Personal health management devices

Abstract

A wearable personal health monitoring device for measuring personal health is provided, configured to detect photoplethysmography (PPG) waves generated by infrared, green light, or red light emitted from the device. The device includes an outward-facing top surface, a side surface, and a bottom surface facing the user's skin, a plurality of electrical contact sensors, a first of the plurality of sensors disposed on one of the top, side, and bottom surfaces of the personal health monitoring device, and a second of the plurality of sensors disposed on one of the top and side surfaces.

Description

[Technical Field]

[0001] This application is a continuation-in-part of U.S. patent application Ser. No. 17 / 494,908, entitled "Personal Healthcare Device," filed on October 6, 2021, which claims priority to U.S. Provisional Application Ser. No. 63 / 088,223, entitled "Personal Healthcare Device," filed on October 6, 2020, and to U.S. Provisional Application Ser. No. 63 / 402,642, entitled "Personal Healthcare Device," filed on August 31, 2022, each of which is incorporated herein by reference.

[0002] This application relates to wearable health management devices, and more particularly to devices that collect data and monitor biometric measurements of a user. [Background technology]

[0003] One of the challenges of using photoplethysmography (PPG) technology in smartwatches and other wearable devices is the difficulty of detecting pulse signals due to scattering of reflected light due to the location of the device. In the case of smartwatches and other wristband-type devices, light-emitting diodes (LEDs) and photodiodes (PDs) are typically positioned on the wrist, where relatively high bone mass, low levels of capillaries and veins, and poor skin sensor sealing cause poor reflection of the various wavelengths of light used in pulse signal detection. This is particularly problematic when high-frequency wavelengths of light are used for pulse signal detection, resulting in weak DC and AC signals, a high signal-to-noise ratio, and poor PPG signal quality.

[0004] In medical and clinical applications, this problem of detecting pulse signals is addressed by using fingertip sensors, an area of ​​the body with little bone and a wide presence of capillaries, allowing light to pass directly from the LED through the skin to the PD, where the skin can form a better seal with the LED and PD, resulting in a more uniform light scattering effect and more precise and efficient detection of the pulse wave.

[0005] Fingertip oximeters are an example of such devices. Typically, a wired or wireless clip is attached to the finger, and an LED emits infrared (IR) and near-infrared (NIR) light with wavelengths between 640 nm and 940 nm. The reflected light is used to precisely detect pulse waves and create a PPG signal, from which the RR interval (the interval between two consecutive heartbeats) and other important parameters, such as blood oxygen concentration (SpO2), heart rate (HR), heart rate variability (HRV), and blood pressure (BP), can be inferred. When other higher-frequency NIR wavelengths are used, it is also possible to estimate hemoglobin and glucose concentrations. However, such attachable fingertip devices are difficult to consider as wearable devices due to their physical configuration and the need to reattach them to the body each time a measurement is required. In addition, when frequent or even continuous measurements are required, current devices restrict the wearer's movement, which is inconvenient for daily or continuous use.

[0006] PPG technology is well established for use at the fingertip (where PPG technology can capture high-quality readings because there are no bones and the skin at the fingertip forms a good seal with the sensor). In addition to traditional devices such as pulse oximeters that capture PPG via fingertip insertion, there are stand-alone devices that also capture high-quality PPG signals from the fingertip, such as smartphones. However, PPG technology does not appear to have been integrated into the side of a wrist-worn device. While side sensors exist for smartwatches, there are no devices that incorporate IR and NIR sensors into the side of a wearable device that would allow the wearer to place their fingertip on the side sensor to obtain high-quality PPG readings. Thus, there is a need for a wearable device that is not so limited.

[0007] Additionally, while wearable biometric monitors are available, most have limited functionality. For example, most are limited to measuring steps / distance walked and heart rate. People interested in a deeper profile and understanding of their health must do so through inconvenient visits to their healthcare professional, often involving invasive procedures. Furthermore, the reliability of the data generated when the device is moving is questionable, and the PPG signal can create artifacts in the measurements. Therefore, there is a need for a wearable device with multiple sensors that detects movement and provides timely, high-quality PPG data continuously or as needed, without the need for invasive procedures. Summary of the Invention

[0008] The present application provides a method, a wearable device, and a computer-readable medium for measuring personal health. According to one embodiment, the method includes detecting a photoplethysmography (PPG) signal by a sensor, where the PPG signal is generated by infrared, green, and / or red light emitted from one or more radiation sources of the personal health management device; transmitting the PPG signal data to a server, where the server processes the PPG signal data to infer biometric statistics based on machine-learned correlations generated from a training set of the PPG signal and the biometric data; receiving the biometric statistics from the server; and generating display data based on the biometric statistics.

[0009] The biometric statistics may include at least one of overall health, health changes, mood, sleep quality, fatigue, and stress. The biometric data includes at least one of heart rate, respiration rate, steps walked, calories burned, distance walked, sleep quality, ECG / EKG, blood pressure, mood, fatigue, body temperature, glucose level, blood alcohol, and blood oxygen. In one embodiment, the machine-learned correlations are based on PPG feature vectors including Kaiser-Teager power energy values, heart rate values, and spectral entropy values.

[0010] According to one embodiment, the wearable device includes at least one radiation source configured to generate a combination of at least two of infrared, red, and green light; a sensor configured to detect a photoplethysmography (PPG) signal based on the combination of light generated from the at least one radiation source; a network communication module configured to transmit the PPG signal to a server and receive biometric statistics from the server, where the server processes the PPG signal and infers the biometric statistics based on machine-learned correlations generated from a training set of the PPG signal and the biometric data; a processor configured to generate display data based on the biometric statistics; and a display configured to display the display data.

