary and non-monetary based values ​​to one or more animal data sets, including one or more simulation data sets. In more detail, one or more artificial intelligence techniques (e.g., machine learning, deep learning techniques) may be utilized to dynamically assign one or more classifications and / or values ​​to one or more data sets.

[0103]

[0115] Systems for generating simulation data and for running one or more simulations using at least a portion of the animal data are applied in various industries. For example, in the context of real-world fitness or wellness systems including personalized / group fitness classes (e.g., cycling, CrossFit, remote home fitness platforms), the simulation data may be used to inform one or more users regarding their real-time or near real-time biological state (e.g., physical state such as current or predicted "energy level") and provide insight into future outcomes that may occur based on one or more actions (e.g., based on current animal data measurements, past animal data, and current activity of running at p miles per hour, a target fitness individual X is predicted to reach a desired energy expenditure in n minutes, s seconds). Furthermore, the simulation system may be used to gamify fitness / wellness classes, making biological data available for integration into simulation games as well as making simulation data available within the games. In one example, one or more users may utilize one or more sensors to provide at least a portion of their animal data (e.g., heart rate data) to the simulation system to obtain a competitive advantage or other compensation compared to other users in the class (e.g., sensor data is converted into a format that is input into the simulation, the simulation is performed with the users participating in the simulation, the animal data incorporated into the simulation allows for benefits (e.g., monetary or non-monetary value) such as more rest periods, free classes based on physiological-based success metrics within the simulation, or prizes based on maximum "energy" exerted in a class, etc.). Advantageously, at least a portion of the animal data is simulation data. Comparative biological metrics, which may include simulation metrics, may be visually displayed for each user to determine who is performing best in any given class.In another example, one or more hardware components of a fitness machine (e.g., a treadmill, a cycling machine) or a fitness display (e.g., a computing device that displays fitness content such as a television or an interactive mirror) may communicate with one or more sensors of a subject that aggregates all sensor data into a single application from which one or more simulations may be performed to generate simulated data that conveys information about the subject's current and future biological state and generate one or more predictions or probabilities regarding animal data. In another example, a fitness machine with an integrated display or interactive computing device that streams fitness content may collect biosensor data and provide its one or more users prior to a workout with one or more biological insights derived from the simulated data (e.g., a predicted "fatigue" level or expected calorie expenditure based on current biological measurements, previously collected animal data, and the predicted duration and intensity of a given workout). In particular, a fitness instructor (e.g., real, virtual AI-based) or a "smart" fitness device (e.g., a device having one or more computing devices or a computing device having a display featuring customized AI-generated fitness content) may perform one or more actions that may adjust the workout based on simulation data derived from collected animal data (e.g., if the instructor or "smart" device predicts that the target subject will not exert enough energy at the end of the workout based on current parameters and predicted power output, or predicts that the target subject is on pace to exert too much energy by the end of the workout based on simulation data derived from the subject's data, such as heart rate or derivatives of performance zones, the instructor or "smart" device may adjust the difficulty or speed to increase or decrease the difficulty of the workout for the user). The one or more actions may be derived, created, modified, or adjusted via one or more artificial intelligence techniques.In particular, one or more users may receive compensation (e.g., money, cryptocurrency, gift cards, free analysis, free classes) for allowing third parties to access their animal data or one or more derivatives thereof (e.g., including simulated animal data utilizing at least a portion of their animal data). Users may have the ability to opt in or out before, during, or after a workout.

[0104]

[0116] In particular, artificial animal data is created for one or more simulation systems (e.g., virtual / simulated events, video games, simulators) for user involvement. In one variation, the computer software follows through on an event or occurrence, including a race, contest, study, etc. Advanced algorithms typically use random number generators to determine the outcome. In particular, a neural network or multiple neural networks may be utilized to determine the outcome, possibly taking into account the skill and state of the participants, as well as the luck factor inherent in real-world events (e.g., sporting events). The artificial animal data may be created by running one or more simulations based at least in part on the actual animal data. In another variation, the computer software utilizes data derived from at least a portion of the actual animal data in the simulation system to enable direct or indirect forms of user involvement (e.g., participation) in the simulation. The one or more simulation systems may include game-based simulation systems (e.g., video game systems that simulate an event, such as playing a sport, and allow for user or multi-user participation; sports betting simulators that allow users to place one or more bets on a virtual event, such as betting on a virtual horse race; video game systems that allow for simulation data related products to be obtained within the simulation, such as simulation database virtual products that can be purchased), simulators, and other systems (e.g., military simulators, virtual reality systems, medical simulators that utilize augmented reality systems, mixed reality systems, extended reality systems, etc.). In particular, game-based systems engage one or more users via at least one of a virtual reality system, an augmented reality system, a mixed reality system, or an extended reality system (e.g., a game-based military simulator that utilizes a virtual reality system).One or more users may participate in the simulation utilizing a variety of computing devices including traditional video game consoles, personal computers, mobile phones, tablets, terminals, virtual reality systems, augmented reality systems, etc. The environments in which users participate in the simulation system may include traditional video game environments (e.g., PC games, handheld games, mobile games, online console games, online games including web browser-based games), virtual reality systems, augmented reality systems, mixed reality systems, extended reality systems, etc. The simulation system that engages the users may have one or more hardware components (e.g., game controllers, game keyboards, game headsets) associated with the simulation system and one or more biometric sensors (e.g., game controllers, game headsets, game keyboards, seat sensors, camera sensors, other gaming sensors) embedded within the hardware associated with the simulation system, or may utilize one or more sensors (e.g., smart watches or on-body sensors that capture biometric data) from one or more users that communicate with the simulation system and are integrated as part of the simulation. Examples include hand and finger pressure sensors located in a simulated game controller (e.g., to see how tightly the controller is being held), ECG / heart rate sensors to monitor the heart rate of a player participating in the simulated game, EEG sensors located in a headset utilized as part of the simulated game, motion sensors embedded in the seats or controllers of players participating in the simulated game, sensors embedded in bicycles to measure power output or wattage based on pedal movement as part of the simulated game, and sensors to measure reaction times of players participating in the simulated game. Advantageously, direct communication between one or more sensors and the simulation (e.g., a video game) may occur via a web browser.Additional details regarding a system capable of communicating directly with sensors via a web browser are disclosed in U.S. Patent Application No. 16 / 274,701, filed February 13, 2019, and International Application No. PCT / US20 / 18063, filed February 13, 2020, the entire disclosures of which are incorporated herein by reference. In particular, the simulation system may integrate and display one or more simulated measurements derived from at least a portion of one or more sensors that collect information from one or more users. For example, participants in a group fitness class may compete against each other in a simulation game with one or more simulated measurements derived from at least a portion of the animal data incorporated within the game. In another example, a medical professional may view a display of a patient's real-time sensor data in an augmented reality system, where the real-time sensor data may include simulated data that provides a real-time probability of the patient experiencing a medical event while undergoing one or more procedures. In one variation, the simulation system may utilize simulation data (e.g., in the case of a medical professional, the medical professional's own derived simulation data or simulation data derived from one or more virtual patients) to train one or more medical professionals to fine-tune a skill or skills. In yet another example, an insurance company may allow users to participate in a simulation game (e.g., utilizing one or more sensors in a simulation that provide a predictive indicator such as a personal risk score based on the simulation data) in which the simulation data allows one or more premiums to be derived from at least a portion of the user's sensor data, whereby one or more premiums may be created, modified, or enhanced. In yet another example, a medical platform (e.g., a telemedicine application) may use a simulation system that utilizes simulation data derived from one or more sensor measurements of a patient to enable a virtual medical examination or consultation from which an outcome from the simulation may result (e.g., a patient may receive a score or other indicator based on simulation data generated from at least a portion of the animal data).Characteristically, the one or more simulated measurements may include one or more non-animal data measurements as one or more inputs. More specifically, the video game or game-based system may create one or more novel data types for characters or objects in the game based at least in part on the actual sensor data provided by the user or one or more derivatives thereof. For example, the simulation game may create novel indexes for objects in the game based on insights derived from at least in part on the captured real-world sensor data or animal data such as fatigue level, heart rate, reaction time, controller pressure, etc. of the real-world object. The simulated data utilized in the game may be animal data converted into a format to be input into the simulation, or animal data generated by running one or more simulations utilizing at least in part on the actual sensor data or one or more derivatives thereof to create artificial data. Providing one or more measurements to the simulation system (e.g., the video game or game-based system) and generating the artificial data may all occur in real time or near real time.

