Wearable devices using machine learned models for individual-specific biometric tracking and outcome predictions

Abstract

A computing system is contemplated that includes a first machine-learned model trained to identify one or more phases of a menstrual cycle of a wearer of the wearable sensing device. The first machine-learned model uses training data obtained from a plurality of individuals that may or may not include the wearer. The computing system also includes a second machine-learned model trained to identify the one or more phases of the menstrual cycle of the wearer of the wearable sensing device. In contrast to the first machine-learned model, the second machine-learned model uses training data obtained only from the wearer. The use of two such models improves the accuracy of the menstrual cycle phase prediction, which can then be used with other biometric data to make health outcome predictions and recommendations.

Description

[0001] Wearable Devices Using Machine Learned Models for Individual-Specific Biometric Tracking and Outcome Predictions

[0002] Cross-Reference to Related Applications

[0003] This application claims the benefit of priority to U.S. Provisional Patent Application No. 63 / 678,962, filed August 2, 2024, U.S. Provisional Patent Application No. 63 / 709,579, filed October 21 , 2024, and U.S. Provisional Patent Application No. 63 / 709,694, filed October 21 , 2024, each of which is hereby incorporated by reference in its entirety.

[0004] Technical Field

[0005] The present disclosure is directed to a wearable sensing device for measuring one or biometric features of the wearer. The systems and methods described herein focus on devices that sense, measure, transmit, and present various data, including, but not limited to data regarding physiological conditions within and / or in proximity to a wearer’s body, including a vital sign, biodata and temperature within an artificially created cavity (created cavity temperature or “CBCT”) in a wearer’s body.

[0006] Background

[0007] Increased consumer interest in personal health has led to a variety of personal health monitoring devices being offered in the market. For example, wearable devices for monitoring personal health are well known in the art, which include electronic measurement devices that can be worn on a finger, wrist, arm, or other body part. Generally, such devices include electronic elements, such as one or more flexible printed circuit boards, processors, sensors, batteries, and the like, with the device being worn close to, on, and / or in contact with a surface of skin, where the devices detect, analyze, and transmit information concerning an individual’s biometric signals such as vital signs, and / or ambient data and which allow in some cases immediate biofeedback to the wearer. Also commonly referred to as wearables, fashion technology, smartwear, tech togs, streetwear tech, skin electronics or fashion electronics, wearable devices such as activity trackers are an example of the Internet of Things, since "things" such as electronics, software, sensors, and connectivity are effectors that enable objects to exchange data through the internet with a manufacturer, operator, and / or other connected devices, without requiring human intervention. Wearable devices are popular in consumer electronics, most commonly in the form factors of smartwatches, smart rings, and implants. Most wearable devices measure a variety of body conditions (i.e., temperature, heart rate variability, blood oxygenation levels, pulse rate, breathing rate, and so on) on a person’s skin, and often perform other functions. Typically, data collected from a wearable device has not been analyzed in the context of woman’s unique menstrual cycle. This can lead to less accurate historical data and predictions geared towards improving a woman’s health.

[0008] As such, a need currently exists for a wearable device and method of tracking an individual woman’s health based on her unique menstrual cycle in order to provide accurate tracking and outcome predictions that are individualized to provide useful feedback, suggestions, and options for a woman to manage her unique health.

[0009] Summary

[0010] Aspects and advantages of embodiments of the present disclosure will be set forth in part in the following description, or can be learned from the description, or can be learned through practice of the embodiments.

[0011] In one aspect of the present disclosure, a computing system is provided. The computing system includes a first machine-learned model trained to identify one or more phases of a menstrual cycle of a wearer of a wearable sensing device, wherein the first machine-learned model uses training data obtained from a plurality of individuals that may or may not include the wearer; a second machine-learned model trained to identify the one or more phases of the menstrual cycle of the wearer of the wearable sensing device, wherein the second machine-learned model uses training data obtained only from the wearer; one or more processors; and memory storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations. The operations include obtaining biometric parameter data from the wearer as measured via the wearable sensing device; inputting the biometric parameter data into the first machine-learned model for a first period of time, the first period of time being sufficient for an adequate amount of biometric parameter data from the wearer to be obtained by the wearable sensing device to be used as the training data to train the second machine-learned model; receiving, as an output of the first machine-learned model, the identification of the one or more phases of the menstrual cycle of the wearer; after the first period of time has elapsed, inputting the biometric parameter data obtained by the wearable sensing device into the second machine-learned model; and receiving, as an output of the second machine-learned model, the identification of the one or more phases of the menstrual cycle of the wearer.

[0012] In another aspect, the first machine-learned model is not used after the first period of time has elapsed so that only the second machine-learned model is used to identify the one or more phases of the menstrual cycle of the wearer after the first period of time has elapsed using biometric parameter data that is continuously input into the second machine-learned model.

[0013] In yet another aspect, the first period of time can include at least one complete menstrual cycle.

[0014] In still another aspect, the one or more phases of the menstrual cycle can include a menstruation phase, a follicular phase, an ovulation phase, and a luteal phase.

[0015] In an additional aspect, the wearable sensing device can include a first wearable sensing device and a second wearable sensing device, wherein the first wearable sensing device and the second wearable sensing device can each be inserted into a created cavity of the wearer. The first wearable sensing device and the second wearable sensing device can measure the one or more biometric data parameters simultaneously, or the first wearable sensing device can measure the one or more biometric data parameters while the second wearable sensing device is not activated and vice versa, such that the system toggles between wearable sensing devices.

[0016] In one more aspect, the one or more biometric data parameters can include created cavity temperature, SpC>2, active energy expenditure, resting energy expenditure, total energy expenditure, sleep metrics, heart rate, heart rate variability, physical activity, or a combination thereof.

[0017] In another aspect, the first machine-learned model, the second machine- learned model, or both can include one or more of a deep artificial neural network, a transformer network with attention mechanisms, a support vector machine, a decision tree, or a linear model. For example, the deep artificial neural networks can include architectures that employ attention algorithms, such as self-attention and multi-head attention mechanisms, to enhance pattern recognition and feature extraction from sensor data.

[0018] In still another aspect, data corresponding with the one or more biometric data parameters and information related to the identification of the one of more phases of the menstrual cycle of the wearer can be input into a third machine-learned model, wherein the third machine-learned model outputs predictions related to the wearer’s physical health, mental health, stress level, readiness, restorative shift, resilience, or a combination thereof. Further, the third machine-learned model can include a large language model, a generative Al model, a foundation model, or a combination thereof, wherein a foundation model is a model trained through self-supervision, a cross-modal model, a knowledge distillation model, a multi-site model, or a combination thereof.

[0019] In yet another aspect, data corresponding with the one or more biometric data parameters and information related to the identification of the one of more phases of the menstrual cycle of the wearer input into a fourth machine-learned model, wherein the fourth machine-learned model outputs recommendations to improve the wearer’s physical health, mental health, stress level, readiness, restorative shift, resilience, or a combination thereof. Further, the fourth machine-learned model can include a large language model, a generative Al model, a foundation model, or a combination thereof, wherein a foundation model is a model trained through self-supervision, a cross-modal model, a knowledge distillation model, a multi-site model, or a combination thereof.

[0020] In one more aspect of the present disclosure, a method is provided that includes obtaining, by a computing system, biometric parameter data from a wearer of a wearable sensing device as measured via the wearable sensing device; inputting, by the computing system, the biometric parameter data into a first machine-learned model for a first period of time, the first period of time being sufficient for an adequate amount of biometric parameter data from the wearer to be obtained by the wearable sensing device to be used as the training data to train a second machine-learned model; receiving, by the computing system as an output of the first machine-learned model, an identification of one or more phases of a menstrual cycle of the wearer; after the first period of time has elapsed, inputting, by the computing system, the biometric parameter data obtained by the wearable sensing device into the second machine-learned model; and receiving, by the computing system as an output of the second machine-learned model, the identification of the one or more phases of the menstrual cycle of the wearer.

[0021] In another aspect, the first machine-learned model can be trained to identify the one or more phases of the menstrual cycle of the wearer using training data obtained from a plurality of individuals that may or may not include the wearer.

[0022] In still another aspect, the second machine-learned model can be trained to identify the one or more phases of the menstrual cycle of the wearer of the wearable sensing device, wherein the second machine-learned model can use training data obtained only from the wearer.

[0023] In yet another aspect, the first machine-learned model is not used after the first period of time has elapsed so that only the second machine-learned model is used to identify the one or more phases of the menstrual cycle of the wearer after the first period of time has elapsed using biometric parameter data that is continuously input into the second machine-learned model.

[0024] In an additional aspect, the wearable sensing device can include a first wearable sensing device and a second wearable sensing device, wherein the first wearable sensing device and the second wearable sensing device are each inserted into a created cavity of the wearer.