[0011] The wearable device may further include a wristband including a plurality of equally spaced holes to accommodate the mount having the stones thereon. The cross section of the holes may be hourglass-shaped to hold similarly shaped legs of the mount. The mount may be inserted through the holes in the wristband from the inside thereof. One set of legs of the mount may fit flush with the outside of the wristband. In one embodiment, the top of the mount includes at least one of gold, silver, copper, germanium, a magnet, and salt therein. In another embodiment, the top of the mount includes the stones.

[0012] The biometric statistics may include at least one of overall health, health changes, mood, sleep quality, fatigue, and stress. The biometric data may include at least one of heart rate, respiration rate, steps walked, calories burned, distance walked, sleep quality, ECG / EKG, blood pressure, mood, fatigue, body temperature, glucose level, blood alcohol, and blood oxygen. The machine-learned correlations may be based on PPG feature vectors including Kaiser-Tiger Power-Energy values, heart rate values, and spectral entropy values.

[0013] The wearable device may further include a planar inline sensor (FIS) including a near-field infrared (NIR) light-emitting diode (LED) with a signal length of approximately 1300 nanometers (nm), a NIR LED with a signal length of approximately 1550 nm, and a photodiode with a wavelength sensitivity range of 900 nm to 1700 nm. Light from the NIR LED may be directed toward the skin via two angled mirrors so that the light is reflected off blood glucose molecules back to the photodiode at a predetermined angle.

[0014] According to one embodiment, a non-transitory computer readable medium includes computer program code for detecting a photoplethysmography (PPG) signal by a sensor, the PPG signal being generated by infrared, green, or red light emitted from one or more radiation sources of the personal health management device; computer program code for transmitting the PPG signal to a server, the server processing the PPG signal to infer biometric statistics based on machine-learned correlations generated from the PPG signal and a training set of biometric data; computer program code for receiving the biometric statistics from the server; and computer program code for generating display data based on the biometric statistics.

[0015] The biometric statistics may include at least one of overall health, health changes, mood, sleep quality, fatigue, and stress. The biometric data may include at least one of heart rate, respiration rate, steps walked, calories burned, distance walked, sleep quality, ECG / EKG, blood pressure, mood, fatigue, body temperature, glucose level, blood alcohol, and blood oxygen. In one embodiment, the machine-learned correlations are based on PPG feature vectors including Kaiser-Tiger Power-Energy values, heart rate values, and spectral entropy values.

[0016] According to one embodiment, one or more of the radiation sources and / or sensors are located on the sides of the wearable device. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a block diagram of a personal health management device in accordance with at least one embodiment herein. [Figure 2] FIG. 1 is a block diagram of a personal healthcare device operating in a network environment in accordance with at least one embodiment herein. [Figure 3A] 1 is a perspective view of a personal health management device according to one embodiment herein; [Figure 3B] FIG. 2 is another perspective view of a personal health management device according to an embodiment herein. [Figure 4A] FIG. 1 is a perspective view of a personal health management device according to another embodiment herein. [Figure 4B] FIG. 1 is a diagram of a personal health management device according to another embodiment herein. [Figure 4C] 1 is a perspective view of a personal health management device in operation according to one embodiment herein; [Figure 5] 1A-1C are a series of illustrations of a wristband for a personal health management device according to one embodiment herein. [Figure 6A] FIG. 1 is a diagram of a planar in-line sensor (FIS) for a personal health monitoring device according to one embodiment herein. [Figure 6B] 1 is a table detailing exemplary LEDs that may be used in a preferred embodiment of an FIS according to one embodiment herein. [Figure 6C] 1 is a table detailing exemplary photodiodes and photodiode combinations that may be used in a preferred embodiment of an FIS according to one embodiment herein. [Figure 6D] FIG. 1 illustrates a functional implementation of an FIS adjacent to the skin according to one embodiment herein. [Figure 6E] FIG. 10 is another diagram illustrating a functional implementation of an FIS adjacent to the skin according to an embodiment herein. [Figure 7] 1 is a flowchart of a method for training a learning machine to interpret data from a personal healthcare device according to one embodiment herein. [Figure 8] 10 is a series of interface screens displayed on a portable device app associated with a personal health management device according to one embodiment herein. [Figure 9A] 10 is a series of interface screens displayed on a portable device app associated with a personal health management device according to one embodiment herein. [Figure 9B]10 is a series of interface screens displayed on a portable device app associated with a personal health management device according to one embodiment herein. [Figure 10] 10 is a series of interface screens displayed on a portable device app associated with a personal health management device according to one embodiment herein. [Figure 11] 10 is a series of interface screens displayed on a portable device app associated with a personal health management device according to one embodiment herein. [Figure 12A] 10 is a series of interface screens displayed on a portable device app associated with a personal health management device according to one embodiment herein. [Figure 12B] 10 is a series of interface screens displayed on a portable device app associated with a personal health management device according to one embodiment herein. [Figure 13] 10 is a series of interface screens displayed on a portable device app associated with a personal health management device according to one embodiment herein. [Figure 14] 10 is a series of interface screens displayed on a portable device app associated with a personal health management device according to one embodiment herein. [Figure 15] 10 is a series of interface screens displayed on a portable device app associated with a personal health management device according to one embodiment herein. [Figure 16A] 1A-1C are various views of a personal health management device according to another embodiment herein. [Figure 16B] 1A-1C are various views of a personal health management device according to another embodiment herein. [Figure 16C] 1A-1C are various views of a personal health management device according to another embodiment herein. [Figure 16D] 1A-1C are various views of a personal health management device according to another embodiment herein. [Figure 16E] 1A-1C are various views of a personal health management device according to another embodiment herein. [Figure 16F] 1A-1C are various views of a personal health management device according to another embodiment herein. [Figure 17A] 10A-10C are various views of a personal healthcare device having a charging dock according to another embodiment herein. [Figure 17B] 10A-10C are various views of a personal healthcare device having a charging dock according to another embodiment herein. [Figure 17C] 10A-10C are various views of a personal healthcare device having a charging dock according to another embodiment herein. [Figure 17D] 10A-10C are various views of a personal healthcare device having a charging dock according to another embodiment herein. [Figure 17E] 10A-10C are various views of a personal healthcare device having a charging dock according to another embodiment herein. [Figure 17F] 10A-10C are various views of a personal healthcare device having a charging dock according to another embodiment herein. [Figure 18] 4 is a table illustrating status signals generated according to an exemplary embodiment of a personal health management device herein. DETAILED DESCRIPTION OF THE INVENTION