[0105]

[0117] In one variation, simulation system-based data, which may be derived from one or more simulations and / or may be artificially created based at least in part on animal data, may be associated with or related to one or more characters (e.g., animals) that are characterized as part of the simulation (e.g., game). The characters may be based on animals that exist in real life (e.g., a real-world professional soccer player may have a character that portrays that particular professional soccer player in a soccer video game), or may be artificially created and may be based on or share one or more characteristics of one or more real animals (e.g., simulated soccer players in a game may share a jersey number, jersey color, or biological characteristics that are recognizable as a human soccer player). The system may enable users of a game-based system (e.g., a video game system) to purchase artificial data that utilizes at least a portion of the real data in the game. In particular, the animal data purchased in the game may be artificial animal data that may be based at least in part on real animal data and may be generated via one or more simulations, or may be converted animal data in a form that can be input into the simulation (e.g., game). This data may be utilized, for example, as an indicator of events in the game. For example, a user playing a game may be able to play against a simulated version of a real-world athlete in a game that utilizes the athlete's "real-world data," which may include the athlete's real-world biological data or one or more derivatives thereof, converted into a form that can be input into the simulation. This may mean, for example, that the real-world athlete's "energy level" data collected over time is integrated into the game. In one particular example, as a match in a video game gets longer or as simulated athletes run distances in a video game, their "energy level" in the video game may be adjusted and influenced based on the actual data collected for the actual athlete. The real-world data may indicate the athlete's fatigue range based on the distance run or length of any given match.This information may then be utilized by the simulation system to adjust the "energy level" within the game. This data may be utilized, for example, to obtain an advantage within the game. In the context of a sports video game, the types of animal data that may be purchased within the game may include the ability to run faster, jump higher, have a longer energy life, hit the ball farther, etc. FIG. 10 illustrates an example of a video game in which a user may purchase derivatives of animal data (e.g., artificially generated animal data such as "energy levels") based in part on real animal data to provide an advantage to the user of the video game (e.g., increased energy levels that increase the likelihood of winning the game). In another example, in-game artificial data derived from and sharing at least one characteristic with the animal data may provide one or more subjects within the game with one or more special powers that may be derived from one or more simulations. More specifically, a user may be able to purchase real-world animal data that has been converted into a format that can be input into and incorporated into a simulation.

[0106]

[0118] In particular, the simulation system converts at least a portion of the actual animal data (e.g., physiological data) into artificial data representing one or more insights, calculated assets, or predictive indicators that are utilized within the game. The insights, calculated assets, or predictive indicators are based on the game's conversion of the animal data into artificial data converted into a form that can be input into the simulation, and the simulation system's interpretation of the data in a manner that allows the simulation (e.g., game) to provide advantages or penalties to one or more users (e.g., an "energy bar" provided in a fight sports game that utilizes real-world sensor data to more accurately predict fatigue of animals such as humans). To perform the conversion and interpretation, the game runs one or more simulations to determine the appropriate advantage or penalties to be provided.

[0107]

[0119] More specifically, the artificial animal data, including at least a portion of the real animal data, is provided to create one or more new markets (e.g., prop bets / wagering) for people to place one or more bets (including virtual bets) or as supplemental information for one or more bets. For example, one or more bets may be based on biological data (e.g., is player A's heart rate in a live tennis match greater than 180 bpm in the first set of match X, which may be a proposed prop bet within any given betting system), and the artificial data may be used as supporting information for one or more real bets (e.g., one or more simulations may be run to predict the likelihood that player A's heart rate will be greater than 180 bpm in the first set of match X in n simulations). In another example, if the bet is "will player A win a match against player B" in a live real-world match, one or more simulations using at least a portion of the animal data may be run to generate one or more probabilities or predictions regarding one or more outcomes. In one variation, the synthetic data may be information on which a bet is made on one or more real-world events (e.g., will player A's "energy level" reach less than n% in the first set of live match X). In another variation, the synthetic data may be utilized as supporting information for one or more bets in a simulation (e.g., a bet "will player A win a match against player B in a simulated match" may be supported by synthetic data generated to provide probability-based information regarding the outcome of the simulated match). In another variation, the synthetic data may be one or more bets in a simulation (e.g., player A's simulated heart rate exceeds 180 bpm in the first set of simulated match X). Advantageously, one or more bets on the synthetic data may occur within one or more simulation systems (e.g., virtual reality, mixed reality, etc.).The one or more bets may be based on biological data (e.g., Player A's heart rate in a live tennis match displayed in a virtual or augmented reality system exceeds 180 bpm in the first set of match X, which may be a prop bet offered in any given betting system) or derivatives (e.g., a simulated "energy level" of a virtual object in a soccer video game will fall below 40% in the first half, where the simulated "energy level" is derived from at least a portion of the biosensor data generated from one or more simulations from a real player or object). The simulated data generated from at least a portion of the biosensor data provided directly to the system may be used to understand the probability of any given outcome occurring and provide one or more predictive indicators via one or more simulations. For example, a bettor may have the opportunity to purchase a simulated "energy level" of Player A in the last 10 minutes of a match in a real match or a simulated game (e.g., a video game) to determine whether Player A will win the match, where one or more simulations are run and one or more artificial intelligence techniques are utilized to recognize patterns in the data (e.g., machine learning techniques) and predict the outcome. Artificial data generated from at least a portion of the biosensor data may be used to influence the outcome of a particular bet (e.g., by providing an advantage or disadvantage to one or more users in a game) or events within a simulated game. For example, a bettor may purchase more virtual "energy" for virtual player A in a video game to increase the likelihood that player A will win the simulated game.

[0108]

[0120] In particular, the simulation data may be used within a simulation system (e.g., a virtual reality system, an extended reality system) and within a portion of a simulation within the system. For example, in the context of medicine, a virtual reality ("VR") system may be utilized to replicate real-life medical procedures. More specifically, a simulation system such as a VR system may provide a representation of an actual medical scenario for various use cases including practicing, evaluating, learning, testing, or understanding biological systems, processes, or human actions. In this regard, simulated animal data may be generated and incorporated by the system (e.g., to represent one or more biological measurements or vitals of a patient within the virtual reality system, which may include simulated ECG, respiration rate, or biofluid data), and the simulated animal data may be changed or modified based on one or more actions of a user of the system (e.g., a doctor provides a medicine injection to a virtual patient, causing one or more biological measurements, such as a heart rate measurement of the virtual patient, to change). However, simulation data may be incorporated to represent biological data derived by the user (e.g., the user's animal data is input into the simulation to convert the user's sensor data measurements into a displayable format. For example, in this scenario the user / doctor performs surgery within the virtual reality system and displays indicators of the doctor's "stress", "nervousness", etc. based on abnormal biomechanical movements such as head movements or hand tremors derived from one or more sensors that may be integrated into one or more doctor-held objects during surgery). In another VR example, simulation data may be incorporated to manifest future outcomes. For example, a user (e.g., doctor) within a medical simulator may have predicted animal data-related outcomes derived from his / her animal data or that of a virtual subject (e.g., patient) that are based at least in part on actual animal data from one or more real-world subjects based on any given or potential actions performed within the system. In some cases, a system such as a VR system may include a live operator operating the simulation system from which scenarios within the virtual reality system may be altered or modified.In such cases, the simulated animal data utilized within the simulation system may be altered or modified based on alterations or modifications made by one or more operators. In particular, the one or more operators within the simulation system may be simulation operators, such as in the case of constructive simulation.

[0109]

[0121] In another variation, animal data may be utilized to affect outcomes or gain a competitive advantage within a simulation (e.g., gaming) system. In this variation, the system integrates users involved in the simulation system (e.g., playing a video game) and their animal data (e.g., physiological data) into the game itself. More specifically, if the gaming system utilizes real-world people or characters that share one or more characteristics of one or more real-world people, the system may utilize at least a portion of the person's real-world animal data or synthetic data based in part on the actual animal data to affect outcomes or provide the ability to affect outcomes through in-game purchases, acquisitions, or achievements of any simulation games played (e.g., sports video games, online virtual world games, group fitness competitions). For example, if a user has high real-world biological measurements (e.g., stress levels or high heart rates) compared to other users playing similar games or compared to the user's baseline biological measurements, one or more virtual subjects in the game may also experience a similar data-related response (e.g., high stress levels, high heart rates), which may provide an advantage, disadvantage, or other indication to the user and / or one or more subjects in the game. Advantageously, this may occur in real-time or near real-time. The advantage, disadvantage or other indication may be immediate and / or of a particular duration. Depending on the game, advantages may include bonus points, additional intensity, access to easier levels of resistance of the fitness equipment being utilized (e.g., in a cycling class competing against other subjects), and the like. Disadvantages within the game may include loss of points, reduced energy levels, increased resistance applied to the fitness equipment being utilized by the subject (e.g., bicycle pedals in a cycling class competing against other subjects), and the like. Similarly, the display of the user's various biology-based animal data measurements may include a viewable portal that provides the user's various biology-based animal data measurements within the game.Use cases may include flight simulations, military simulators, medical simulators, high frequency trading simulation programs that monitor how traders react in any given situation on financial markets, sports video games, fitness classes, wellness simulators (including behavioral health), and the like. As an example, if a user is playing a web browser based shooter game, the stability of the shooter's zoom lens in the game may be reduced by indicating actual stress or elevated heart rate. In another example, a virtual body in which a surgeon is operating may provide an indicator (e.g., change from one color to another) if the surgeon's heart rate, stress level, or biomechanical movement (e.g., hands) indicates something is wrong with the surgeon (e.g., the surgeon is stressed or the surgeon's hands are unsteady). On the other hand, indicating peak biological activity (e.g., steady hands and a steady heart rate) may provide an advantage in the game to the user and their corresponding virtual character or subject (e.g., shooter) in the game. These biologically-based animal measurements (e.g., real-time heart rate) may be displayed by one or more opponents or third parties, upon which tactics may be developed to put the opponent at a disadvantage (e.g., to increase the opponent's real-time heart rate and weaken the opponent in some way within the game), feedback may be provided, and rewards or other consideration (e.g., monetary) may be given. In one detail, a controller with sensors embedded within the controller, or a non-controller-based animal data sensor (e.g., a smart watch, on-body sensor, or embedded sensor) communicates with the game itself. In a further detail, communication between the sensor and the system is via a web browser. In another detail, simulation data may be purchased based at least in part on sensor data collected by the video game or game-based system. As previously described, this data may be utilized, for example, to gain an advantage within the game.In the context of a sports video game, the types of artificial data based on real sensor data that may be purchased within the game may include increased energy levels or the ability to run faster, fly higher, have a longer energy life, hit the ball further, which may increase the likelihood of winning the game. The types of simulated biological data provided may also include one or more special powers for one or more objects within the game, which utilize at least one associated characteristic for the biological data, which may be derived from one or more simulations, or may be generated from one or more statistical models or artificial intelligence techniques. Updates to the simulated data within the game may be provided or derived in real time or near real time as data is collected by the simulation system. In this variation, stimulus animal data generated by the above method may be provided to the simulation system.