[0025] In one more aspect, the method can include measuring the one or more biometric data parameters with the first wearable sensing device and the second wearable sensing device simultaneously.

[0026] In another aspect, the method can include measuring the one or more biometric data parameters with the first wearable sensing device while the second wearable sensing device is not activated and vice versa.

[0027] In yet another aspect, the one or more biometric data parameters can include created cavity temperature, SpC>2, active energy expenditure, resting energy expenditure, total energy expenditure, sleep metrics, heart rate, heart rate variability, physical activity, or a combination thereof.

[0028] In another aspect, the first machine-learned model, the second machine- learned model, or both can include one or more of a deep artificial neural network, a transformer network with attention mechanisms, a support vector machine, a decision tree, or a linear model. For example, the deep artificial neural networks can include architectures that employ attention algorithms, such as self-attention and multi-head attention mechanisms, to enhance pattern recognition and feature extraction from sensor data.

[0029] In one more aspect, the method can include inputting, via the computing system, data corresponding with the one or more biometric data parameters and information related to the identification of the one of more phases of the menstrual cycle of the wearer into a third machine-learned model; and outputting, by the computing system via the third machine learned model, predictions related to the wearer’s physical health, mental health, stress level, readiness, restorative shift, resilience, or a combination thereof. Further, the third machine-learned model can include a large language model, a generative Al model, a foundation model, or a combination thereof, wherein a foundation model is a model trained through selfsupervision, a cross-modal model, a knowledge distillation model, a multi-site model, or a combination thereof.

[0030] In a further aspect, the method can include inputting, via the computing system, data corresponding with the one or more biometric data parameters and information related to the identification of the one of more phases of the menstrual cycle of the wearer into a fourth machine-learned model; and outputting, by the computing system via the fourth machine-learned model, recommendations to improve the wearer’s physical health, mental health, stress level, readiness, restorative shift, resilience, or a combination thereof. Further, the fourth machine- learned model can include a large language model, a generative Al model, a foundation model, or a combination thereof, wherein a foundation model is a model trained through self-supervision, a cross-modal model, a knowledge distillation model, a multi-site model, or a combination thereof.

[0031] These and other features, aspects, and advantages of various embodiments of the present disclosure will become better understood with reference to the following description and appended claims. The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate example embodiments of the present disclosure and, together with the description, serve to explain the related principles.

[0032] Brief Description of the Drawings A full and enabling disclosure of the present disclosure to one skilled in the art, including the best mode thereof, is set forth more particularly in the remainder of the specification, including reference to the accompanying figures, in which:

[0033] FIG. 1 illustrates an exploded perspective view of one exemplary embodiment of a wearable sensing device in the form and shape of a post-type earring;

[0034] FIG. 2 illustrates the assembled wearable sensing device of FIG. 1 ;

[0035] FIGs. 3A through 3D illustrate various views of a flexible circuit board and battery of the wearable sensing device of FIG. 1 ;

[0036] FIG. 4 illustrates a wearable sensing device system according to one embodiment of the present disclosure;

[0037] FIG. 5 illustrates various devices and components capable of communicating with each other according to one embodiment of the present disclosure; and

[0038] FIG. 6 illustrates an example computing system for a machine-learning based outcome predictor determination or a wearer-specific recommendation determination based on the one or more biometric data parameters for the wearer.

[0039] Repeat use of reference characters in the present specification and drawings is intended to represent the same or analogous features or elements of the present disclosure.

[0040] Detailed Description of Representative Embodiments

[0041] It is to be understood by one of ordinary skill in the art that the present discussion is a description of exemplary embodiments only, and is not intended as limiting the broader aspects of the present disclosure. Any of the features, components, or details of any of the arrangements or embodiments disclosed in this application are interchangeably combinable with any other features, components, or details of any of the arrangements or embodiments disclosed herein to form new arrangements and embodiments.

[0042] The present disclosure is generally directed towards a wearable sensing device to measure various physiological conditions of a wearer. Such measurements can include sensing a useful body temperature and / or variations thereof over time, heart rate, breathing rate, blood oxygenation, pulse, movement, activity, calories, distance traveled, steps, blood pressure, glucose monitoring, angular velocity measurements, location (e.g., global positioning), magnetic field measurements, and ambient noises and other conditions.

[0043] Among other things, the device is capable of measuring the created cavity temperature (OCT) within an artificial created cavity in living tissue or the wearer’s body into which the device is inserted. This current device provides a wearable thermometer that continuously or periodically measures temperature for a convenient, comfortable method of continuously tracking the wearer’s temperature. Alternatively or in combination, the device may include LED sensors for heart rate, breathing rate, blood oxygenation and / or pulse measurements. Alternatively or in combination, the device may include an accelerometer for movement, activity, calories, distance traveled and / or steps. Alternatively or in combination, the device may include LED sensors and / or Pressure sensors for blood pressure. Alternative or in combination, the device may include a CGM sensor for glucose monitoring. Alternatively or in combination, the device may include a gyroscope for angular velocity measurements. Alternatively or in combination, the device may include a global positioning system sensor. Such features are used to collect data from the wearer that can then be used to make predictions and provide recommendations specific to the wearer based on the wearer’s own biometric data and menstrual cycle information with respect to physical health, mental health, sleep, stress, readiness, restoration, resilience, etc.

[0044] Specifically, In one embodiment of the present disclosure, a computing system is contemplated that includes a first machine-learned model trained to identify one or more phases of a menstrual cycle of a wearer of the wearable sensing device. The first machine-learned model uses training data obtained from a plurality of individuals that may or may not include the wearer. The computing system also includes a second machine-learned model trained to identify the one or more phases of the menstrual cycle of the wearer of the wearable sensing device. In contrast to the first machine-learned model, the second machine- learned model uses training data obtained only from the wearer. The system also includes one or more processors and memory storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations. The operations include: obtaining biometric parameter data from the wearer as measured via the wearable sensing device; inputting the biometric parameter data into the first machine-learned model for a first period of time, the first period of time being sufficient for an adequate amount of biometric parameter data from the wearer to be obtained by the wearable sensing device to be used as the training data to train the second machine-learned model; receiving, as an output of the first machine-learned model, the identification of the one or more phases of the menstrual cycle of the wearer; after the first period of time has elapsed, inputting the biometric parameter data obtained by the wearable sensing device into the second machine-learned model; and receiving, as an output of the second machine-learned model, the identification of the one or more phases of the menstrual cycle of the wearer.

[0045] Wearable Sensing Device Components

[0046] Referring now to FIGs. 1-6, the particular components of the wearable sensing device are shown. Specifically, FIG. 1 depicts a simplified exploded perspective view of a body-mounted wearable sensing device 100 in the form and shape of a post-type earring, while FIG. 2 depicts the assemble view of the wearable sensing device 100. The wearable sensing device 100 includes a central body 110, a base 120, a post 130, a surface dome or cover 140 and a clasp or backing 150. Further, it is to be understood that although FIGs. 1 and 2 only shows one wearable sensing device 100, a pair of wearable sensing devices 100 are contemplated by the present disclosure for, inter alia, improved accuracy, additional data points, battery conservation, and the like.

[0047] As best seen in FIGs. 3A through 3D, the central body 300 can include a semi-flex board, flex board or flexible circuit board 310 having a first rigid section 320, a second rigid section 330, and a flexible linkage 340 positioned therebetween and connecting the two rigid sections. The shape of the two rigid sections is desirably round to fit within the dome enclosure. Exemplary dimensions for the flexboard components can include a thickness of 825 pm (±120 pm) and a diameter of approximately 10 mm for each rigid section, with a 133 pm thickness (± 50 pm) for a 6 mm long flexible section having a width of 3.48 mm and a 0.55 mm radius of flex.

[0048] In the depicted embodiment, a power supply or battery 350 can be positioned or sandwiched between the first and second rigid sections for sizing considerations and / or a variety of other reasons, although a wide variety of component configurations and / or arrangements can be utilized by those of ordinary skill in the art of circuit design. In various embodiments, the disclosed devices desirably utilize minimal power and have the capability to measure a wearer’s temperature or other vital signs or biodata via the earring post within the ear piercing (and / or other measurements) at least every 15 mins for a minimum of eight (8) hours before requiring recharging. More preferably, the disclosed devices allow for the capability to measure a wearer’s temperature via the earring post within the ear piercing every 15 seconds for a minimum of twenty-four (24) hours before requiring recharging. As described herein, the carrying and / or storage case for the earrings may incorporate battery charging features which are integrated into the portable storage case.

[0049] Various types of batteries may be utilized for any of the components described herein, including films, flex, rechargeable, non-chargeable, electronic charging, solar-power charging, trickle charging, battery maintainers etc.