[0018] Referring to FIG. 1 , a wearable personal health monitoring device 100 according to at least one embodiment includes a processor 202 coupled to computer memory 204. The device 100 may be a smartwatch, smart band, or other wearable electronic device. While the device 100 may be shown and described as a watch, it is understood that the functions may be implemented in other devices, including smartphones or tablets, or any other device capable of performing the functions disclosed herein, and thus the meaning of "device" as used herein is not so limited. The memory stores software therein that, when executed, causes the device 100 to perform the functions discussed herein. The processor 202 is preferably further coupled to a transmitter / receiver 206 that enables communication between the device 100 and other devices, as discussed below. The device 100 preferably includes at least one radiation source 208 and one or more sensors 210. The radiation source 208 is generally a device that emits energy that is converted and received by the sensor 210. The radiation source 208 and sensor 210 are controlled by the processor 202 to emit energy and process the converted energy received by the sensor 210 into usable biometric data as discussed herein. One or more sensors (and radiation source) may include a sensor package that is connected directly to the device's main board for signal acquisition and processing. Various types of radiation sources may be used with the device 100, including light (visible and invisible spectrum), heat, sound, electrical conduction, etc.

[0019] Device 100 may include multiple respective radiation sources / sensors, such as a combination of infrared and red light, and corresponding sensors. The device may further include one or more sensors 210 operable to collect hemodynamic or other data, from which device 100 uses signal processing and / or other refinements in processor 202 to reduce signal noise before transmitting the data remotely for further processing into more meaningful parameters, such as heart rate, respiration rate, body fat percentage, walking steps, ECG / EKG, blood pressure, body temperature, glucose concentration, blood alcohol, and blood oxygen. Noise may be mechanically reduced with raised edges on the edge of the sensor glass (as shown in FIGS. 4A-4B) to provide a better seal between the LED, PD, and fingertip. Additionally, the device may detect ambient light (e.g., with a separate sensor) when reading and erase or otherwise eliminate interference or noise caused by ambient light. The raw data collected by the device from these sensors 210 may be processed and / or aggregated remotely on a server to infer, for example, overall health, health changes, mood, sleep quality, fatigue, or stress.

[0020] Device 100 is thus operable to collect data to enable a wealth of personal health data including one or more of the following: heart rate, respiration rate, steps walked, calories burned, distance walked, sleep quality, ECG / EKG, arrhythmia detection, bioimpedance (BIA), accelerated plethysmography (APG), blood pressure, mood, fatigue, body temperature, glucose levels, blood alcohol, blood oxygen, etc. Device 100 may also include one or more of the following features: iPhone® / Android connectivity, or a panic button (sounds an audible and visual alarm, delivers GPS location and message to a pre-set address, etc.) as a standalone IoT (Internet of Things) device to enable remote monitoring of vital activities by, for example, medical personnel, houses germanium stones, provides mosquito protection, displays location-based air quality, detects toxic gases, etc.

[0021] In one embodiment, device 100 may automatically measure certain biometric data through an internal timer. The frequency at which measurements are taken may be preset or remotely configured by the wearer, a caregiver, or an authorized third party. For example, the frequency may be every 30 minutes, every 60 minutes, etc., selected from a drop-down menu of available frequencies. Device 100 may also continuously collect data for use, such as to infer some conclusions from the data, while also displaying and tabulating periodic measurements. For example, device 100 may continuously collect heart rate data and use it to determine heart rate variability, while also tabulating only hourly measurements. In another embodiment, device 100 may additionally or alternatively include sensors to evaluate biometric data as needed, i.e., when the user chooses to perform measurements.

[0022] In at least one embodiment, device 100 includes at least one sensor for gathering information about environmental conditions at the location of device 100, such as temperature, humidity, weather conditions (e.g., rain or snow), and air quality. Air quality can be assessed, for example, using a gas sensor configured to determine the presence and level of hazardous or unhealthy materials and / or conditions. For example, the sensor may monitor low or high levels of temperature, humidity, oxygen, ozone, carbon monoxide, VOCs, TVOCs, odors, sulfur, flammable gases, air quality, or any other air quality criterion. Device 100 may determine the level, for example, on a scale ranging from 1 to 5, where level 1 may indicate a normal condition and level 5 may indicate an unacceptable condition. The level may be based on one or more environmental readings; for example, a combination of a moderate VOC level and a moderate carbon monoxide level may be combined into a higher combined level of 4 or 5.

[0023] Referring to FIG. 2 , the personal health management device 100 is preferably operable to communicate with other devices in a network environment. For example, the device 100 may communicate directly with a portable device 102 (such as a phone or tablet) or a personal computer 104 via a short-range wireless connection such as Bluetooth®, or via the “Internet of Things” (IoT). In addition, the device 100 may be operable to indirectly communicate with these and other devices over a wireless LAN or via a GSM or LTE connection. The device may also be configured with a near-field communication system / device, allowing the device to be used for NFC payments. Finally, the device 100 may operate to provide the functionality discussed herein in conjunction with one or more server computers 106, which are further coupled to one or more databases 108 via the Internet 110.