[0110]

[0122] In one variation, simulation data derived from the biological data of the target subject may be utilized within the simulation to alter, modify, or enhance one or more other data types to inform one or more subjects. For example, in a fitness simulator, the system may utilize simulation data derived from the target subject to predict future biological measurements of any given activity from which other data (e.g., visual representations of the target subject) may be created, modified, or enhanced. The derived simulation data may include inputs such as exercise plans, nutrition plans, animal data of the target subject (including data sets collected via one or more biosensors utilized by the target subject), as well as current and / or modified visual representations of the target subject, which may include the subject's body (e.g., modified renderings of the subject's "target" future body via an avatar or other visual representation that may be adjustable by the target subject). The one or more simulation outputs may include optimal exercise, nutrition and daily lifestyle plans (e.g., including hours of sleep per night, social habits, etc.) based on at least a portion of the generated simulated animal data or one or more derivatives thereof, an altered visual representation of the target subject which may include the subject's body (e.g., an altered rendering of the subject's future body via an avatar or other visual representation after n days following a particular exercise and diet plan using at least a portion of the simulation data), and the like.

[0111]

[0123] In another variation, one or more users of a simulation (e.g., a video game or game-based system) may include their own animal data as part of the game and may compete against (1) other real-world subjects (e.g., professional sports athletes, fitness instructors, consumers looking to compare themselves to other consumers in fitness challenges, gamers, etc.) or (2) visual participants (e.g., avatars) that may share at least one characteristic with the one or more subjects. The system may run one or more simulations to convert the real-world animal data into simulated data to be used in the simulation game and / or convert the animal data into a format that can be input into the simulation system. For example, a user may want to participate in a head-to-head tennis match against athlete X in a simulation game (e.g., a virtual reality game) that includes simulated animal data based at least in part on actual animal data from both the one or more users and athlete X. Both the user and Athlete X may utilize one or more sensors that transmit various biological data to the system (e.g., ECG, heart rate, biomechanical data such as racquet swing data, positional data), which may be further calculated into one or more additional measurements (e.g., stress level, swing speed), converted into game-based metrics (e.g., "energy level" bars, "swing power" bars), or incorporated into one or more simulations to generate simulated data outputs. In particular, one or more simulated game users or spectators participating in or watching the game may place bets based on the game / competition (e.g., on matches played against Athlete X within the gaming system), create or modify one or more products, determine probabilities or odds of the occurrence of an outcome of an event, modify previously determined probabilities or odds for an event, or develop a strategy. In exchange for providing at least a portion of the animal data, one or more participants in the game or competition (e.g., Athlete X and / or one or more game players) may receive a portion of consideration from bets placed (e.g., from winning bets) or purchases made within the competition that directly or indirectly utilize their data.For example, a star tennis player may contribute his / her biological data to a video game simulation so that game users can play as or against a virtual representation of the star tennis player. In this situation, the user may pay a fee to the simulation operator (e.g., a video game company) to access the data or a derivative thereof (e.g., synthetic data generated based at least in part on the actual animal data), a portion of which may be donated to the star tennis player. Alternatively, the simulation operator may pay a license fee or provide other compensation (e.g., a percentage of game sales or data-related products sold) to the athlete for use of the data within the simulated game. In another example, the simulation operator allows one or more bets / wagers to be placed on the game itself (e.g., between the user and the star tennis player) or prop bets to be placed within the game (e.g., micro-bets based on various aspects within the game). In particular, the one or more prop bets are based at least in part on the animal data and / or one or more derivatives thereof (including the simulated data). In this situation, the user and / or star tennis player may receive a portion of compensation from each bet placed, the total number of bets placed, and / or one or more products created, offered, and / or sold based at least in part on the data. In another example, in a fitness class, an instructor may be able to receive compensation for a bet placed between the instructor and a user (e.g., who can pedal the most miles in 10 minutes), or a gamer may receive compensation for a prop bet that incorporates at least a portion of their biological data. In particular, one or more subjects that provide at least a portion of their animal data to one or more third parties as part of one or more simulations may receive compensation for providing access to their data.

[0112]

[0124] In particular, the simulation system creates artificial data for one or more simulated objects (which may represent one or more real-world objects) characterized by one or more characteristics that are desirable for a user or group of users. The created artificial data may then be used in one or more further simulations (e.g., as part of a baseline) to create one or more artificial data sets that can be used to calculate, calculate, derive, extract, extrapolate, simulate, create, modify, assign, enhance, estimate, evaluate, infer, establish, determine, transform, infer, observe, communicate, or take action based on one or more predictions or probabilities. For example, if a healthcare provider wishes to determine the effectiveness of a particular dosage of a drug on a target patient characterized by a particular characteristic (e.g., age, weight, height, medical history, social habits, particular medical condition), the healthcare provider may run one or more simulations utilizing other patient data that share one or more common characteristics with the target patient, including patients who have been provided with a particular dose of the drug or medication to determine the effect of the drug on the target patient, and the healthcare provider also utilizes the simulation data to evaluate one or more other potential outcomes (e.g., the likelihood that administration of the drug will lead to a heart attack, the likelihood that the drug will induce severe nausea, etc.). If the healthcare provider does not have a large enough data set or requires additional data to run one or more simulations to determine the effect of a drug on a subject with a particular desired characteristic, the healthcare provider may create an artificial data set by running one or more simulations or by other methods described herein that feature the particular characteristics of the target patient with one or more variables (e.g., amount of drug) that the healthcare provider desires as part of the probability assessment. The healthcare provider may then utilize the one or more artificial data sets as part of a baseline in one or more additional simulations to determine the probability of an event occurring. In particular, the healthcare provider may bill insurance companies (or vice versa) for each simulation run, which may result in a benefit being provided to one or more parties (e.g., one or more simulations may provide the healthcare provider with a probability of an event occurring).In one variation, a target patient, insurance provider, medical provider, or combination thereof may choose to conduct one or more simulations prior to administering inpatient treatment to determine one or more effects of a given action being taken, or the methods used by the medical provider may be applied to the patient.

[0113]

[0125] In particular, one or more artificial intelligence techniques may be utilized to evaluate one or more biosensor outputs and perform one or more data quality assessments on both the collected sensor data and the generated artificial data values. Even more particularly, one or more neural networks may be trained to generate one or more data values ​​that can be used to test one or more biosensor outputs (e.g., signals, measurements) and algorithms utilized to generate the one or more sensor outputs.

[0114]

[0126] More specifically, one or more artificial data values ​​are generated when detecting and replacing one or more values ​​(e.g., outliers, missing values) generated from one or more biosensors. Often, one or more sensors generate measurements (e.g., analog-derived measurements such as raw AFE data) that are provided to a server, and the server applies a method or technique to filter the data to generate one or more values ​​(e.g., heart rate values). However, when the signal-to-noise ratio of the data is very low or when one or more values ​​are missing, a pre-filter logic may be needed to generate the artificial data values. In one aspect, a pre-filter method is proposed in which the system "corrects" the data generated from the sensor to ensure that the one or more data values ​​generated are clean and fit within a predefined range. The pre-filter logic takes data from the sensor, detects outliers or "bad" values, replaces these values ​​with expected values ​​or "good" artificial values, and communicates the "good" artificial values ​​as the calculation of one or more animal data values ​​(e.g., heart rate values). The term "correction" refers to the function of creating one or more substitute data values ​​(i.e., "good" values) to align with the time series of generated values ​​and replace values ​​that may fall outside of pre-established thresholds with one or more "good" data values ​​that fall within the pre-established thresholds. These steps occur before any logic takes action on the received biological data to calculate one or more biological data values ​​(e.g., heart rate values).