[0050] FIGs. 3C and 3D depict the exemplary flexible circuit board 310 of FIGs. 3A and 3B after being flexed to a desired “stacked” configuration, with a battery 350 (e.g., a 3.7 Volt 13 m-A-hr Li-ION battery commercially available from Shenzhen Grepow Battery Co., Ltd. of Shenzhen, China) desirably positioned between the rigid sections 320 and 330 with battery leads 360 shown extending along and / or connected to ports on the second rigid section 330. The first rigid section will desirably house the MCU, memory and RF antenna, with this section preferably thermally isolated to some degree from the more heat generating second rigid section, as well as isolating the digital signals from the analog signals. The second rigid section will desirably house the power, analog sensors, post assembly and battery (which can be soldered as a post process of SMD placements). The battery leads are desirably soldered to castellated edge connections on this section. The first rigid section can also incorporate castellated edge connections for SWD Debug / Programming Interface between MCU and External Debugger / Programmer.

[0051] In various embodiments, a miniature transmitter is included on the flexboard and is used for interfacing the wearable sensor, LED sensors, accelerometer, and other sensors to a measurement tracking or control device. The transmitter can be positioned on an end of the device and is desirably located outside of the artificial created cavity. The transmitter contains the capability to isolate, amplify, filter noise, linearize, and convert input signals from the data sensors and wearable sensor and send a standardized output signal to the computing / control device. Common electrical output signals ranges are used.

[0052] As best seen in FIGs. 3B through 3D, a central post 370 will desirably extend through and be secured within an opening in the second rigid section 330. In a preferred embodiment, an overall dimension of the earring device can be an outer diameter of 12 mm to 13 mm, with the dome shaped and / or colored to present a pearl-like appearance or other desired colors or styles. In various alternative embodiments, the device can have a width / diameter of approximately less than 16 mm, and more preferably less than 12 mm. The post can have a length of approximately 6 mm to 8 mm and a diameter of approximately 1 mm, although other posts having lengths of 4.5 mm, 6 mm, 8 mm and 10 mm and diameters / post thicknesses such as 0.8 mm, and 1 .2 mm (and / or other sized known in the art) are contemplated herein with various design changes. The earrings can each have a total weight of less than 8 grams, and more preferably less than 5 grams, and desirably incorporates a comfortable and lightweight design and outer profile to allow the wearer to wear the earrings during sleep.

[0053] As best seen in FIGs. 1 and 2, the base component 120 will desirably fit partially and / or fully within the cover 140 (e.g., preferably via seamless outer and inner molding), with a lower surface of the base positioned adjacent to or against a skin surface of the wearer (e.g., an ear surface for an earring embodiment) when the post is contained within the piercing channel. In various embodiments, the device components will desirably be fully sealed and capable of full water immersion, including during bathing and / or showers. In some embodiments, additional water protections may be provided, such as 1 meter, 3 meter and / or 100 meter waterproof I water resistance measures.

[0054] For one non-limiting example, the cover component can include a small, spherical housing, approximately 0.25 inch diameter, containing all electrical components needed to operate the wearable sensors and / or biosensors and / or temperature sensors and / or fluid sensors and associated components. Such a design can desirably emulate the profile, shape and / or coloration of a small, pearl earring or similar design.

[0055] While the disclosed embodiment is a one-piece earring, it is contemplated that an alternative design could incorporate a two-piece construction, such as where the wearable sensing device 100 includes two parts, including a proximal portion that is able to be disconnected from a distal portion. For example, the distal portion could include a miniature transmitter, other components and / or at least one wearable sensor and biosensor and temperature sensor, while the proximal removable portion could include a miniature battery. The proximal portion can be connected to the distal portion by means of the post (e.g., containing a temperature sensor and / or other biosensors and / or wearable sensors and / or microphone and / or speaker), with the proximal portion encompassing the capability to detach by manually sliding off the post, similar to an earring back. By removing the proximal portion of the device, inserting the distal portion of the device through a created cavity, and returning the proximal portion onto the device to provide energy for the various components thereof, CCT or other biometrics can be measured and tracked.

[0056] In various embodiments, as shown in FIG. 1 , the post 130 and / or base 120 can desirably incorporate open and / or clear / transparent portions which allow various components to access the skin surface and / or transmit / receive information from the wearer’s anatomy, such as LED transmitters and / or sensors to detect and / or calculate wearer anatomical measurements such as heart rate, pulse, breathing rate, blood oxygenation and / or CO2 levels. A variety of sensor types can be incorporated into the device, including a wide variety of biochemical (enzymebased, tissue-based, immunosensors, DNA biosensors, and thermal and piezoelectric biosensors), chemical, electromechanical, optical and / or electrical sensors. For example, a chemical sensor may be included to measure the concentration levels of chemicals in blood, sweat or other bodily fluids, such as glucose monitors for diabetics and / or lactate level measurements, as well as sensors to measure proteins or hormones, fertility hormones or other chemical constituents (e.g., stress hormones) in sweat. Similarly, an electromechanical sensor can be incorporated to use electrical measurements to track mechanical movements, such as an accelerometer to measure physical activity and / or device / wearer orientation, inertial measurement units to measure angular changes and / or linear acceleration (e.g., for rotational velocity and / or position tracking) or GPS. Optical sensors can be incorporated to detect various biological signals like heart rate, heart rate variability, pulse, breathing rate, oxygen saturation and / or blood pressure (as well as temperature, galvanic skin response and / or stress sensors), with these sensors typically including a light source (transmitter or emitter) and photodiode sensors (detectors or receivers) that measure how much light is absorbed, reflected back out and / or passed through adjacent tissues via spectroscopy analysis. Electrical sensors (including bioelectrical sensors and electrochemical sensors) can be included to detect, measure, and evaluate electrical signals in the wearers tissues, including to measure heart rate or brain activity, including electrocardiogram information (e.g., ECG or heart rate monitor), EEG measurements (electroencephalograms) electromyography measurements (EMG or muscle movement monitor) and / or electrode dermal sensors to measure sweat levels (e.g., perspiration monitoring). Other sensors contemplated herein include pressure sensors, continuous glucose monitoring sensors, gyroscopes, GPS receivers and other wearable sensors.

[0057] In various embodiments, combinations of the following measurements and / or sensors are contemplated (including in any combinations thereof): Temperature, Pulse, Resting Heart Rate, Heart Rate Variability (HRV), Heart Beat Sensor, Perfusion, Oxygen Level, Blood Oxygen Level (SpO2), Breathing Rate, Blood Pressure, Glucose, Hormones, GPS, Accelerometer, Motion Sensors, ambient and / or cavity microphones and speakers.

[0058] Where a measurement sensor or wearable sensor, for a non-limiting example wearable sensor is incorporated into the device, this is desirably a small linear or non-linear rod-like structure, located within and / or on a structure that passes into or through the created artificial cavity, with the sensor in electronic communication with a circuit board containing operational software for the sensor. The wearable sensor can desirably sense or measure temperature or temperature changes constantly or periodically using specific or nonspecific time intervals. In one non-limiting example, this sensor functions to accurately detect small temperature changes, for measurement orders of about 1 degrees to about 0.01 degrees Fahrenheit. The temperature can be a unitary sensor unit, or a plurality of sensors can be used.

[0059] In one exemplary embodiment disclosed herein, shown in FIG. 1 , the wearable sensing device 100 desirably incorporates a plurality of wearable sensors, including at least one temperature sensor (e.g., thermistor sensor SC30F103AN, commercially available from Amphenol Thermometries, Inc. of St. Marys, PA, USA) positioned within the post 130, an ambient temperature sensor (e.g., NTC thermistor NCP03XH103J05RL commercially available from Murata Electronics North America, inc. of Smyrna, GA USA) on the central body 110, an accelerometer (e.g., accelerometer MC3635 commercially available from Memsic Semiconductor Co., Ltd. of Zhubei City, Hsinchu County, Taiwan), and an optical sensor package which incorporates an optical biosensor with proximity sensor and ambient light sensing features (Renesas OB1203SD-C4, commercially available from Renesas Electronics Corporation of Tokyo, Japan). The Renasas photoplethysmography (PPG) biosensor integrates light sources and drivers, analog digital conversion and I2C communication in a single optical package, with data from the OB1203 biosensor potentially being used to determine heart rate (HR), oxygen saturation (SpO2), respiration rate (RR), pulse and / or heart rate variability (HRV - a measure of stress). In various embodiments, this device can desirably measure one or more of the following: heart rate, heart rate variability, oxygen saturation, respiration rate, 3-axis accelerometer, ear lobe cavity temperature and / or ambient temperature.