[0024] In at least one embodiment, device 100 communicates with a portable device 102 or personal computer 104 running an application that manages the results of information received from device 100. The application may display current biometric data as well as historical biometric data (collected over time), as shown, for example, in FIGS. 6-11. Data (biometric and environmental) may be stored locally on portable device 102, on personal computer 104, or preferably remotely on the "cloud." The latter allows a user to access the data online via a browser application. In another embodiment, the biometric data may be accessed locally on device 100.

[0025] Referring to FIG. 3A, in one embodiment, device 100 is in the form of a device worn on a user's wrist. In this regard, device 100 may include a wristband having ends that are detachably connected to one another. Device 100 may further include a mode button that, when pressed for a current time, toggles through specific functions. For example, two seconds may power device 100 on, whereas eight seconds may power device 100 off. Sequential clicks may activate other functions, such as a double selection to activate a panic button function. Device 100 may also include an outward-facing top surface ("exterior") that includes a display 312, which may be a simple LED or more robust display such as an LCD that displays alphanumeric biometric data, or a touchscreen display. In one embodiment, device 100 includes a first sensor 308 on the exterior surface of device 100 and / or a second sensor 310 on the side of device 100. The device 100 may optionally include a third sensor 309 on the exterior surface of the device 100, as shown. For example, the device may include one or more biometric impedance sensors 308, 309. These sensors may also be on the bottom surface of the device 100, as shown in FIG. 3B (402, 404). Additionally, top and bottom sensors may be used, in which the user touches the top sensor 308 (and preferably 309) with one or two fingers to complete a circuit between them, and / or touches the top sensor 308 (and preferably 309) with the bottom sensor 402 touching the wrist, to collect additional data that allows biometric impedance to be measured and body composition to be inferred in near real time. The fingers on sensors 308 and 309 may further detect electrical signals generated by the user's heart to generate an ECG suitable for clinical and fitness applications, as shown in FIG. 15. The environmental sensor may be located anywhere on the device, but is preferably located on an exterior surface so that interference from contaminants can be minimized.

[0026] Referring to FIGS. 4A and 4B, in another embodiment, device 100 includes electrical contact sensors 308 and 309 on the top surface of device 100 and a sensor 310 on the side. Sensor 310 is preferably positioned within a ridge 314 that extends outward from the side of the device and preferably around the periphery of sensor 310, so that sensor 310 is recessed within ridge 314. Referring to FIG. 4C, a lateral sensor may be used in which a user's fingertip forms a good seal with sensor 310 via ridge 314, and IR and NIR light is emitted from the lateral sensor and detected to generate a PPG signal from which SpO2 is inferred in real time, as shown in FIG. 4C. BP can be inferred by repeating a similar process. These higher quality measurements inferred from sensor 310 can be remotely compared by server computer 106 with measurements inferred from the PPG signal from rear continuous measurement sensor 404. Any discrepancies can be used to correlate and inform machine learning processes to improve measurements inferred by sensor 404. Device 100 may further include a sensor / radiator pair, preferably in-line, on the side or rear of device 100, arranged to reflect signals between each other, as with other in-line sensor / radiator embodiments discussed herein.

[0027] Finally, device 100 may include a unique mechanism for attachment of stones 304 that are directed toward the wearer's skin. The stones are preferably mounted on a modular platform that allows them to be interchangeably added to the wristband of the device, as shown in FIG.

[0028] 6A, according to another embodiment, sensors 310 and 404 may be a planar in-line sensor (FIS) 600 that may be used to obtain biometric data such as heart rate, respiration rate, ECG / EKG, blood pressure, glucose levels, blood alcohol, blood oxygen, etc., via photoplethysmography (PPG) signals. The FIS may be mechanically applied adjacent to the body at the surface 608 of the skin, where the blood vessels lie beneath. By applying the FIS adjacent to the body at the surface of the skin, it is possible to obtain useful PPG signals of different frequencies that may be used to determine or otherwise derive blood glucose levels within these vessels.

[0029] In a preferred embodiment, blood glucose measurements are taken on the underside of the wrist, fingertips, or other sufficiently flat surface with good access to capillaries or veins for the FIS to register a low-noise PPG signal.

[0030] In a preferred embodiment, the sensor package FIS600 is connected directly to the device's main board for PPG signal acquisition and processing.

[0031] In a preferred embodiment, the FIS may be a surface-mounted device (SMD) package containing multiple near-field infrared (NIR) LEDs 602 and 606 and a photodiode (PD) 604, with the LEDs and PD mounted to a base 609 of the SMD package such that the PD is arranged between the LEDs in the FIS and / or SMD. In a preferred embodiment, the SMD package may include one NIR LED (606) having a wavelength of approximately 1300 nanometers (nm), one NIR LED (602) having a wavelength of approximately 1550 nm, and one photodiode (PD) (604) having a wide wavelength sensitivity from approximately 900 nm to approximately 1700 nm. It is understood that the order and / or location of the LEDs may vary. Thus, the in-line SMD package is an exemplary embodiment and is therefore not limiting. The term "about" is used herein to represent applicable tolerances in the manufacturing of such diodes.

[0032] In another embodiment, the SMD package may include one LED 606 having a wavelength range of about 1300 nm ±10%, one LED 602 ​​having a wavelength of about 1550 nm ±10%, and one PD 604 having a wide wavelength sensitivity from about 900 nm to about 1700 nm ±10%.