[0115]

[0127] Advantageously, the pre-filter logic and methodology for identification and replacement of one or more data values ​​may be applied to any type of collected sensor data, including both raw and processed outputs. For purposes of illustration, raw data, such as analog front-end measurements (AFE), may be converted to other waveforms, such as surface electromyogram (sEMG) signals, but the presently disclosed and claimed subject matter focuses on conversion to ECG and heart rate (HR) values. However, the presently disclosed and claimed subject matter is not limited to the type of collected sensor data. As previously mentioned, the pre-filter logic becomes important in scenarios where the signal-to-noise ratio of the time series of AFE values ​​generated from one or more sensors is zero or close to zero, or is numerically small. In this case, one or more systems or methods that generate one or more heart rate values ​​may ignore one or more such values, and in some cases, no heart rate values ​​may be generated, or the generated heart rate values ​​may fall outside pre-established parameters, patterns, and / or thresholds. Such AFE values ​​may result from the subject performing an action that increases one or more other physiological parameters (e.g., muscle activity), or may result from competing signals introduced or degrading connections derived from the same sensor or from other variables, which may result in inconsistent HR series.

[0116]

[0128] To solve this problem, a method has been established in which one or more data values ​​are created by looking at future values ​​rather than previously generated values. More specifically, the system may detect one or more outlier signal values ​​and replace the outliers with one or more signal values ​​that are within an expected range (e.g., established upper and lower limits), thus acting to smooth the sequence while reducing the variance between the values. The established expected range may take into account a number of different variables including the individual, the type of sensor, one or more sensor parameters, one or more sensor characteristics, one or more environmental factors, one or more characteristics of the individual, the activity of the individual, etc. The expected range may be created by one or more artificial intelligence or machine learning techniques that use at least a portion of previously collected sensor data and / or one or more derivatives thereof, and possibly one or more of the aforementioned variables, to predict what the expected range may be. The expected range may change over a period of time and may be dynamic in nature, adjusting based on one or more variables (e.g., the activity the person is engaged in or the environmental conditions). In one variation, one or more artificial intelligence techniques may be employed, at least in part, to generate one or more artificial signal values ​​within an expected range (e.g., upper and lower limits) derived from at least a portion of the collected sensor data and / or one or more derivatives thereof from one or more sensors.

[0117]

[0129] To achieve the desired result of creating one or more values ​​based on future values, the system first samples one or more of the sensor's "normal" or "expected" AFE values ​​and applies statistical tests and exploratory data analysis to determine acceptable upper and lower limits for each AFE value generated by the sensor, which may include outlier detection techniques such as interquartile range (IQR), distribution and percentile cutoffs, kurtosis, etc. The normal or expected AFE values ​​may be determined by utilizing at least a portion of the previously collected sensor data. What is considered to be a normal or expected AFE value may vary depending on the sensor, the sensor parameters, or other parameters / characteristics that may be taken into account for what is determined to be normal or expected (e.g., the subject, the activity the subject engages in).

[0118]

[0130] Once an outlier is identified, the pre-filter logic then uses a backwards fill method to fill one or more outliers (i.e., AFE values ​​that are outside the accepted lower and upper limits) with the next available value that is within the normal range for the current window of samples. This results in a cleaner, more predictable time series of values ​​that is free of intractable noise. In particular, by utilizing one or more artificial intelligence techniques in which a model is trained to predict the next AFE value given a past sequence of AFE values, and / or one or more values ​​are generated as a replacement for one or more outliers to enable the sequence of values ​​to be within the normal range. In one variation, the user may utilize a heuristic or formula-based method that describes a waveform similar to that which is the AFE signal generated from the sensor.

[0119]

[0131] In the case of heart rate values, the system may increase the amount of data used by the pre-filter logic that processes the raw data to include n seconds worth of AFE data. Increasing the amount of data collected and utilized by the system allows the system to create a more predictable pattern of HR generated values ​​as the number of intervals used to identify QRS complexes increases. This occurs because HR is an average of HR values ​​calculated in 1 second subintervals. The number of n seconds is an adjustable parameter that may be pre-determined or dynamic. In particular, one or more artificial intelligence techniques may be utilized to predict the n seconds of AFE data required to generate one or more values ​​that fall within a predetermined range based on one or more previously collected data sets.

[0120]

[0132] Table 3 shows an example of pseudocode for generating artificial animal data (e.g., artificial sensor values) using an LSTM method, one or more of whose parameters may be tunable (variable), for training and testing AFE predictions under noisy inputs. TIFF0007656942000007.tif180170

[0121]

[0133] FIG. 11 illustrates out-of-sample testing of AFE data (labeled “AFE” in the figure) derived from a biosensor with predicted AFE values ​​generated using an LSTM neural network (labeled “PRED.AFE” in the figure) with an input layer having 15 nodes and a hidden layer having 10 nodes. The number of nodes can be a variable parameter. The history of AFE values ​​used to make this prediction is 20 (i.e., looking back at 20 timestamps). The number of history values ​​can be a variable parameter. In this figure, the model is trained on AFE data that is free of noise or disturbances that occur during physical activity that may introduce signals that are not true AFE values. This allows the system to predict what normal AFE values ​​will be given a past sequence of normal AFE values. In addition to training a model of baseline normal AFE values, the system can be trained on other collected data, which can include both animal data and non-animal data sets. Such training can be done using historical data and current (e.g., active) data sets (e.g., data collected in real time or near real time, such as at a live sporting event) as the data is received by the system. The system may utilize such collected data to generate more accurate predictions, probabilities, or possibilities, and may adjust one or more predictions, probabilities, or possibilities for one or more target subjects based on previously collected data attributes.

[0122]

[0134] While the pre-processing of the data may not reproduce a possible R peak of the QRS complex, it brings one or more noisy values ​​into the range of a normal or expected signal, thereby allowing downstream filters and systems that generate HR values ​​to generate one or more HR values ​​that are within the expected range in the absence of a quality signal. Additional details regarding systems for measuring heart rate and other biological data are disclosed in U.S. Application No. 16 / 246,923, filed January 14, 2019, and International Application No. PCT / US20 / 13461, filed January 14, 2020, the disclosures of which are incorporated herein by reference in their entireties.

[0123]

[0135] In particular, the simulation system generates artificial data values ​​to complete the dataset. For example, a sensor collecting any given biological data (e.g., heart rate) may have an event that prevents the sensor from collecting, analyzing, and / or distributing data to the simulation (e.g., one or more sensors are removed from the subject, the sensor runs out of power and stops collecting data, etc.). In this example, the simulation system may create one or more artificial datasets to complete the dataset (e.g., if the subject runs for 40 minutes and the heart rate sensor runs out of battery after 30 minutes, the simulation system may artificially generate the last 10 minutes of heart rate data, which may take into account one or more variables including previously collected data and datasets, speed, distance, environmental conditions, etc.).

[0124]

[0136] More specifically, a user provides one or more commands, and one or more computing devices (e.g., a simulation system, a computing device, or a third party) execute one or more actions that utilize at least a portion of the simulation data or a derivative thereof to fulfill at least a portion of the one or more commands. The commands may be initiated through a number of methods, including a physical cue (e.g., clicking an icon of an application) or a verbal cue (e.g., speaking to a voice-activated virtual assistant or other communication medium). The one or more commands may be neurologically initiated. For example, a computing device (e.g., a brain-computer interface) may acquire one or more of the subject's brain signals from neurons, analyze the one or more brain signals, and convert the one or more brain signals into commands that are relayed to an output device to perform a desired action. Acquisition of the brain signals may occur through a number of different mechanisms, including one or more sensors that may be implanted in the subject's brain). The one or more actions performed by the one or more computing devices based on the one or more commands include at least one of: (1) recommending whether or not to place one or more bets; (2) creating, enhancing, modifying, acquiring, offering, or dispensing one or more products; (3) evaluating, calculating, deriving, modifying, enhancing, or communicating one or more predictions, probabilities, or possibilities; (4) formulating one or more strategies; (5) performing one or more actions, including placing one or more bets; (6) mitigating or preventing one or more risks; (7) recommending one or more actions; or (8) any combination thereof. For example, a user may verbally communicate to a voice-activated assistant that he or she would like to place a particular bet. The voice-activated assistant may inform the user whether or not to place a bet by evaluating probabilities or odds based in part on at least a portion of the simulated animal data. The bet may then be made. In another example, a user may verbally communicate to a voice-activated assistant that the user would like to engage in physical activity when the user has a heart condition.The voice-activated assistant may inform the user whether or not to engage in an activity by assessing probabilities or odds based at least in part on the user's simulated animal data to determine cardiac and other health risks associated with the physical activity.