[0060] Referring now to FIG. 4, components of an example system 400 of the wearable sensing device 100 that can be utilized in accordance with various embodiments are illustrated. In particular, as shown, the system 400 may also include at least one controller 402 communicatively coupled to the plurality of biometric sensors 170 contained within or on the wearable sensing device 100. Moreover, in an embodiment, the controller(s) 402 may be a central processing unit (CPU) or graphics processing unit (GPU) for executing instructions that can be stored in a memory device 404, such as flash memory or DRAM, among other such options.

[0061] For example, in an embodiment, the memory device 404 may include RAM, ROM, FLASH memory, or other non-transitory digital data storage, and may include a control program comprising sequences of instructions which, when loaded from the memory device 404 and executed using the controller(s) 402, cause the controller(s) 402 to perform the functions that are described herein. As would be apparent to one of ordinary skill in the art, the system 400 can include many types of memory, data storage, or computer-readable media, such as data storage for program instructions for execution by the controller or any suitable processor. The same or separate storage can be used for images or data, a removable memory can be available for sharing information with other devices, and any number of communication approaches can be available for sharing with other devices.

[0062] In addition, as shown, the system 400 can include any suitable external display 406, such as a touch screen, organic light emitting diode (OLED), or liquid crystal display (LCD) on a mobile phone, tablet, or computer, although devices might convey information via other means, such as through audio speakers, projectors, and the present disclosure contemplates casting the display or streaming data to another device, such as a mobile phone, tablet, or computer, wherein an application on the mobile phone displays the data obtained from the wearable sensing device 100. The system 400 may also include one or more wireless components 412 operable to communicate with one or more electronic devices within a communication range of the particular wireless channel. The wireless channel can be any appropriate channel used to enable devices to communicate wirelessly, such as Bluetooth, cellular, NFC, Ultra-Wideband (UWB), or Wi-Fi channels. It should be understood that the system 400 can have one or more conventional wired communications connections as known in the art.

[0063] The system 400 also includes one or more power components 408, such as may include a battery operable to be recharged through conventional plug-in approaches, or through other approaches such as capacitive charging through proximity with a power mat or other such device. In further embodiments, the system 400 can also include at least one additional I / O device 410 able to receive conventional input from a wearer. This conventional input can include, for example, a push button, touch pad, touch screen, wheel joystick, keyboard, mouse, keypad, or any other such device or element whereby a wearer can input a command to the system 400. In another embodiment, the I / O device(s) 410 may be connected by a wireless infrared or Bluetooth or other link as well in some embodiments. In some embodiments, the system 400 may also include a microphone or other audio capture element that accepts voice or other audio commands. For example, in particular embodiments, the system 400 may not include any buttons at all, but might be controlled only through a combination of visual and audio commands, such that a wearer can control the wearable sensing device 100 without having to be in contact therewith. In certain embodiments, the I / O elements 410 may also include one or more of the biometric sensors 170 described herein, optical sensors, barometric sensors (e.g., altimeter, etc.), and the like. Still referring to FIG. 4, the system 400 may also include a driver 414 and at least some combination of one or more emitters 416 and one or more detectors 418 (referred to herein as an optics package 415) for measuring data for one or more metrics of a human body, such as for a person wearing the wearable sensing device 100. In such embodiments, as shown in FIG. 4, for example, the optics package 415 may be arranged within the central body 110, base 120, post 130, surface dome or cover 140, clasp or backing 150 and at least partially exposed through an exterior surface of the wearable sensing device 100. Thus, as shown and further explained herein, the biometric sensors 170 may be positioned around the optics package 415 on an exterior surface of the wearable sensing device 100. In alternative embodiments, the various components of the optics package 415 may be positioned around the biometric sensors 170 and / or in another other suitable configuration such as adjacent to, interspersed with, surrounded by, or on top of the optics package 415. In certain embodiments, for example, wherein the biometric sensors 170 are transparent, the biometric sensors 170 may be arranged atop the optics package 415.

[0064] The emitters 416 and detectors 418 of FIG. 4 may also be capable of being used, in one example, for obtaining optical polyplethysmography (PPG) measurements. Some PPG technologies rely on detecting light at a single spatial location, adding signals taken from two or more spatial locations, or an algorithmic combination thereof. Both of these approaches result in a single spatial measurement from which the heart rate (HR) estimate (or other physiological metrics) can be determined. In some embodiments, a PPG device employs a single light source coupled to a single detector (i.e. , a single light path). Alternatively, a PPG device may employ multiple light sources coupled to a single detector or multiple detectors (i.e., two or more light paths). In other embodiments, a PPG device employs multiple detectors coupled to a single light source or multiple light sources (i.e., two or more light paths). In some cases, the light source(s) may be configured to emit one or more of green, red, infrared (IR) light, as well as any other suitable wavelengths in the spectrum (such as long IR for metabolic monitoring). For example, a PPG device may employ a single light source and two or more light detectors each configured to detect a specific wavelength or wavelength range. In some cases, each detector is configured to detect a different wavelength or wavelength range from one another. In other cases, two or more detectors are configured to detect the same wavelength or wavelength range. In yet another case, one or more detectors configured to detect a specific wavelength or wavelength range different from one or more other detectors). In embodiments employing multiple light paths, the PPG device may determine an average of the signals resulting from the multiple light paths before determining an HR estimate or other physiological metrics.

[0065] Moreover, in an embodiment, the emitters 416 and detectors 418 may be coupled to the controller 402 directly or indirectly using driver circuitry by which the controller 402 may drive the emitters 216 and obtain signals from the detectors 418. The host computer 422 can communicate with the wireless networking components 412 via the one or more networks 420, which may include one or more local area networks, wide area networks, UWB, and / or internetworks using any of terrestrial or satellite links. In some embodiments, the host computer 422 executes control programs and / or application programs that are configured to perform some of the functions described herein.

[0066] Referring now to FIG. 5, a schematic diagram of an environment 500 in which aspects of various embodiments can be implemented is illustrated. In particular, as shown, a wearer might have a number of different devices that are able to communicate using at least one wireless communication protocol. For example, as shown, the wearer might have a wearable sensing device 100, which the wearer would like to be able to communicate with a smartphone 504 and a tablet computer 506. The ability to communicate with multiple devices can enable a wearer to obtain information from the wearable sensing device 100, e.g., data captured using a sensor on the wearable sensing device 100, using an application installed on either the smartphone 504 or another such device 506 associated with that wearer, such as but not limited to a tablet, personal computer, smartwatch, and the like. The system may also automatically and periodically sync data from the wearable sensing device 100 to a service provider 508, or other such entity, which is able to obtain and process data from the wearable sensing device 100 and provide functionality that may not otherwise be available on the wearable sensing device 100 or the applications installed on the individual devices. In this embodiment, the wearable sensing device 100 automatically communicates via Bluetooth® with one of the individual devices (such as smartphone 504 or tablet computer 506) to periodically sync data in the background without requiring user intervention, and the individual device then communicates with the service provider 508 through at least one network 220, such as the Internet or a cellular network. There may be a number of other types of, or reasons for, communications in various embodiments.

[0067] In addition to being able to communicate, the system may also need to communicate in a number of ways or with certain aspects. For example, communications between the devices should be secure, particularly where the data may include personal health data or other such communications. The device or application providers may also be required to secure this information in at least some situations. The devices should be able to communicate with each other concurrently, rather than sequentially, to enable seamless background data synchronization. This may be particularly true where pairing may be required, as it is preferable that each device be paired at most once, such that no manual pairing is required for subsequent automatic sync operations. The communications should also be as standards-based as possible, not only so that little manual intervention is required but also so that the devices can automatically communicate with as many other types of devices as possible, which is often not the case for various proprietary formats. The system is thus designed to enable automatic communication between devices with little to no effort required from the wearer, allowing for seamless background data synchronization when devices are within communication range. In various conventional approaches, a device will utilize a communication technology such as Wi-Fi to communicate with other devices using wireless local area networking (WLAN). Smaller or lower capacity devices, such as many Internet of Things (loT) devices, instead utilize a communication technology such as Bluetooth®, and in particular Bluetooth Low Energy (BLE) which has very low power consumption.

[0068] In further embodiments, the environment 500 illustrated in FIG. 5 enables data to be captured, processed, and displayed in a number of different ways. For example, data may be captured using sensors on the wearable sensing device 100, but due to limited resources on the wearable sensing device 100, the data may be transferred to the smartphone 504 or the service provider 508 (or a cloud resource) for processing, and results of that processing may then be presented back to the wearer of the wearable sensing device 200 via a smartphone 504, and / or another such device 506 associated with that wearer, such as but not limited to a table, personal computer, smartwatch, and the like. In at least some embodiments, a wearer may also be able to provide input such as health data using an interface on any of these devices, which can then be considered when making that determination.