[0033] In operation, the user of device 100, or the user's caregiver or authorized third party, can initiate a measurement directly through the device's menu via the touchscreen display or through the menu of a connected device communicating with device 100. The user places their fingertip on sensor 310 for several seconds (typically 30-60 seconds) until a light is emitted, as shown in FIG. 4C. The flow for operating the side sensor, in certain embodiments, may include selecting the desired function on the device or a connected device, placing the fingertip on the side sensor, selecting the desired length of time, maintaining the fingertip on the sensor, viewing the results on the device's display, and synchronizing the results with a connected device (app) and / or a cloud platform supporting the device's remote monitoring capabilities via, for example, Bluetooth, IoT, etc. The user can perform as many measurements as desired without the need to attach or apply any external device to their fingertip. Device 100 can continue to perform all other functions and run its applications, providing the user with all the benefits of a fingertip device, without any limitations or compromises to measurement performance or quality. The sequence of steps for initiating a side sensor measurement, from initiating the measurement to outputting the results, may be the same as or similar to the steps performed by a conventional fingertip pulse oximeter, but is not necessarily so constrained. The environmental measurement may be performed automatically, for example, periodically, or in response to input from a user. For example, the user may press an input, such as a button, that causes the device to perform an environmental measurement.

[0034] The device's various sensors can work together to provide more robust results. For example, to perform an SpO2 measurement, the NIR and IR LEDs can be activated simultaneously to generate overlapping PPGs that allow for SpO2 measurements using data from both the wrist and fingertip. At night, the LEDs on the back of the device can be contacted with the user's wrist and activated while the user sleeps for SpO2 data. The accuracy of this data can be improved by correlating the wrist sensor reading with SpO2 data from a previous reading using the side sensor.

[0035] Device 100 may also automatically measure certain biometric and / or environmental data through an internal timer. The frequency at which measurements are taken may be preset or remotely configured by the wearer, a caregiver, or an authorized third party. For example, the frequency may be every 30 minutes, every 60 minutes, etc., selected from a drop-down menu of available frequencies. Device 100 may also continuously collect data for use, such as to infer some conclusions from the data, while also displaying and tabulating periodic measurements. For example, device 100 may continuously collect heart rate data and use it to determine heart rate variability, while also tabulating only hourly measurements. Device 100 may further include sensors to evaluate biometric data as needed, i.e., when the user chooses to perform measurements.

[0036] In a preferred embodiment, during operation, light from two NIR LEDs 602, 606 is directed toward the skin surface 608 via reflection from two angled mirrors 603, 605, respectively, mounted to the base 609 of the SMD package. The angles of the angled mirrors 603, 605 ensure that the light reflected back is at a predetermined angle when subsequently captured by the PD 604, as shown by example in FIG. 6A. Specifically, light from LED 602 ​​reflects off the skin surface 608 at an angle of approximately 45 degrees, and light from LED 606 reflects off the skin surface at an angle of approximately 90 degrees, with each ray of light being captured by the PD 604. In another embodiment, the angular range of reflection can be ±10%.

[0037] In a preferred embodiment, sensors 310 and 404 are covered with specially designed glass, e.g., with a thickness and curvature (or lack thereof, i.e., flat) that precisely directs LED light toward the fingertip, and the PD receives a signal from the reflected change in light absorption in oxygenated or deoxygenated blood. As can be seen, sensors 308 and 404 can be arranged alongside or in-line with one another.

[0038] 6A, in some cases, rather than the emitter / LED being directed outward, mirrors and / or prisms may be used to reflect the signal / light emitted from the emitter / LED, which is then reflected at the subject's skin to the receiver / sensor / PD. This arrangement may allow for a more compact design and may allow the emitter / LED to be split and used for multiple purposes and receivers / sensors / PDs.

[0039] In a particular configuration, using PPG technology and two specific NIR LEDs (approximately 1300 nm and approximately 1550 nm) of very high frequency wavelength, the high frequency wavelength light is reflected at a specific angle when it is reflected off blood glucose molecules, for example, 90 degrees for 1550 nm wavelength light and 45 degrees for 1300 nm wavelength light.

[0040] FIG. 6B outlines details of typical LEDs that may be used in preferred embodiments of the FIS. For example, typical LEDs have wavelengths of approximately 640 nm, approximately 940 nm, approximately 1300 nm, and approximately 1550 nm. For heart rate detection and monitoring, a combination of LEDs with wavelengths of approximately 640 nm, approximately 940 nm, approximately 1300 nm, and approximately 1550 nm may be used. For blood oxygen detection and monitoring, a combination of LEDs with wavelengths of approximately 640 nm and approximately 940 nm may be used. For blood glucose detection and monitoring, a combination of LEDs with wavelengths of 640 nm, 940 nm, 1300 nm, and 1550 nm may be used.

[0041] FIG. 6C outlines details of typical photodiodes and photodiode combinations that may be used in accordance with preferred embodiments of the FIS.

[0042] 6D displays a functional implementation of the FIS next to the skin for LED 1300 and LED 1550 when the light is reflected at a depth of about 4 mm below the skin. The angle between 602 and 604 is 45 degrees, and the wider angle between 606 and 604 is 90 degrees.

[0043] FIG. 6E shows a functional implementation of the FIS next to the skin for LED 940 and LED 640 when the light is reflected at a depth of about 4 mm below the skin.

[0044] 3B, there is shown the bottom or underside of device 100. This side may include one or more sensors thereon, at or around sensor 402. The bottom of device 100 preferably includes multiple sensors including at least synchronized red and near-infrared radiation sources / sensors.

[0045] Referring to FIG. 5, details of the wristband are shown. Specifically, the band includes a plurality of equally spaced holes to accommodate the mounts with stones therein. The cross-section of the holes is preferably hourglass-shaped to hold similarly shaped legs of the mount. The mount is inserted through the holes from the inside of the band. One set of legs of the mount fits flush with the outside of the band, as shown. The top of the mount includes at least one stone. The stones are typically materials believed to have beneficial properties, such as gold, silver, copper, germanium, magnets, salt, etc. The mount advantageously places the stone in contact with the wearer's skin.