[0125]

[0137] In particular, the one or more computing devices may take one or more series of steps to obtain a response to a command provided by the user. For example, the user may request a program on the computing device to generate information through one or more simulations from which a response may be provided to the user (e.g., an optimal exercise plan including time, activity, and target biological measurements to burn a target number of calories). In another example, when the user verbally provides a command to the computing device or a third party system to determine whether to place a bet, the voice-activated assistant may run at least one simulation to inform the user whether to place a bet. In one variation, the computing device performs one or more actions on behalf of the user based on one or more thresholds set by the user, the one or more actions being initiated directly or indirectly as a result of at least a portion of the simulation data or one or more derivatives thereof. For example, the computing system or a third party may have the capability to monitor various inputs and variables in real time or near real time that may change the probability of an event occurring. By running one or more simulations using at least a portion of the animal data at any given time, the system may provide modified odds of an event occurring that may trigger a bet. In relation to this scenario and sports betting, a user may configure the system to place a bet on the probability of a threshold being reached (e.g., if the probability of Player A's heart rate exceeding 200 bpm in the fourth game of the third set of Match X is greater than 85%, then the user may configure the system to bet that Player A's heart rate will exceed 200 in the fourth game of the third set of Match X), or may provide the user with a notification that a bet should be made (e.g., a verbal notification, a kinesthetic notification such as a vibration on a smartwatch, a visual notification such as a pop-up in a virtual or augmented reality system, a text message on a phone, an alert in an app, etc.). Derivatives may include one or more calculated assets, insights and / or predictive indicators.In one variation, at least a portion of the simulation data is utilized to generate one or more insights, calculated assets, or predictive indicators. In this scenario, a computing system or a third party may generate probabilities of events occurring based at least in part on the simulation data generated by running one or more simulations.

[0126]

[0138] More specifically, the simulation data may be used as part of a health monitoring system. For example, a health monitoring system, such as a health platform or application, may be operable to utilize one or more artificial intelligence techniques to correlate the datasets and identify known biologically relevant issues from one or more target individuals or groups of target individuals, as well as to identify hidden patterns in the one or more datasets to identify biologically relevant issues based on the collected data. This may include discovering entirely new patterns in the data that have not previously been correlated with known issues, or discovering novel patterns in the one or more datasets that may identify novel issues. The application may be operable to further perform one or more simulations to generate one or more artificial datasets, which may enable a user to at least one of: (i) evaluate, calculate, derive, modify, enhance, or communicate one or more predictions, probabilities, or possibilities; (ii) develop one or more strategies; (iii) perform one or more actions; (iv) mitigate or prevent one or more risks; (v) recommend one or more actions; or (vi) any combination thereof. The artificial data generated from one or more simulations may be utilized to create, modify or enhance one or more insights, calculated assets or predictive indicators, or may be utilized in another one or more simulations that may create, enhance or modify one or more insights, calculated assets or predictive indicators. The communication of the animal data and one or more derivatives thereof (e.g., simulation data, predictive indicators, calculated assets) to one or more users of the monitoring system occurs in real-time or near real-time to provide a holistic view of the target subject and potential future risks derived from the simulation data. This may occur via displays within the application (e.g., biological data such as heart rate, respiration rate, biological fluid levels, and insights such as "energy levels," probability of experiencing a biological event, immediate risk threat regarding one or more biological measurements based on one or more simulations performed) or may be communicated in other forms (e.g., verbally via a virtual assistant, visually as part of an augmented reality display).Such health monitoring products may be used in a variety of industries, such as fitness, telemedicine / healthcare (including remote patient monitoring), insurance, general business (e.g., employee wellness), aviation, automotive, etc. In one variation, the health monitoring system may detect a health or medical condition based on one or more running simulations that utilize at least a portion of the collected sensor data, which may trigger any of the alerts that are sent to another computing device or devices (e.g., hospitals, medical professionals).

[0127]

[0139] Advantageously, such information may be communicated in real-time or near real-time via direct sensor communication with a web browser. In particular, the simulation data may be generated from or utilized as part of a data tracking system, and in particular a biological (animal) data tracking system including a connectivity application and a server. The connectivity application establishes wireless communication with a plurality of wireless communication enabled data sensors (e.g., wearable biosensors and / or other wireless communication enabled data sensors), receives one or more data streams from the sensors, and displays, via a browser application, one or more measurements derived from at least a portion of the streaming data. The server may transmit the connectivity application to the browser in response to a user accessing a web page. The connectivity application transmits one or more streams of data to the server, which then calculates one or more measurements. The plurality of wireless communication enabled sensors may include at least two different types of sensors, including sensors that communicate with the connectivity application using different communication interfaces. At least one of the measurements, including the simulation data or one or more derivatives thereof, may be derived from at least a portion of the data streams from the two different sensors. The one or more data streams may be inputs in one or more simulations from which the simulation data may be created. In one variation, the one or more measurements may also be derived from one or more simulations utilizing at least a portion of one or more data streams from two or more different sensors. The connectivity application may be operable to send one or more commands to a remote control device and / or a subset of the sensors to change one or more sensor settings, which may occur sequentially or simultaneously.

[0128]

[0140] More specifically, the simulation data may be generated from or utilized as part of a data tracking system, and in particular a biological (animal) data tracking system that includes one or more wireless communication enabled data sensors (e.g., biological data sensors), a computing device, and a connectivity application. The one or more wireless communication enabled data sensors may include at least two different types of sensors that may communicate with the connectivity application using different communication interfaces. The computing device includes a network connection (e.g., Internet) and a browser application (e.g., running browser software). The connectivity application executing within the browser is configured (e.g., programmed) to establish one or more wireless communication links with each of the one or more sensors, receive one or more data streams from the one or more sensors, and display, via the browser application, one or more measurements derived from at least a portion of the streaming data. The connectivity application may send one or more commands to the one or more wireless communication enabled data sensors to change, adjust, and / or modify one or more sensor settings. At least one of the measurements that may include the simulation data may be derived from at least a portion of one or more data streams from two or more different sensors. The one or more data streams may be inputs in one or more simulations from which the simulation data may be created. In one variation, the one or more measurements may also be derived from one or more simulations utilizing at least a portion of one or more data streams from two or more different sensors. The system may include a server configured to receive the one or more data streams via a network connection (e.g., the Internet) and calculate the measurements. The server may be operable to transmit a connection application program to the browser in response to a user accessing a web page.

[0129]

[0141] More specifically, the simulation data may be generated from or utilized as part of a method for tracking biological (animal) data, the method including transmitting a connectivity application from a server to a browser, detecting at least one wireless communication enabled sensor, wirelessly receiving one or more data streams from the at least one sensor, and displaying one or more measurements. The server transmits the application to the browser in response to a user accessing a web page. The connectivity application detects the one or more sensors and directly receives at least a portion of the one or more data streams. The one or more measurements are derived from at least a portion of the one or more data streams and are displayed at least in part on the browser. The one or more measurements may include simulation data. Characteristically, the one or more data streams may be inputs for one or more simulations from which simulation data may be created. The method may include transmitting at least a portion of the one or more data streams from the browser to the server, and transmitting the one or more measurements from the server to the browser. The connectivity application may be operable to send one or more commands to the one or more sensors to change one or more sensor settings. The one or more sensors may include at least two types of sensors that may transmit one or more data streams to the connectivity application using two different communication interfaces. Additional details regarding the data tracking system utilizing simulation data are disclosed in U.S. Patent Application No. 16 / 274,701, filed February 13, 2019, and International Application No. PCT / US20 / 18063, filed February 13, 2020, the disclosures of which are incorporated by reference in their entireties herein.

[0130]

[0142] Advantageously, the simulation data may be utilized as part of an unmanned aerial vehicle-based sensor data collection and distribution system. The unmanned aerial vehicle-based data collection and distribution system may include a source of electronically transmittable animal data. The source of animal data may include at least one biosensor. The animal data may be collected from at least one target individual or a group of target individuals. The system may include one or more unmanned aerial vehicles (e.g., drones, high altitude long endurance aircraft, high altitude pseudolites, atmospheric satellites, balloons, multirotor drones, airships, fixed wing aircraft, low altitude systems) that receive the animal data from one or more sensors and communicate with one or more other computing devices (e.g., a home station, another computing system). Characteristically, the one or more unmanned aerial vehicles include a transceiver operable to receive one or more signals or measurements from a source of animal data, collect data from one or more sensors that are part of the UAV (e.g., optical sensors, temperature sensors, etc., mounted on, integrated with, connected to, or associated with the one or more UAVs), and / or provide (e.g., transmit) data to another computing device or make the data accessible via a cloud. One or more UAVs may operate as part of a network of UAVs (e.g., a cellular network that utilizes a network of drones for one or more data-related functions), a network composed of one or more UAVs and non-UAVs (e.g., ground stations), or multiple networks.