[0069] Created Body Cavity

[0070] It should be understood that the disclosed device(s) can be utilized in a variety of locations on / in a human body, although for many individuals a body piercing location such as one or both ear lobes may be particular preferred (and such locations may already be pierced to accept a variety of ornamentation). Various piercing locations may be suitable for biometric measurement and sensing of temperature, heart rate at rest, heart rate during activity, heart rate variability, heart beat sensing, oxygen sensing, blood oxygen level, blood pressure, resting pulse rate, active pulse rate, perfusion, breathing rate, movement, sleep stages (light, non-REM, REM, deep, awake), active energy expenditure, resting energy expenditure, total energy expenditure, SpC , and / or glucose, measurements of bacteria, white blood cell count, proteins, lipids, salts, fats (or other fluid characteristics, including lymph or other fluid characteristics) as well as a variety of other data including location (via GPS), sound sensors, accelerometer data, etc. In some cases, localized body conditions proximate to a specific piercing location may be particularly well suited for measurement of various physiological conditions, such as a tongue piercing for tracking glucose measurement and / or blood sugar levels.

[0071] The disclosed CCT is a measurement that may be obtained from an artificially created cavity in the wearer’s body, where the created cavity is formed artificially, as in a non-limiting example, an earring piercing procedure. Additionally, two or more artificially created cavities can be utilized for measuring CCT in one body. As one non-limiting example, two different biosensing devices or wearable sensing devices can be used in two different created cavities to measure two CCT values (or other anatomical metrics) simultaneously within separate created cavities. Further, these different sensing devices can be used to derive a single CCT (or other anatomical metrics). It should be understood that the CCT (or other anatomical metrics) from one created cavity in a wearer’s body may by different from the CCT (or other anatomical metrics) obtained from another created cavity in the same wearer’s body, as anatomical differences may induce localize temperature variations (e.g., the CCT or other metric of an ear piercing may not be the same as the CCT or other metric of a belly button or tongue piercing of the same wearer, or the CCTs or other metrics in opposing ear piercings may be different for a variety of environment and / or anatomical reasons).

[0072] The disclosed device can desirably measure CCT or other metrics within the created cavity to track the wearer’s useful temperature or for other purposes. The aspects of the created cavity are further described below. CCT may or may not be the same temperature measurement as core body temperature (CBT), internal body temperature (IBT), basal body temperature (BBT) body cavity temperature (BCT), and / or surface body temperature (SBT). This can be recognized by reason that the temperature inside the created cavity does not need to equal and / or correlate on a relationship basis with CBT, IBT, BBT, BCT and / or SBT. Moreover, it is suitable if CCT measurements are a perfect match, higher, lower, or not a 1 :1 relationship to body temperature.

[0073] The disclosed device can offer a practical design that provides means for a comfortable, convenient, and unobtrusive method of continuously or periodically tracking the wearer’s temperature or other biometrics while sleeping, rest or activity. For example, but not limited to, the shape of the device resembles a small, stud earring structure that is wearable on the body, through a wearer’s earlobe, wearer’s belly button, tongue, nose, eyebrow, lip, genitalia and / or other body locations. This device allows for wireless communication to an external system for tracking continuous or periodic measurements, which can be used for the nonlimiting example of determination of fertility by identifying changes in body temperature associated with the biological event of ovulation.

[0074] By means of the disclosed device, cavity temperature measurements and / or other measurements are obtainable from a wearer over a defined period of time. In a non-limiting example, cavity temperature measurements can help assist in the determination of fertility. Within this example, readings are evaluated to differentiate the local minimum during a given night’s sleep, where the local minimum is defined by the lowest reading within a pre-established time period. Then local minima can be plotted in relation to time for fertility trends. Additionally, measurements can be tracked over 24 hours and temperatures compared across different days and nights to determine patterns, for a non-limiting example, to identify or predict the wear’s date of ovulation.

[0075] Cycle and Ovulation / Fertility Detection

[0076] 12.3 percent of women (7.5 million) in the United States ages 15-44 have impaired ability getting or staying pregnant, according to the Centers for Disease Control and Prevention (CDC). In 40% of such women, ovulatory defects are present but difficult to characterize because ovulation is an internal, normally unmonitored clinical state that evolves quickly over short time frames. Defining the rapidly changing physiology of ovulation for each affected woman requires frequent monitoring during suspected ovulation windows. Moreover, while the rhythm method of birth control (e.g., tracking a menstrual cycle on a calendar to predict ovulation) is over 75% effective at preventing unwanted pregnancy, less than 1% of women 15 to 44 years of age currently use this highly effective natural form of birth control.

[0077] In predicting ovulation for both pregnancy and birth control objectives, the need for frequent monitoring currently translates to significant patient burdens. The use of basal body temperature (BBT) recordings is a known and safe method to monitor ovulation and has the advantages of patient monitoring themselves at home (no clinical scheduling or attendance) at minimal cost and risk compared to a blood tests or ultrasounds. However, application of BBT methods is limited by the inconvenience of taking and recording a waking temperature at the same time each morning - patient compliance is especially poor. Even for those who comply, the information is difficult to interpret and often frustrating for the patient.

[0078] Traditional fertility thermometers assist in tracking a woman’s ovulation trends by measuring her basal body temperature (BBT) through a natural body cavity. These devices do not offer the most accurate ovulation results due to temperature measurements taken after awakening with a non-convenient thermometer. In contrast, the disclosed devices and associated system components will desirably eliminate any need to awaken before temperature measurements can be taken and more accurately identify the low temperature within a given night’s sleep, because the low temperature does not necessarily occur at the time of waking. This is desirably accomplished by a small, wearable sensor such as a temperature sensor located within an artificial created cavity in the body that offers continuous or periodic readings of CCT that are wirelessly transmitted and analyzed by the wearer’s associated smart device.

[0079] In one exemplary embodiment, the disclosed earring technology and systems can track a wearer’s unique menstrual cycle (menstruation phase, follicular phase, ovulation phase, luteal phase) and provide insights to the specific wearer based on their current phase in terms of needs related to rest, sleep, nutrition, exercise, and the like. The system conveniently monitors ovulation by measuring basal body temperature (BBT) utilizing earring sensors - similar in visual appearance to the earrings that many women in the US wear daily. Since the monitoring occurs at home, the inconvenience and cost to the patient is mitigated and care can be delivered more equitably. Moreover, the applications of this disclosed device can go far beyond infertility and can extend to use in identifying ovulation I avoiding pregnancy in fertility management, assist with wellness tracking, perform infection I COVID monitoring and / or other uses, provide early pregnancy detection, assist with natural birth control, identify pregnancy, labor and menopause onset, and / or track a wearer’s menstrual cycle and / or other health metrics. In various embodiments, the earring may include features which may alert the wearer (i.e., using sound, vibration, light and / or electrical pulses to the wearer) of vital information, such as device proximity or non-proximity (i.e., phone theft prevention) or the receipt of emails, text messages, information updates or phone calls, etc. The disclosed devices and related system components also provide capabilities to obtain information based on early detection of one’s health issues or conditions, like infection, COVID19, virus, pregnancy, early labor, perimenopause, women specific conditions or needing to take specific medications or vitamins. This device could also be used for other medical applications, by way of non-limiting example as seen in detection of fever and / or sickness. Further, this device can be used for deriving information for medical tracking and implemented for hospital use in circumstances that require constant or periodic temperature and vital and biometric monitoring, and it should be understood that the device can be used in the context of hospital systems or other medical environments for in-patient and / or remote patient monitoring, where the device can be used to relay information and can be linked to medical provider records. Additional applications contemplated herein include, but are not limited to, fertility, infertility, natural family planning, COVID infection and / or condition, infectious disease symptoms and / or susceptibility, cancer diagnoses and treatment, long term hospital patient monitoring, fever detection and treatment, thyroid issues, gut health, inflammation, hormone health, hormone levels, energy levels, cardiovascular health, post-partum conditions, puberty, PCOS, diabetes detection and management, obesity, asthma, heart disease, chronic obstructive pulmonary disease, women's health issues, wellness management, stress management, sleep monitoring and disorder detection and treatment, diet and calorie counting, activity / fitness tracking, sports, government health, military, agriculture, etc.

[0080] In one embodiment of the present disclosure, a computing system is contemplated that includes a first machine-learned model trained to identify one or more phases of a menstrual cycle of a wearer of the wearable sensing device. The first machine-learned model uses training data obtained from a plurality of individuals that may or may not include the wearer. The computing system also includes a second machine-learned model trained to identify the one or more phases of the menstrual cycle of the wearer of the wearable sensing device. In contrast to the first machine-learned model, the second machine-learned model uses training data obtained only from the wearer. The system also includes one or more processors and memory storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations. The operations include: obtaining biometric parameter data from the wearer as measured via the wearable sensing device; inputting the biometric parameter data into the first machine-learned model for a first period of time, the first period of time being sufficient for an adequate amount of biometric parameter data from the wearer to be obtained by the wearable sensing device to be used as the training data to train the second machine-learned model; receiving, as an output of the first machine-learned model, the identification of the one or more phases of the menstrual cycle of the wearer; after the first period of time has elapsed, inputting the biometric parameter data obtained by the wearable sensing device into the second machine-learned model; and receiving, as an output of the second machine-learned model, the identification of the one or more phases of the menstrual cycle of the wearer.