[0046] 7 presents a flowchart of a method for training a learning machine to interpret data from a personal health monitoring device according to one embodiment herein. In at least one embodiment, device 100 uses synchronized red and near-infrared light to generate a photoplethysmography (PPG) signal that is analyzed to determine a biometric measurement, such as a blood glucose level (BGL) or any of the other biometrics discussed herein. The PPG signal is received from device 100 by an analysis server, step 702. Steps in evaluating a biometric measurement using PPG may involve correlations between PPG and BGL, blood pressure, etc. Similarly, correlations between PPG and essential nutrients and / or endothelial health, as well as environmental data, may also be analyzed and correlated.

[0047] This analysis can be accomplished by, for example, sampling 1000 people for PPG signal data, standard BGL, blood pressure, environment, etc. to create training data for machine learning. Testing can be performed, for example, every day before breakfast for 14 consecutive days.

[0048] Algorithms for determining BGL and other biometrics from PPG data can be derived from the following exemplary process.

[0049] Obtaining Biometric Data—Biometric data is received, step 704—e.g., using a personal health monitoring device with a planar inline sensor at 660 nm red light and 940 nm near-infrared light to obtain PPG data. The device takes two readings, each lasting one minute. Next, the BGL can be tested using a medical-grade, minimally invasive blood glucose monitor. Blood pressure is also measured with a cuff sphygmomanometer, ensuring two additional readings are taken. For each individual, testing continues for two weeks, twice daily at the same time each day: one in the morning before breakfast and one in the afternoon one hour after lunch. The individual's gender, age, height, weight, nationality, ethnicity, history of cardiovascular and cerebrovascular disease, history of metabolic disease, history of family disease, ongoing and any ongoing medications, or medical condition history may be provided along with the biometric data. Location, caffeine intake, whether and how much one smokes, mood, fatigue, environmental conditions, etc. may also be recorded.

[0050] The generated PPG data can be processed to obtain a clear signal. Feature vector data is extracted (step 706). The PPG signal can be filtered with a bandpass filter that allows signals between about 0.5 Hz and about 5 Hz, and then an adaptive noise canceller can be applied using recursive least squares or a similar method. The key to usable data is to find a valid reference signal and extract a feature vector. From the clear signal, a feature vector can be identified and extracted, and then supervised machine learning can be applied to calculate correlations. The resulting equation can be evaluated against a subset of test data to predict the effectiveness of the algorithm.

[0051] A typical PPG feature vector: Kaiser Tiga Power Energy Value:KTE n =x(n) 2 -x(n+1)x(n-1), where x is the electromyographic value, n is the number of samples, and the split real-time power-energy value: KTE n , average KTE n μ , mean square deviation KTEn σ , quarter distance KTE n α , slewness KTE n β , and the corresponding segment KTE μ , K.T.E. σ , K.T.E. α , K.T.E. β can be obtained.

[0052] Heart rate value: From the PPG wave, the corresponding HR μ , H.R. σ , H.R. α , H.R. β can be calculated.

[0053] Spectral entropy is calculated by the FFT (Fast Fourier Transform) of the split signal after regularization. n ← FFT(x(n),L) n Knowing this, the entropy can then be calculated, H←Px n Log(Px n ). The split data is H n μ , H n σ , H n α , H n β If an arithmetic overflow occurs, the logarithmic function LogE←Log(x(n)) is used, and LogE σ and LogE α Learn about

[0054] The red and near-infrared peak values ​​Pr and Pi can also be calculated independently from the power values. To avoid respiratory effects, the segmentation duration can be 5 to 10 seconds, where the signal is x(n) and the corresponding matrix is ​​Xi. When performing this calculation, new significant vector elements can be added, and signals with subtle effects can be removed.

[0055] The correlation between the PPG signal and the biometric data can be determined using supervised machine learning, step 708. The dimension of the vector can be 10 to 20 from the PPG data. Randomly, 90% of the data can be placed in a training set and another 10% can be placed in a test set. Machine learning algorithms can be used to compare least squares linear recursion, logistic regression, support vector machines (SVMs), classification and regression trees (CARTs), random forests, neural networks (NNs), AdaBoost, etc. SVMs can use sequential minimal optimization (SMO) and a kernel function, which may use a radial basis function (RBF). NNs can use backpropagation (BP) and Hopfield for testing. Based on training and testing via machine learning, certain aspects of the biometric data can be correlated with certain PPG waves.

[0056] The device's high-quality photoplethysmography (PPG) obtained using side sensor 310 illuminates the fingertip and measures changes in light absorption due to changes in blood volume in the tissue's microvascular bed. The second derivative of this PPG can be used to determine the degree of vascular aging and atherosclerosis, which can be presented on a human interface screen on a scale of 1 to 7, as shown in FIG.

[0057] 8 presents an interface screen for displaying a user's step count data determined from a personal health management device according to one embodiment herein. The interface may include step count data by day, week, or month. The personal health management device may record total steps, distance, and calorie burn totals. Step totals may also be provided in a table over a given period of time.

[0058] 9 presents an interface screen for displaying a user's heart rate data determined from a personal health management device according to one embodiment herein. The interface may include a record of maximum and minimum beats per minute, along with an indicative range for normal / abnormal heart rate. The heart rate interface may also calculate the user's average heart rate and display a table of the user's heart rate over a given period of time.

[0059] 10 presents an interface screen for displaying a user's respiration rate data determined from a personal health management device according to one embodiment herein. The interface may present the determined respiration rate (times per minute) for the user as determined by the personal health management device. The user's determined respiration rate may be displayed in a table over a given period of time.