[0131]

[0143] In particular, one or more simulations incorporating the collected sensor data may be performed to predict one or more animal data measurements (e.g., location, movement) of the target individuals and to optimize one or more UAVs. The one or more simulations may include the collected sensor data, one or more characteristics of the one or more target individuals (e.g., activities in which the one or more target individuals are engaged), one or more types of non-animal data (e.g., weather, search results or content from one or more mobile devices), and the like. For example, collecting location data from one or more target individuals to predict one or more movements via one or more simulations may enable: increasing the overall efficiency of one or more UAVs, including optimizing UAV formations (e.g., three-dimensional formations) to ensure optimal line of sight with one or more target individuals; mapping the UAVs; routing the UAVs (e.g., maximizing the efficiency of any given route to minimize energy consumption); sharing data between UAVs and other computing devices (e.g., determining data that may need to be shared or made available to other UAVs or computing devices versus one or more (e.g., what information may need to be stored based on one or more predicted movements of the target individual, what information needs to be replicated between UAVs to ensure seamless handoff based on the predicted movements, etc.), communications between systems (e.g., maximizing the likelihood of target detection or connection between one or more UAVs and one or more sensors based on the location of the target individual), antenna positioning, the type of antenna utilized to communicate with one or more sensors or systems, positioning of antenna arrays, optimization of beam patterns and orientation based on the predicted location of the target individual, placement / formation of one or more UAVs based on the predicted individual locations of targets (e.g., including predicted changes in altitude, elevation), etc.One or more actions performed by one or more UAVs on the simulation data may result in bandwidth optimization (e.g., more available bandwidth), increased energy savings of one or more UAVs (e.g., allowing the UAV to use energy for additional capabilities or increased flight time), more reliable communication between sensors and the UAVs (e.g., stronger signal strength, reduced data packet loss), maximizing coverage area, etc.

[0132]

[0144] More specifically, the artificial data may be generated utilizing one or more statistical models or artificial intelligence techniques from which one or more simulations may be performed to provide information that enables one or more UAVs to perform one or more actions. Based at least in part on the received sensor data from one or more target individuals, one or more UAVs may be operable to provide (e.g., transmit) data to one or more computing devices to perform one or more simulations or to perform one or more simulations on one or more UAVs. Based on output from the one or more simulations, one or more UAVs may perform one or more actions. For example, collected biosensor data from one or more target individuals may trigger one or more UAVs or a home station controlling one or more UAVs to perform one or more simulations with respect to one or more target individuals from which one or more predictions, probabilities, or possibilities may be calculated, calculated, derived, extracted, extrapolated, simulated, created, modified, enhanced, estimated, evaluated, inferred, established, determined, estimated, observed, communicated, or performed. More specifically, the one or more UAVs may detect or capture information to detect biological-based information based on one or more sensors (e.g., that a target subject is experiencing a medical event such as a heart attack or stroke), analyze the collected sensor data (e.g., utilizing one or more machine learning techniques to find patterns in the data to generate predictive or probability-based information) or provide the data for analysis via another computing device that has access to the data (e.g., via the cloud), and perform one or more actions (e.g., sending an alert to another system, such as a hospital system, notifying the system of the alert, delivering one or more medications or drugs as a result of the UAV's analysis of one or more signals or measurements; receiving the analyzed information from the computing device that provides the analysis and sending an alert to a third party).The alert may include one or more biological measurements (e.g., a summary of the measurements, the location of the target individual where the biological measurements were captured), along with information about the one or more UAVs (e.g., a predictive indicator conveying the likelihood of a medical event occurring based on the collected information). More specifically, the one or more UAVs may detect biological-based information that triggers the one or more UAVs to run one or more simulations, or triggers another computing device to receive or obtain data from the one or more UAVs to run one or more simulations, from which one or more predictions, probabilities, or likelihoods are derived (e.g., the collected biological sensor data provides measurements that indicate an anomaly in the data associated with a particular medical episode, so the system runs one or more simulations to determine the likelihood that the target individual will experience a medical episode within n time periods), and one or more actions are performed (e.g., the UAV delivers a first aid kit or other medical equipment to assist in dealing with the medical episode, or sends an alert to another system, such as a hospital system or emergency system, or sends an alert to the target individual that a medical episode is about to occur). More specifically, one or more UAVs may detect animal data and one or more other UAVs may perform an action (e.g., one UAV detects biological data, another UAV runs one or more simulations, another UAV interprets the captured sensor data and generated artificial information to predict the likelihood of a medical event occurring, and another UAV delivers one or more medications, prescriptions, or sensing / medical devices to address and mitigate / prevent the risk of a medical episode). More specifically, one or more UAVs may detect animal data and one or more other computing devices may perform an action (e.g., a UAV captures sensor data, transmits the data to a third party to run a simulation, and delivers the appropriate medication / prescription / device based on the output).

[0133]

[0145] Simulation data derived from at least a portion of the UAV-collected sensor data or one or more derivatives thereof may be used directly or indirectly: (1) as a market in which one or more bets are made or accepted; (2) to create, modify, enhance, acquire, offer or distribute one or more products; (3) to evaluate, calculate, derive, modify, enhance or communicate one or more predictions, probabilities or possibilities; (4) to develop one or more strategies; (5) to take one or more actions; (6) to mitigate or prevent one or more risks; (7) to recommend one or more actions; (8) as one or more signals or measurements utilized in one or more simulations, calculations or analyses; (9) as part of one or more simulations, the output of which directly or indirectly engages one or more users; (10) as one or more core components or supplements to one or more consumption media; (11) in one or more promotions; or (12) any combination thereof. For example, one or more simulations may be performed for the individual locations of a group of target individuals to predict expected individual locations in order to position one or more UAVs or a network of UAVs to ensure optimal placement. Additional details of an unmanned aerial vehicle-based data collection and distribution system for sensor data that can incorporate a simulation system are disclosed in U.S. patent application Ser. No. 16 / 517,012, filed July 19, 2019, and International Application No. PCT / US20 / 42705, filed July 20, 2020, the disclosures of which are incorporated by reference in their entireties into this specification.

[0134]

[0146] The simulation data may be used to change, adjust or modify one or more sensor settings. In particular, one or more simulations are performed that result in the change, adjustment or modification of one or more sensor settings. In one variation, a simulation system or computing device that receives the simulation data or one or more derivatives thereof (e.g., an alert based on the simulation data) automatically changes, adjusts or modifies one or more sensor settings based on the results of one or more simulations that may occur using one or more artificial intelligence techniques. For example, if a simulation generates simulation data that indicates that one or more sensor measurements of the target subject are likely to change from normal measurements to abnormal measurements based on one or more sensor measurements from a target subject (e.g., in the case of a simulation that predicts that the target subject will have a heart attack), the simulation system or other computing device may automatically change one or more sensor settings (e.g., increase the sampling rate of the sensor; increase the frequency at which one or more signals or measurements are provided; change / adjust / modify one or more computing devices that receive the sensor data. In this example, if the measurements of the data are expected to become irregular, one or more measurements may be automatically sent to a medical-related system or medical professional). In another example, if the simulation produces simulation data that indicates the target subject is expected to have normal ECG measurements, the simulation system of the computing device receiving the simulation data or one or more derivatives thereof may modify the sampling rate of the sensor (e.g., reduce the rate) or reduce the frequency of data provided by the sensor to the one or more computing devices (e.g., change the data streaming rate from continuous to intermittent to preserve battery life). In another example, the simulation data may trigger the target subject to conduct a virtual consultation with a medical professional via the simulation system (which may be, for example, a remote health monitoring / telemedicine platform) or another computing device.Based on the simulation data, the medical professional may change, modify or adjust one or more sensors utilized by the target subjects (e.g., cameras) so that the medical professional can tailor the one or more sensors to their particular needs (in the example of a camera, examining one or more particular parts of the body). In one variation, and based on the simulation data, the medical professional may change, modify or adjust one or more sensor settings utilized by one or more target subjects so that the one or more sensors can capture data that is more relevant to the monitoring, supervision or care of the one or more target subjects utilizing the one or more sensors. For example, a clinician may adjust the sensor settings to more accurately capture relevant animal data needed to make a patient monitoring, treatment, admission, review or follow-up decision (e.g., manually override settings) or adjust the sensor settings.

[0135]

[0147] Although exemplary embodiments have been described above, it is not intended that these embodiments describe all possible forms of the invention. Rather, the terms used herein are words of description rather than limitation, and it is understood that various changes may be made without departing from the spirit and scope of the invention. Moreover, features of various embodiments may be combined to form further embodiments of the invention.