[0081] In addition, it is to be understood that the first machine-learned model is not used after the first period of time has elapsed so that only the second machine- learned model is used to identify the one or more phases of the menstrual cycle of the wearer after the first period of time has elapsed using biometric parameter data that is continuously input into the second machine-learned model.

[0082] Further, the one or more biometric data parameters can include created cavity temperature, SpO2, active energy expenditure, resting energy expenditure, total energy expenditure, sleep metrics, energy levels, hormone levels, heart rate, heart rate variability, physical activity, or a combination thereof. Further, data corresponding with the one or more biometric data parameters and information related to the identification of the one of more phases of the menstrual cycle of the wearer can be input into a machine-learned model that outputs predictions related to the wearer’s physical health, mental health, stress level, readiness, restorative shift, resilience, or a combination thereof, as discussed in the appendices that follow, which are incorporated herein by reference.

[0083] Additionally, it should be understood that the machine-learned models contemplated by the present disclosure can provide insights or recommendations to a user based on the user's individual biometric data. These insights and / or recommendations may all be calculated or provided from the foundation on a current phase of the user’s menstrual cycle (i.e., when the user naturally cycling in reproductive years) or not related to the user’s menstrual cycle phases (i.e., the user is not naturally cycling, the user is on hormonal birth control, the user is postmenopausal, etc). For example, sleep recommendations may or may not be based on the user's current menstrual cycle phase.

[0084] Referring now to FIG. 6, an example computing system 600 for a machinelearning based outcome predictor determination or a wearer-specific recommendation determination based on one or more biometric data parameters for the wearer is provided. FIG. 6 depicts an example computing system 600 for machine-learning-based identification of the phases of a wearer’s menstrual cycle, predictions of health features (physical, mental, stress, sleep, readiness, restoration, resilience, etc.), and recommendations to provide to the wearer based on the biometric parameter data, among other outputs, according to example embodiments of the present disclosure. The example system 600 includes a computing device 602 and a machine learning computing system 630 that are communicatively coupled over a network 680.

[0085] The computing device 602 includes one or more processors 612 and a memory 614. The one or more processors 612 can be any suitable processing device (e.g., a processor core, a microprocessor, an ASIC, a FPGA, a controller, a microcontroller, etc.) and can be one processor or a plurality of processors that are operatively connected. The memory 614 can include one or more non-transitory computer-readable storage media, such as RAM, ROM, EEPROM, EPROM, one or more memory devices, flash memory devices, etc., and combinations thereof.

[0086] The memory 614 can store information that can be accessed by the one or more processors 612. For instance, the memory 614 (e.g., one or more non- transitory computer-readable storage mediums, memory devices) can store data 616 that can be obtained, received, accessed, written, manipulated, created, and / or stored. In some implementations, the computing device 602 can obtain data from one or more memory device(s) that are remote from the device 602.

[0087] The memory 614 can also store computer-readable instructions 618 that can be executed by the one or more processors 612. The instructions 618 can be software written in any suitable programming language or can be implemented in hardware. Additionally, or alternatively, the instructions 618 can be executed in logically and / or virtually separate threads on processor(s) 612.

[0088] For example, the memory 614 can store instructions 618 that when executed by the one or more processors 612 cause the one or more processors 612 to perform any of the instructions, operations, and / or functions described herein. According to an aspect of the present disclosure, the computing device 602 can store or include one or more machine-learned models 610. For example, the models 610 can be or can otherwise include various machine-learned models such as a random forest classifier; a logistic regression classifier; a support vector machine; one or more decision trees; a neural network; and / or other types of models including both linear models and non-linear models. Example neural networks include feed-forward neural networks, recurrent neural networks (e.g., long short-term memory recurrent neural networks), convolutional neural networks, deep neural networks, transformer networks utilizing attention mechanisms, or other forms of neural networks. The models 610 can also include deep learning architectures that employ attention algorithms, such as self-attention and multihead attention mechanisms, which can be particularly useful for processing sequential sensor data and identifying relevant patterns across different time periods. The models 610 can also include large language models, generative Al models, foundation models, or a combination thereof. It should be understood that a foundation model can be a model trained through self-supervision, a cross-modal model, a knowledge distillation model, a multi-site model, or a combination thereof, and such models allow for the inclusion of information gained from either physiological data collected from devices on other body locations or other data modes (e.g., language, image, electronic health record data, etc.) with the data gathered from the wearable sensing device (e.g., earring). Further, the models can be, in some instances, trained with one or more prompts.

[0089] In some implementations, the computing device 602 can receive the one or more machine-learned models 610 from the machine learning computing system 630 over network 680 and can store the one or more machine-learned models 610 in the memory 614. The computing device 602 can then use or otherwise run the one or more machine-learned models 610 (e.g., by processor(s) 612).

[0090] The machine learning computing system 630 includes one or more processors 632 and a memory 634. The one or more processors 632 can be any suitable processing device (e.g., a processor core, a microprocessor, an ASIC, a FPGA, a controller, a microcontroller, etc.) and can be one processor or a plurality of processors that are operatively connected. The memory 634 can include one or more non-transitory computer-readable storage media, such as RAM, ROM, EEPROM, EPROM, one or more memory devices, flash memory devices, etc., and combinations thereof.

[0091] The memory 634 can store information that can be accessed by the one or more processors 632. For instance, the memory 634 (e.g., one or more non- transitory computer-readable storage mediums, memory devices) can store data 636 that can be obtained, received, accessed, written, manipulated, created, and / or stored. In some implementations, the machine learning computing system 630 can obtain data from one or more memory device(s) that are remote from the system 630.

[0092] The memory 634 can also store computer-readable instructions 638 that can be executed by the one or more processors 632. The instructions 638 can be software written in any suitable programming language or can be implemented in hardware. Additionally, or alternatively, the instructions 638 can be executed in logically and / or virtually separate threads on processor(s) 632. For example, the memory 634 can store instructions 638 that when executed by the one or more processors 632 cause the one or more processors 632 to perform any of the operations and / or functions described herein.

[0093] In some implementations, the machine learning computing system 630 includes one or more server computing devices. If the machine learning computing system 630 includes multiple server computing devices, such server computing devices can operate according to various computing architectures, including, for example, sequential computing architectures, parallel computing architectures, or some combination thereof.

[0094] In addition or alternatively to the model(s) 610 at the computing device 602, the machine learning computing system 630 can include one or more machine- learned models 640. For example, the models 640 can be or can otherwise include various machine-learned models such as a random forest classifier; a logistic regression classifier; a support vector machine; one or more decision trees; a neural network; and / or other types of models including both linear models and nonlinear models. Example neural networks include feed-forward neural networks, recurrent neural networks (e.g., long short-term memory recurrent neural networks), convolutional neural networks, deep neural networks, transformer networks utilizing attention mechanisms, or other forms of neural networks. The models 640 can also include deep learning architectures that employ attention algorithms, such as self-attention and multi-head attention mechanisms, which can enable the machine learning computing system 630 to process complex sensor data patterns and perform advanced analytics that may not be feasible on the local computing device 602. The models 640 can also include large language models, generative Al models, foundation models, or a combination thereof. It should be understood that a foundation model can be a model trained through selfsupervision, a cross-modal model, a knowledge distillation model, a multi-site model, or a combination thereof, and such models allow for the inclusion of information gained from either physiological data collected from devices on other body locations or other data modes (e.g., language, image, electronic health record data, etc.) with the data gathered from the wearable sensing device (e.g., earring). Further, the models can be, in some instances, trained with one or more prompts.

[0095] As an example, the machine learning computing system 630 can communicate with the computing device 602 according to a client-server relationship. For example, the machine learning computing system 640 can implement the machine-learned models 640 to provide a web service to the computing device 602.

[0096] Thus, machine-learned models 610 can located and used at the computing device 602 and / or machine-learned models 640 can be located and used at the machine learning computing system 630.

[0097] In some implementations, the machine learning computing system 630 and / or the computing device 602 can train the machine-learned models 610 and / or 640 through use of a model trainer 660. The model trainer 660 can train the machine-learned models 610 and / or 640 using one or more training or learning algorithms. One example training technique is backwards propagation of errors (“backpropagation”).