[0060] FIG. 11 presents an interface screen for displaying a user's energy and mood data determined from a personal health monitoring device according to one embodiment herein. The user's energy and mood may be calculated according to a predetermined scale. For example, an energy level of "25" may be normal, and a mood of "50" may indicate that the user is calm. Energy and mood may be displayed in a table over a given period of time.

[0061] 12 presents an interface screen for displaying a user's sleep data determined from a personal health management device according to one embodiment herein. The sleep data may include total sleep duration, duration of deep sleep, duration of light sleep, and how many times the user woke up during sleep. One or more data points about the sleep may be plotted over a given period of time.

[0062] FIG. 13 presents an exemplary home screen of a personal health management device according to one embodiment herein. The home screen may include a quick summary of the data described with respect to FIGS. 8-12. Each data section may be presented in a selection tile that can be expanded to view the data in more detail. Additionally, the home screen may provide features for taking "quick checks," for example, of respiration rate, heart rate, energy, and mood.

[0063] 14 depicts an interface screen for displaying a user's APG result data determined from a personal health monitoring device according to one embodiment herein. As can be seen, the results may be presented relative to a scale for a given date and time in the interface. In this example, the scale is presented as L1-L7, and the APG results determined by the system are displayed on the scale accordingly. The interface screen may also include a vascular health analysis, which may include a table showing readings over time and trends between low or high peaks in the table.

[0064] FIG. 14 depicts an interface image for displaying a user's ECG result data determined from a personal healthcare device.

[0065] FIGS. 16A-16F depict another embodiment of a personal health monitoring device. In this embodiment, the device 1600 includes an integrated strap 1610 (as shown in FIG. 16C) that incorporates a space or volume for housing the electronics that provide the functionality discussed herein. Preferably, the strap 1610 includes a pair of electrical contact sensors or sensor plates 1602 on each of one of the sides of the device 1600. These plates may be made of a precious metal, such as gold or silver, for improved sensitivity. The strap 1610 may have a rounded or similar transition between the top or outward-facing surface and the side surfaces (on either side of the band). Additionally, the plates 1604 preferably have an outer surface that extends in a "waterfall" fashion from the top surface onto the side surfaces, as shown, mimicking the transition between these surfaces elsewhere on the device. This configuration provides multiple surfaces for a user to easily access sensor plate 1604 with the user's thumb and index finger for the various readings discussed herein. Finally, the band includes a clasp or buckle 1612 for adjustably fitting the band to the user's wrist.

[0066] In a preferred embodiment, device 1600 includes a visual and / or audio output device 1606, such as an LED light. The LED light is preferably configured in a circular pattern with button 1602 at the center of the circle, as shown. This provides a target to ensure the user successfully hits the button as desired. The LED light is preferably capable of displaying multiple colors in one or more patterns, with each color and / or pattern representing the status of device 1600, as shown in FIG. 18. Button 1602 can be a physical button or a touch sensor, i.e., a sensor that senses a user's touch. Button 1602 generally provides a means for user input related to any of the functions discussed herein.

[0067] In one embodiment, device 1600 includes a BP and / or SPO2 sensor 1608 (as discussed above with respect to the side sensor shown in FIG. 4B) on the exterior surface of band 1610, with clear glass, preferably black glass, and a raised edge on the edge of the sensor glass.

[0068] 16B depicts a side view of device 1600. As can be seen, sensor plate 1604 extends only partially across the lateral surface of device 1600. In one embodiment, the electronics are secured to case 1702, which is at least partially inserted into the space or volume. Preferably, case 1702 forms the back surface of device 1600 that contacts the user's wrist. This surface may therefore include one or more ECG, BIA, and skin temperature sensors 1704, and multiple metal ECG plates 1706, as shown.

[0069] Referring to FIG. 16C, an exploded view of the band 1610 and case 1702 is shown, which includes a volume for receiving electronics, in this case the case 1702, as discussed above. The strap 1610, or more specifically the volume, has a hole 1804 therein for the BP and / SPO2 sensor 1608. The case 1702 preferably includes multiple slots for receiving the sensors 1704, 1706, respectively. Preferably, the case also includes a slot for receiving the charging electrode 1802. Referring to FIG. 16D, the strap 1610 may include an NFC payment chip 1806 embedded therein on the opposite side of the button from the BP and / SPO2 sensor 1608.

[0070] Figure 16E depicts a rear view of the device, showing the case attached to the strap 1610 and the location of the various sensors discussed above located within the case. Figure 16F is another front view of the device, showing the outer surface 1614 of the strap 1610 and multiple slots 1616 for engaging the connection system (buckles). The strap is preferably made from liquid silica gel, and the slots 1616 are formed from a TPU material.

[0071] 17A-17F depict a dock 1900 for use with device 1600. The dock generally provides charging functionality. In this regard, as shown, with respect to charging posts in particular, the dock includes slots shaped to receive protruding features on the exterior of the case so that the charging posts align with corresponding posts on the dock.

[0072] Figure 18 depicts typical LED light statuses. For example, the LEDs may provide patterns to indicate low charge, charging, taking biometrics, malfunction, power on, power off, etc.

[0073] In one embodiment, a proof-of-sensing protocol is employed that uses cryptographic techniques and blockchain technology to secure, verify, and / or anonymize health and wellness data generated by the wearable devices disclosed herein. This protocol creates an unbreakable and tamper-proof chain of verification from the wearable device to a service provider computer in the "cloud." Each wearable device that supports the proof-of-sensing protocol preferably includes a VSC PoS security chip, a high-end security solution that provides a trust anchor for connecting IoT devices to the cloud and gives every IoT device its own unique identity. The VSC PoS is preferably a high-end CC EAL6+ (high) certified security controller.