Claims

1. A step of: a computing device receiving one or more sets of actual animal data derived at least in part from one or more sensors that receive, store and / or transmit information regarding one or more target individuals, the one or more sensors including a biological sensor configured to measure electrical signals on or from the body of the target individuals, the one or more target individuals including athletes in a sporting event; converting the electrical signals into one or more heart rate values ​​and / or electrocardiogram (ECG) data; generating, by the computing device, simulated animal data from at least a portion of real animal data or one or more derivatives thereof, wherein one or more parameters or variables of the one or more target individuals are modified, the simulated animal data including artificially created data that shares at least one biological function with humans or other animals; assessing fatigue of an athlete in a sporting event having elevated heart rate, diastolic blood pressure, systolic blood pressure, sweat rate, and / or distance traveled based on said simulated animal data over a period of time; The simulated animal data is generated by a method using a recurrent neural network (RNN), the method comprising: (a) receiving actual animal data from a plurality of events, where readings of the training animal data are time-stamped and the data is recorded at predetermined time intervals; (b) training an untrained recurrent neural network model using the real animal data to form a trained recurrent neural network model, the training comprising: (i) using a set of observations as input; (ii) running training over multiple epochs using an optimizer for updating the error metric and the network weights; and (c) generating simulated animal data by applying the trained recurrent neural network model, wherein the trained recurrent neural network model predicts physiological data including heart rate based on the actual animal data; and (d) performing in-sample prediction, where the simulated animal data is generated from a subset of the actual animal data, to predict future values ​​based on known outcomes; (e) performing out-of-sample prediction, wherein the simulated animal data is generated from observations that were not part of the actual animal data, predicting future values ​​based on previously unseen data; The method includes:

2. The method of claim 1 , wherein one or more simulations are performed to generate the simulated animal data.

3. The method of claim 2 , wherein at least a portion of the generated simulated animal data or one or more derivatives thereof is used to create, enhance or modify one or more insights, calculated assets or predictive indicators.

4. 3. The method of claim 2, wherein at least a portion of the generated simulated animal data or one or more derivatives thereof are used in one or more simulation systems, the one or more simulation systems being at least one of a game-based system, an augmented reality system, a virtual reality system, a mixed reality system, or an extended reality system.

5. The method of claim 4, wherein one or more computing devices utilized as part of one or more simulation systems are operable, directly or indirectly, to: (1) offer or accept one or more bets; (2) create, enhance, modify, acquire, offer or dispense one or more products; (3) evaluate, calculate, derive, modify, enhance or communicate one or more predictions, probabilities or possibilities; (4) develop one or more strategies; (5) execute one or more actions; (6) mitigate or prevent one or more risks; (7) recommend one or more actions; (8) engage one or more users; or (9) perform any combination thereof.

6. The method of claim 2 , wherein at least a portion of the generated simulated animal data is used as one or more inputs in one or more simulations to generate simulated animal data.

7. The method of claim 6 , wherein at least a portion of the generated simulated animal data or one or more derivatives thereof is used to create, enhance or modify one or more insights, calculated assets or predictive indicators.

8. 7. The method of claim 6, wherein the generated simulated animal data or one or more derivatives thereof are used in one or more simulation systems, the one or more simulation systems being at least one of a game-based system, an augmented reality system, a virtual reality system, a mixed reality system, or an extended reality system.

9. The method of claim 8, wherein one or more computing devices utilized as part of one or more simulation systems are operable, directly or indirectly, to: (1) offer or accept one or more bets; (2) create, enhance, modify, acquire, offer or dispense one or more products; (3) evaluate, calculate, derive, modify, enhance or communicate one or more predictions, probabilities or possibilities; (4) develop one or more strategies; (5) execute one or more actions; (6) mitigate or prevent one or more risks; (7) recommend one or more actions; (8) engage one or more users; or (9) perform any combination thereof.

10. The method of claim 1 , wherein the one or more parameters or variables modified to generate simulation data comprise non-animal data.

11. The method of claim 1 , wherein the simulated animal data is generated by randomly sampling at least a portion of the actual animal data.

12. 2. The method of claim 1 , wherein the simulated animal data is generated by fitting the actual animal data to a function with one or more independent variables or one or more adjustable parameters that are optimized to provide a fit to the actual animal data.

13. The method of claim 12 , wherein the function is a line, polynomial, exponential, Gaussian, Lorentzian, piecewise linear, or spline between actual data points.

14. The method of claim 12 , wherein the one or more independent variables or adjustable parameters include time, such that one or more biological parameters are associated with one or more virtual participants in a simulation as a function of time.

15. The method of claim 1 , wherein the simulated animal data is generated by adding one or more offset values ​​to each value of the actual animal data.

16. The method of claim 1 , wherein at least a portion of the actual animal data is converted to simulated data by adding one or more random numbers to each value of the actual data set.

17. The method of claim 1 , wherein at least a portion of the simulated animal data is converted into a lookup table for use by the simulation.

18. 2. The method of claim 1 , wherein at least a portion of the simulated animal data is generated by fitting the real animal data to a probability distribution and then randomly sampling the probability distribution to assign one or more biological parameters to one or more virtual subjects.

19. 20. The method of claim 18, wherein the probability distribution is selected from the group consisting of a Bernoulli distribution, a uniform distribution, a binomial distribution, a normal distribution, a Poisson distribution, an exponential distribution, and a Lorentz distribution.

20. 20. The method of claim 18, wherein the one or more sets of actual animal data include one or more non-animal data variables or parameters that are applied as one or more parameters or variables in the simulation.

21. 10. The method of claim 1, wherein a trained neural network generates simulated animal data, the trained neural network being trained with at least a portion of the actual animal data or one or more derivatives thereof.

22. 22. The method of claim 21, wherein the trained neural network is trained using at least a portion of simulation data.

23. 22. The method of claim 21, wherein the one or more parameters or variables modified to generate simulation data comprise non-animal data.

24. 22. The method of claim 21, wherein the trained neural network is a recurrent neural network.

25. 22. The method of claim 21, wherein the trained neural network is a long short-term memory recurrent neural network.

26. The method of claim 21 , wherein the trained neural network is a generative adversarial network.

27. The trained neural networks utilized to generate simulated animal data may be forward propagation, perceptrons, deep forward propagation, radial basis networks, gated recurrent units, autoencoders (AE), variational AE, denoising AE, sparse AE, Markov chains, Hopfield networks, Boltzmann machines, restricted BM, deep belief networks, deep convolutional networks, deconvolutional networks, deep convolutional inverse graphics networks, liquid state machines, extreme learning machines, echo state networks, deep residual networks, Kohonen networks, support vector machines, neural Turing machines, group methods of data processing, probabilistic, time delay, convolutional, decon ...

22. The method of claim 21, comprising one or more types of neural networks: deep stacking networks, general regression neural networks, self-organizing maps, learning vector quantization, simple recursive, reservoir computing, echo state, bidirectional, hierarchical, probabilistic, genetically scaled, modular, committee of machines, associative, physical, instantly trained, spike, regulatory feedback, neocognitron, composite hierarchical deep model, deep predictive coding network, multi-layer kernel machine, dynamic, cascade, neuro-fuzzy, constituent pattern generation, memory networks, one-shot associative memories, hierarchical temporal memories, holographic associative memories, semantic hashing, pointer networks, or encoder-decoder networks.

28. 22. The method of claim 21, wherein multiple neural networks are utilized on the same at least a portion of the animal data or one or more derivatives thereof to generate simulated data.

29. 22. The method of claim 21, wherein at least a portion of the simulated animal data is used, directly or indirectly: (1) as one or more markets in which bets are made or accepted; (2) to create, modify, enhance, acquire, offer or distribute one or more products; (3) to evaluate, calculate, derive, modify, enhance or communicate one or more predictions, probabilities or possibilities; (4) to develop one or more strategies; (5) to execute one or more actions; (6) to mitigate or prevent one or more risks; (7) to recommend one or more actions; (8) as one or more signals or measurements utilized in one or more simulations, calculations or analyses; (9) as part of one or more simulations, the output of which directly or indirectly engages one or more users; (10) as one or more core components or supplements to one or more consumption media; (11) in one or more promotions; or (12) any combination thereof.

30. 22. The method of claim 21, wherein one or more simulations utilizing at least a portion of the actual animal data or one or more derivatives thereof are generated to create simulation data that is used directly or indirectly: (1) as one or more markets in which bets are made or accepted; (2) to create, modify, enhance, acquire, offer or distribute one or more products; (3) to evaluate, calculate, derive, modify, enhance or communicate one or more predictions, probabilities or possibilities; (4) to develop one or more strategies; (5) to execute one or more actions; (6) to mitigate or prevent one or more risks; (7) to recommend one or more actions; (8) as one or more signals or measurements used in one or more simulations, calculations or analyses; (9) as part of one or more simulations, the output of which directly or indirectly engages one or more users; (10) as one or more core components or supplements to one or more consumer media; (11) in one or more promotions; or (12) any combination thereof.

31. The method of claim 1 , wherein the simulation is based on one or more target individuals participating in at least one of a fitness activity, a sporting event, a health evaluation, or an insurance evaluation.