[0098] In some implementations, the model trainer 660 can perform supervised training techniques using a set of labeled training data 662. In other implementations, the model trainer 660 can perform unsupervised training techniques using a set of unlabeled training data. The model trainer 660 can perform a number of generalization techniques to improve the generalization capability of the models being trained. Generalization techniques include weight decays, dropouts, or other techniques. The model trainer 660 can be implemented in hardware, software, firmware, or combinations thereof.

[0099] The computing device 602 can also include a network interface 624 used to communicate with one or more systems or devices, including systems or devices that are remotely located from the computing device 602. The network interface 624 can include any circuits, components, software, etc. for communicating with one or more networks (e.g., 680). In some implementations, the network interface 624 can include, for example, one or more of a communications controller, receiver, transceiver, transmitter, port, conductors, software and / or hardware for communicating data. Similarly, the machine learning computing system 630 can include a network interface 664.

[0100] The computing device 602 can also include one or more sensors 604. For example, the one or more sensors 604 can include any type of sensor useful to collect test data from a wearer, including, for example, various forms of biometric data as described above. The computing device 602 can also include a wearer input component 620. For example, the wearer input component 620 can include a microphone, a keypad, a keyboard, a click-wheel, buttons, and / or a touch-sensitive screen.

[0101] The computing device 602 can also include an output component 622. For example, the output component 622 can include a speaker, a haptic output component, and / or a display (e.g., a touch-sensitive display).

[0102] As another example, the computing device 602 can transmit information to one or more additional devices 670 (e.g., a smartphone, smartwatch, tablet, personal computer, etc.). The computing device 602 can communicate with the additional computing device(s) 670 over the network 680 and / or via a local, short- range wireless communication protocol (e.g., Bluetooth).

[0103] The network(s) 680 can be any type of network or combination of networks that allows for communication between devices. In some embodiments, the network(s) can include one or more of a local area network, wide area network, the Internet, secure network, cellular network, mesh network, peer-to-peer communication link and / or some combination thereof and can include any number of wired or wireless links. Communication over the network(s) 680 can be accomplished, for instance, via a network interface using any type of protocol, protection scheme, encoding, format, packaging, etc.

[0104] FIG. 6 illustrates one example computing system 600 that can be used to implement the present disclosure. Other computing systems can be used as well. For example, in some implementations, the computing device 602 can include the model trainer 660 and the training dataset 662. In such implementations, the machine-learned models 610 can be both trained and used locally at the computing device 602. As another example, in some implementations, the computing device 602 is not connected to other computing systems.

[0105] Boilerplate

[0106] While various embodiments of the present disclosure have been described above, it should be understood that they have been presented by way of example only, and not by way of limitation. Likewise, the various diagrams may depict an example architectural or other configuration for the disclosure, which is done to aid in understanding the features and functionality that can be included in the disclosure. The disclosure is not restricted to the illustrated example architectures or configurations but can be implemented using a variety of alternative architectures and configurations. Additionally, although the disclosure is described above in terms of various exemplary embodiments and implementations, it should be understood that the various features and functionality described in one or more of the individual embodiments are not limited in their applicability to the particular embodiment with which they are described. They instead can be applied, alone or in some combination, to one or more of the other embodiments of the disclosure, whether or not such embodiments are described, and whether or not such features are presented as being a part of a described embodiment. Thus, the breadth and scope of the present disclosure should not be limited by any of the abovedescribed exemplary embodiments.

[0107] Unless otherwise defined, all terms (including technical and scientific terms) are to be given their ordinary and customary meaning to a person of ordinary skill in the art, and are not to be limited to a special or customized meaning unless expressly so defined herein. It should be noted that the use of particular terminology when describing certain features or aspects of the disclosure should not be taken to imply that the terminology is being re-defined herein to be restricted to include any specific characteristics of the features or aspects of the disclosure with which that terminology is associated. Terms and phrases used in this application, and variations thereof, especially in the appended claims, unless otherwise expressly stated, should be construed as open ended as opposed to limiting. As examples of the foregoing, the term ‘including’ should be read to mean ‘including, without limitation,’ ‘including but not limited to,’ or the like; the term ‘comprising’ as used herein is synonymous with ‘including,’ ‘containing,’ or ‘characterized by,’ and is inclusive or open-ended and does not exclude additional, unrecited elements or method steps; the term ‘having’ should be interpreted as ‘having at least;’ the term ‘includes’ should be interpreted as ‘includes but is not limited to;’ the term ‘example’ is used to provide exemplary instances of the item in discussion, not an exhaustive or limiting list thereof; adjectives such as ‘known’, ‘normal’, ‘standard’, and terms of similar meaning should not be construed as limiting the item described to a given time period or to an item available as of a given time, but instead should be read to encompass known, normal, or standard technologies that may be available or known now or at any time in the future; and use of terms like ‘preferably,’ ‘preferred,’ ‘desired,’ or ‘desirable,’ and words of similar meaning should not be understood as implying that certain features are critical, essential, or even important to the structure or function of the present disclosure, but instead as merely intended to highlight alternative or additional features that may or may not be utilized in a particular embodiment of the present disclosure. Likewise, a group of items linked with the conjunction ‘and’ should not be read as requiring that each and every one of those items be present in the grouping, but rather should be read as ‘and / or1unless expressly stated otherwise. Similarly, a group of items linked with the conjunction ‘oh should not be read as requiring mutual exclusivity among that group, but rather should be read as ‘and / or1unless expressly stated otherwise.

[0108] Where a range of values is provided, it is understood that the upper and lower limit, and each intervening value between the upper and lower limit of the range is encompassed within the embodiments. For instance, when a plurality of ranges are provided, any combination of a minimum value and a maximum value described in the plurality of ranges are contemplated by the present disclosure. For example, if ranges of ‘from about 20% to about 80%’ and ‘from about 30% to about 70%’ are described, a range of ‘from about 20% to about 70%’ or a range of ‘from about 30% to about 80%’ are also contemplated by the present disclosure.

[0109] With respect to the use of substantially any plural and / or singular terms herein, those having skill in the art can translate from the plural to the singular and / or from the singular to the plural as is appropriate to the context and / or application. The various singular / plural permutations may be expressly set forth herein for sake of clarity. The indefinite article ‘a’ or ‘an’ does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims should not be construed as limiting the scope.

[0110] It will be further understood by those within the art that if a specific number of an introduced claim recitation is intended, such an intent will be explicitly recited in the claim, and in the absence of such recitation no such intent is present. For example, as an aid to understanding, the following appended claims may contain usage of the introductory phrases ‘at least one’ and ‘one or more’ to introduce claim recitations. However, the use of such phrases should not be construed to imply that the introduction of a claim recitation by the indefinite articles ‘a’ or ‘an’ limits any particular claim containing such introduced claim recitation to embodiments containing only one such recitation, even when the same claim includes the introductory phrases 'one or more” or ‘at least one’ and indefinite articles such as ‘a’ or ‘an’ (e.g., ‘a’ and / or ‘an’ should typically be interpreted to mean ‘at least one’ or “one or more”); the same holds true for the use of definite articles used to introduce claim recitations. In addition, even if a specific number of an introduced claim recitation is explicitly recited, those skilled in the art will recognize that such recitation should typically be interpreted to mean at least the recited number (e.g., the bare recitation of ‘two recitations,’ without other modifiers, typically means at least two recitations, or two or more recitations). Furthermore, in those instances where a convention analogous to ‘at least one of A, B, and C, etc.’ is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., ‘a system having at least one of A, B, and C’ would include but not be limited to systems that have A alone, B alone, C alone, A and B together, A and C together, B and C together, and / or A, B, and C together, etc.). In those instances where a convention analogous to ‘at least one of A, B, or C, etc.’ is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., ‘a system having at least one of A, B, or C’ would include but not be limited to systems that have A alone, B alone, C alone, A and B together, A and C together, B and C together, and / or A, B, and C together, etc.). It will be further understood by those within the art that virtually any disjunctive word and / or phrase presenting two or more alternative terms, whether in the description, claims, or drawings, should be understood to contemplate the possibilities of including one of the terms, either of the terms, or both terms. For example, the phrase ‘A or B’ will be understood to include the possibilities of ‘A or ‘B’ or ‘A and B.’

[0111] All numbers expressing quantities of ingredients, reaction conditions, and so forth used in the specification are to be understood as being modified in all instances by the terms ‘about,’ ‘approximately,’ or ‘generally.’ Accordingly, unless indicated to the contrary, the numerical parameters set forth herein are approximations that may vary depending upon the desired properties sought to be obtained. At the very least, and not as an attempt to limit the application of the doctrine of equivalents to the scope of any claims in any application claiming priority to the present application, each numerical parameter should be construed in light of the number of significant digits and ordinary rounding approaches. As used herein, the terms ‘about,’ ‘approximately,’ or ‘generally,’ when used to modify a value, indicate that the value can be raised or lowered by 5% and remain within the disclosed embodiment.