[0074] The operational mode of a PoS-enabled wearable device may involve the wearable device collecting the user's health data as discussed herein. These measurements may be performed automatically or whenever the user initiates a measurement. The measurement data is preferably grouped and packaged for transmission to a smartphone app via Bluetooth Low Energy. Before transmission to the app, the data packets are passed through a cryptographic hash function to generate a message digest. This message digest, along with the device's unique private key and several other identifying variables, is used to generate a digital signature. The signature may then be added to the data packet by the wearable device, and the updated data packet may then be sent to the app / user's mobile device. The entire data packet may then be sent from the app to a service provider server over an encrypted, secure connection. At the server, the original data packet is extracted and a digital signature is calculated using the uploaded data and the device's public key, which is securely stored and uniquely associated with each individual device. Using a consensus method, the device and server signatures may be compared, and if they match, the data packet is added to the Smart Chain.

[0075] Although the present invention has been described in some detail for purposes of clarity and understanding, it will be recognized by those skilled in the art upon reading and understanding this disclosure that various changes in form and detail may be made therein without departing from the true scope of the invention.

Claims

1. 1. A wearable device for measuring the health of an individual, comprising: a one-piece strap having an outwardly facing top surface, first and second lateral surfaces, and a bottom surface, the bottom surface facing the user's skin; a plurality of electrical contact sensors, a first electrical contact sensor of the plurality of electrical contact sensors disposed on the top surface and the first lateral surface, and a second electrical contact sensor of the plurality of electrical contact sensors disposed on the top surface and the second lateral surface of the wearable device, the first electrical contact sensor and the second electrical contact sensor configured to complete a circuit therebetween when a user contacts the first sensor with a first surface of the user's skin and the second sensor with the first surface or the second surface of the user's skin, and the wearable device detects PPG waves at the wearable device; a network communication module configured to transmit the detected PPG waves to a server, wherein the server processes the PPG waves and infers biometric data therefrom based on machine-learned correlations generated from a training set of PPG waves and biometric data; a processor configured to generate and display biometric data on an interface screen; A wearable device comprising:

2. The wearable device of claim 1 , further comprising an output device configured to display at least one pattern associated with a status of the wearable device.

3. The wearable device of claim 2 , wherein the output device is a circular LED, and the device further comprises an input button within the circular LED.

4. 10. The wearable device of claim 1, further comprising a third sensor disposed on the outward-facing upper surface, the third sensor recessed within a ridge extending outward from the outward-facing upper surface of the device around a periphery of the third sensor, the ridge configured to form a seal with the first or second surface of the user's skin when the user contacts the third sensor with the first or second surface of the user's skin.

5. 5. The wearable device of claim 4, wherein the third sensor comprises a first NIR LED and a second NIR LED and a photodiode, the third sensor being covered with glass configured to direct light from either the first NIR LED or the second NIR LED toward the user's skin and receive reflected light from the first NIR LED and the second NIR LED with the photodiode.

6. 10. The wearable device of claim 1, wherein the server processes the PPG waves and infers biometric statistics therefrom, the biometric statistics including at least one of overall health, health changes, mood, sleep quality, fatigue, and stress.

7. The wearable device of claim 1 , wherein the machine-learned correlation is based on a PPG feature vector including a Kaiser-Tiger power-energy value, a heart rate value, and a spectral entropy value.

8. The wearable device of claim 1 , wherein the PPG waves are generated by infrared, green, or red light emitted from the personal health monitoring device.

9. 10. The wearable device of claim 1, wherein the personal health management device further comprises an in-line sensor (IS) comprising a first near-field infrared (NIR) light emitting diode (LED), a second NIR LED, and a photodiode having a wavelength sensitivity range of about 900 nm to about 1700 nm ±10%, the photodiode being disposed on the IS between the first NIR LED and the second NIR LED and configured in association therewith to receive reflected light from the first NIR LED and the second NIR LED; and a first angle mirror and a second angle mirror configured to reflect light from either the first NIR LED or the second NIR LED onto a user's skin and to have the user's skin reflect light back to the photodiode, wherein the personal health management device generates the detected PPG wave based on the light reflected by the user's skin.

10. 10. The method of claim 9, wherein the first NIR LED has a first wavelength in the near infrared spectrum and the second NIR LED has a second wavelength in the near infrared spectrum.

11. 10. The method of claim 9, wherein a first intermediate detected PPG wave is generated from light from the first NIR LED reflected off the user's skin, a second intermediate detected PPG wave is generated from light from the second NIR LED reflected off the user's skin, and the detected PPG wave is generated from a combination of the first intermediate detected PPG wave and the second intermediate detected PPG wave.

12. 10. The method of claim 9, wherein the first NIR LED has a wavelength of about 1550 nm ±10% and the second NIR LED has a wavelength of about 1300 nm ±10%.

13. 10. The method of claim 9, wherein the light from the first NIR LED is directed toward the user's skin through the first angle mirror and the light from the second NIR LED is directed toward the user's skin through the second angle mirror such that the light from the first NIR LED is reflected off blood glucose molecules back to the photodiode at a first predetermined angle and the light from the second NIR LED is reflected off blood glucose molecules back to the photodiode at a second predetermined angle.

14. 14. The method of claim 13, wherein the first predetermined angle is approximately 45 degrees and the second predetermined angle is approximately 90 degrees.

15. 10. The method of claim 9, wherein the inline sensor is fitted within a ridge that extends outward from the outward-facing top surface of the device around a periphery of the sensor, the ridge configured to form a seal with the first or second surface of the user's skin when the user contacts at least one of the plurality of sensors with the first or second surface of the user's skin.

16. 10. The wearable device of claim 9, wherein the in-line sensor is covered with glass configured to direct light from either the first NIR LED or the second NIR LED toward the user's skin and receive reflected light from the first NIR LED or the second NIR LED with the photodiode.

Images