32. 32. The method of claim 31, wherein at least a portion of the simulated animal data is used directly or indirectly by one or more computing devices: (1) as one or more markets in which bets are made or accepted; (2) to create, modify, enhance, acquire, offer or distribute one or more products; (3) to evaluate, calculate, derive, modify, enhance or communicate one or more predictions, probabilities or possibilities; (4) to develop one or more strategies; (5) to execute one or more actions; (6) to mitigate or prevent one or more risks; (7) to recommend one or more actions; (8) as one or more signals or measurements utilized in one or more simulations, calculations or analyses; (9) as part of one or more simulations, the output of which directly or indirectly engages one or more users; (10) as one or more core components or supplements to one or more consumable media; (11) in one or more promotions; or (12) any combination thereof.

33. The method of claim 1 , wherein at least a portion of the simulated animal data or one or more derivatives thereof are used to create or modify one or more insurance services, identifications, classifications, rates, reimbursements, or combinations thereof.

34. 2. The method of claim 1 , wherein at least a portion of the simulated animal data or one or more derivatives thereof is used in one or more simulation systems, the one or more simulation systems being at least one of a game-based system, an augmented reality system, a virtual reality system, a mixed reality system, or an extended reality system.

35. The method of claim 34, wherein one or more computing devices utilized as part of one or more simulation systems are operable, directly or indirectly, to: (1) offer or accept one or more bets; (2) create, enhance, modify, acquire, offer or dispense one or more products; (3) evaluate, calculate, derive, modify, enhance or communicate one or more predictions, probabilities or possibilities; (4) develop one or more strategies; (5) execute one or more actions; (6) mitigate or prevent one or more risks; (7) recommend one or more actions; (8) engage one or more users; or (9) perform any combination thereof.

36. 35. The method of claim 34, wherein at least a portion of the simulation user's animal data, or one or more derivatives thereof, is utilized as part of one or more simulations, and wherein the simulation user's animal data is derived at least in part from one or more sensors.

37. 37. The method of claim 36, wherein the device in communication with the one or more simulation systems has one or more sensors that contact, are embedded, attached, reside or are integrated with the device and provides at least a portion of the simulated user's data to a computing device, including the user's animal data or one or more derivatives thereof.

38. 2. The method of claim 1, wherein the one or more sensors and / or one or more appendages thereof are attached to, in contact with, or transmit one or more electronic communications related to or derived from one or more target individuals, including the body, eye, biological organ, muscle, hair, veins, biological fluids, blood vessels, tissue, or skeletal system of the one or more target individuals, are embedded in, reside on, or implanted in, are ingested by, are integrated to contain at least a portion of, or are integrated into, attached to, or embedded in, as part of, a fiber, fabric, garment, material, fixture, object, or device that contacts or communicates with one or more target individuals, directly or through one or more intermediaries.

39. 10. The method of claim 1, wherein the one or more sensors include at least one biosensor that collects physiological, biometric, chemical, biomechanical, locational, environmental, genetic, genomic or other biological data from one or more target individuals.

40. 10. The method of claim 1, wherein the one or more sensors collect or provide information that can be converted into at least one of the following types of data: facial recognition data, eye tracking data, blood flow data, blood volume data, blood pressure data, biofluid data, body composition data, biochemical data, pulse data, oxygenation data, core body temperature data, skin temperature data, galvanic skin response data, sweat data, location data, position data, audio data, biomechanical data, hydration data, cardiac-based data, neurological data, genetic data, genomic data, skeletal data, muscle data, respiratory data, kinesthetic data, ambient temperature data, humidity data, barometric pressure data, or altitude data.

41. 41. The method of claim 40, wherein the simulated animal data is utilized directly or indirectly by one or more computing devices that provide information regarding one or more insights, calculated assets, or predictive indicators collected or derived from the one or more sensors.

42. 10. The method of claim 1, wherein a user selects one or more parameters or variables for one or more simulations, one or more simulations occur, and one or more users obtain at least a portion of the simulated animal data or one or more derivatives thereof for a fee.

43. 2. The method of claim 1 , wherein a user provides one or more commands, and one or more computing devices perform one or more actions that utilize at least a portion of the simulated animal data or one or more derivatives thereof to satisfy at least a portion of the one or more commands.

44. The method of claim 1 , wherein at least a portion of the simulated animal data is utilized to create, enhance or modify one or more insights, calculated assets or predictive indicators.

45. 45. The method of claim 44, wherein the one or more insights are personal scores or other indices for one or more target individuals or groups of target individuals that utilize at least a portion of the simulation data to (1) evaluate, assess, prevent or mitigate risk based on animal data, (2) evaluate, assess and optimize performance based on animal data, or a combination thereof.

46. 2. The method of claim 1 , wherein one or more computing devices perform one or more actions on behalf of the user based on one or more thresholds set by the user, the one or more actions being initiated directly or indirectly as a result of at least a portion of the simulated animal data or one or more derivatives thereof.

47. 2. The method of claim 1, wherein the system detects at least one of one or more outliers generated from the one or more sensors or one or more missing values ​​in the data generated from the one or more sensors, and replaces the one or more outliers or missing values ​​with one or more artificial data values.

48. 48. The method of claim 47, wherein the one or more artificial values ​​are aligned with a time series of generated values ​​and fit within a pre-established threshold or range.

49. 48. The method of claim 47, wherein the one or more artificial values ​​are used at least in part as one or more inputs to derive animal data.

50. 10. The method of claim 1, wherein one or more health classifications, treatments, procedures, identifications, rates, reimbursements or services are created, modified or assigned, directly or indirectly, based at least in part on the simulated animal data or one or more derivatives thereof.

51. 1. A system for generating and distributing simulated animal data, the system comprising: one or more sensors for receiving, storing or transmitting information about one or more target individuals, the one or more sensors including a biological sensor configured to measure electrical signals of or obtained from the body of the target individuals, the one or more target individuals including athletes in a sport; receiving one or more sets of actual animal data derived at least in part from one or more sensors that receive, store or transmit information about one or more target individuals, the simulated animal data including artificially created data that shares at least one biological function with humans or other animals; generating simulated animal data from at least a portion of the actual animal data or one or more derivatives thereof, wherein one or more parameters or variables of the one or more target individuals are modified; assessing fatigue of an athlete in a sporting event having elevated heart rate, diastolic blood pressure, systolic blood pressure, sweat rate, and / or distance traveled based on said simulated animal data over a period of time; A computing device configured to execute Including, At least a portion of the artificially created data is processed by a trained recurrent neural network model, the method for said processing comprising: (a) receiving actual animal data from a plurality of events, where readings of the training animal data are time-stamped and the data is recorded at predetermined time intervals; (b) training an untrained recurrent neural network model using the real animal data to form a trained recurrent neural network model, the training comprising: (i) using a set of observations as input; (ii) running training over multiple epochs using an optimizer for updating the error metric and the network weights; and (c) generating simulated animal data by applying the trained recurrent neural network model, wherein the trained recurrent neural network model predicts physiological data including heart rate based on the actual animal data; and (d) performing in-sample prediction, where the simulated animal data is generated from a subset of the actual animal data, to predict future values ​​based on known outcomes; (e) performing out-of-sample prediction, wherein the simulated animal data is generated from observations that were not part of the actual animal data, predicting future values ​​based on previously unseen data; Including, the system.

52. 52. The system of claim 51, wherein one or more simulations are performed to generate the simulated animal data.

53. 52. The system of claim 51, wherein a trained neural network generates simulated animal data, the trained neural network being trained using at least a portion of the animal data or one or more derivatives thereof.

54. 54. The system of claim 53, wherein the trained neural network is trained using at least a portion of simulation data.

55. The system of claim 51, wherein the simulated animal data is used, directly or indirectly: (1) as one or more markets in which bets are made or accepted; (2) to create, modify, enhance, acquire, offer or distribute one or more products; (3) to evaluate, calculate, derive, modify, enhance or communicate one or more predictions, probabilities or possibilities; (4) to develop one or more strategies; (5) to execute one or more actions; (6) to mitigate or prevent one or more risks; (7) to recommend one or more actions; (8) as one or more signals or measurements utilized in one or more simulations, calculations or analyses; (9) as part of one or more simulations, the output of which directly or indirectly engages one or more users; (10) as one or more core components or supplements to one or more consumption media; (11) in one or more promotions; or (12) any combination thereof.

56. 52. The system of claim 51, wherein the simulation is based on one or more target individuals participating in at least one of a fitness activity, a sporting event, a health evaluation, or an insurance evaluation.

57. 52. The system of claim 51, wherein at least a portion of the simulated animal data or one or more derivatives thereof is used to create, modify or assign one or more health procedures, services, treatments, codes, identifications, classifications, rates, reimbursements, or combinations thereof.

58. 52. The system of claim 51, wherein a user selects one or more parameters or variables for one or more simulations, one or more simulations occur, and one or more users obtain at least a portion of the simulated animal data or one or more derivatives thereof for a fee.

Images