[0112] All of the features disclosed in this specification (including any accompanying exhibits, claims, abstract and drawings), and / or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of such features and / or steps are mutually exclusive. The disclosure is not restricted to the details of any foregoing embodiments. The disclosure extends to any novel one, or any novel combination, of the features disclosed in this specification (including any accompanying claims, abstract and drawings), or to any novel one, or any novel combination, of the steps of any method or process so disclosed.

[0113] While the present subject matter has been described in detail with respect to various specific example embodiments thereof, each example is provided by way of explanation, not limitation of the disclosure. Those skilled in the art, upon attaining an understanding of the foregoing, can readily produce alterations to, variations of, and equivalents to such embodiments. Accordingly, the subject disclosure does not preclude inclusion of such modifications, variations and / or additions to the present subject matter as would be readily apparent to one of ordinary skill in the art. For instance, features illustrated or described as part of one embodiment can be used with another embodiment to yield a still further embodiment. Thus, it is intended that the present disclosure covers such alterations, variations, and equivalents.

[0114] Privacy

[0115] It should be understood that biometric data collection may include sensitive and confidential information relating to a wearer, such as but not limited to genetic predispositions, cardiovascular health, pregnancy, etc. Accordingly, and further to the descriptions above, some or all data acquired using the products or end products of the aforementioned systems and methods will be kept private and confidential for each specific wearer unless the wearer agrees to share such data. Thus, it should not be construed that any information discovered or inferred from the use of the products or end products produced using the aforementioned systems and methods will be improperly used or published. For example, information acquired from the wearable devices described herein may be treated so that no person without express or implied consent is capable of accessing said information. Thus, the information acquired from the wearable device may be kept confidential and access to the information may be controlled exclusively by the wearer of the wearable device whose information is determined by using the wearable device.

Claims

WHAT IS CLAIMED IS:1 . A computing system, the computing system comprising: a first machine-learned model trained to identify one or more phases of a menstrual cycle of a wearer of a wearable sensing device, wherein the first machine- learned model uses training data obtained from a plurality of individuals that may or may not include the wearer; a second machine-learned model trained to identify the one or more phases of the menstrual cycle of the wearer of the wearable sensing device, wherein the second machine-learned model uses training data obtained only from the wearer; one or more processors; and memory storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations, the operations comprising: obtaining biometric parameter data from the wearer as measured via the wearable sensing device; inputting the biometric parameter data into the first machine-learned model for a first period of time, the first period of time being sufficient for an adequate amount of biometric parameter data from the wearer to be obtained by the wearable sensing device to be used as the training data to train the second machine-learned model; receiving, as an output of the first machine-learned model, an identification of the one or more phases of the menstrual cycle of the wearer; after the first period of time has elapsed, inputting the biometric parameter data obtained by the wearable sensing device into the second machine-learned model; and receiving, as an output of the second machine-learned model, the identification of the one or more phases of the menstrual cycle of the wearer.

2. The computing system of claim 1 , wherein the first machine-learned model is not used after the first period of time has elapsed so that only the second machine-learned model is used to identify the one or more phases of the menstrualcycle of the wearer after the first period of time has elapsed using biometric parameter data that is continuously input into the second machine-learned model.

3. The computing system of claim 1 or 2, wherein the first period of time comprises at least one complete menstrual cycle.

4. The computing system of any of the preceding claims, wherein the one or more phases of the menstrual cycle comprise a menstruation phase, a follicular phase, an ovulation phase, and a luteal phase.

5. The computing system of any of the preceding claims, wherein the wearable sensing device comprises a first wearable sensing device and a second wearable sensing device, wherein the first wearable sensing device and the second wearable sensing device are each inserted into a created cavity of the wearer.

6. The computing system of claim 5, wherein the first wearable sensing device and the second wearable sensing device measure the one or more biometric data parameters simultaneously.

7. The computing system of claim 5, wherein the first wearable sensing device measures the one or more biometric data parameters while the second wearable sensing device is not activated and vice versa.

8. The computing system of any of the preceding claims, wherein the one or more biometric data parameters comprises created cavity temperature, SpC>2, active energy expenditure, resting energy expenditure, total energy expenditure, sleep metrics, heart rate, heart rate variability, physical activity, or a combination thereof.

9. The computing system of any of the preceding claims, wherein the first machine-learned model, the second machine-learned model, or both comprises one or more of a deep artificial neural network, a transformer network with attention mechanisms, a support vector machine, a decision tree, or a linear model.

10. The computing system of any of the preceding claims, wherein data corresponding with the one or more biometric data parameters and information related to the identification of the one of more phases of the menstrual cycle of the wearer are input into a third machine-learned model, wherein the third machine- learned model outputs predictions related to the wearer’s physical health, mental health, stress level, readiness, restorative shift, resilience, or a combination thereof.11 . The computing system of claim 10, wherein the third machine-learned model comprises a large language model, a generative Al model, a foundation model, or a combination thereof.

12. The computing system of any of the preceding claims, wherein data corresponding with the one or more biometric data parameters and information related to the identification of the one of more phases of the menstrual cycle of the wearer input into a fourth machine-learned model, wherein the fourth machine- learned model outputs recommendations to improve the wearer’s physical health, mental health, stress level, readiness, restorative shift, resilience, or a combination thereof.

13. The computing system of claim 12, wherein the fourth machine-learned model comprises a large language model, a generative Al model, a foundation model, or a combination thereof.

14. A method, the method comprising: obtaining, by a computing system, biometric parameter data from a wearer of a wearable sensing device as measured via the wearable sensing device; inputting, by the computing system, the biometric parameter data into a first machine-learned model for a first period of time, the first period of time being sufficient for an adequate amount of biometric parameter data from the wearer to be obtained by the wearable sensing device to be used as training data to train a second machine-learned model; receiving, by the computing system as an output of the first machine-learned model, an identification of one or more phases of a menstrual cycle of the wearer; after the first period of time has elapsed, inputting, by the computing system, the biometric parameter data obtained by the wearable sensing device into the second machine-learned model; and receiving, by the computing system as an output of the second machine- learned model, the identification of the one or more phases of the menstrual cycle of the wearer.

15. The method of claim 14, wherein the first machine-learned model is trained to identify the one or more phases of the menstrual cycle of the wearer using training data obtained from a plurality of individuals that may or may not include the wearer.

16. The method of claim 14 or 15, wherein the second machine-learned model is trained to identify the one or more phases of the menstrual cycle of the wearer of the wearable sensing device, wherein the second machine-learned model uses training data obtained only from the wearer.

17. The method of any of claims 14 to 16, wherein the first machine-learned model is not used after the first period of time has elapsed so that only the second machine-learned model is used to identify the one or more phases of the menstrual cycle of the wearer after the first period of time has elapsed using biometric parameter data that is continuously input into the second machine-learned model.

18. The method of any of claims 14 to 17, wherein the wearable sensing device comprises a first wearable sensing device and a second wearable sensing device, wherein the first wearable sensing device and the second wearable sensing device are each inserted into a created cavity of the wearer.

19. The method of claim 18, further comprising measuring the one or more biometric data parameters with the first wearable sensing device and the second wearable sensing device simultaneously.

20. The method of claim 18, further comprising measuring the one or more biometric data parameters with the first wearable sensing device while the second wearable sensing device is not activated and vice versa.21 . The method of any of claims 14 to 20, wherein the one or more biometric data parameters comprises created cavity temperature, SpO2, active energy expenditure, resting energy expenditure, total energy expenditure, sleep metrics, heart rate, heart rate variability, physical activity, or a combination thereof.

22. The method of any of claims 14 to 21 , wherein the first machine-learned model, the second machine-learned model, or both comprises one or more of a deep artificial neural network, a transformer network with attention mechanisms, a support vector machine, a decision tree, or a linear model.

23. The method of any of claims 14 to 22, further comprising: inputting, via the computing system, data corresponding with the one or more biometric data parameters and information related to the identification of the one of more phases of the menstrual cycle of the wearer into a third machine- learned model; and outputting, by the computing system via the third machine learned model, predictions related to the wearer’s physical health, mental health, stress level, readiness, restorative shift, resilience, or a combination thereof.

24. The method of claim 23, wherein the third machine-learned model comprises a large language model, a generative Al model, a foundation model, or a combination thereof.

25. The method of any of claims 14 to 24, further comprising: inputting, via the computing system, data corresponding with the one or more biometric data parameters and information related to the identification of the one of more phases of the menstrual cycle of the wearer into a fourth machine- learned model; and outputting, by the computing system via the fourth machine-learned model, recommendations to improve the wearer’s physical health, mental health, stress level, readiness, restorative shift, resilience, or a combination thereof.

26. The method of claim 25, wherein the fourth machine-learned model comprises a large language model, a generative Al model, a foundation model, or a combination thereof.

Images