# US 12,727,794 B2 · Wearable device and a method for selecting and interpreting light intensity data values applicable thereto > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,727,794 B2 - Document ID: US12727794B2 - Status: Granted - Kind: B2 - Application number: 18/400,268 - Filed: 2023-12-29 - Published: 2024-07-04 - Granted: 2026-09-08 - Earliest claimed: 2022-12-30 - Assignees: National Taipei University of Technology; Taipei Medical University - Applicants: National Taipei University of Technology; Taipei Medical University - Inventors: Cheng-Chun Chang; Po-Wen Lu - Web page: https://patentgrep.com/patent/US12727794B2 ## CPC - A61B 5/00: Measuring for diagnostic purposes (radiation diagnosis A61B6/00; diagnosis by ultrasonic, sonic or infrasonic waves A61B8/00); Identification of persons - A61B 5/1455: using optical sensors, e.g. spectral photometrical oximeters - A61B 5/02416: {using photoplethysmograph signals, e.g. generated by infrared radiation (A61B5/14552 takes precedence)} - A61B 5/14532: {for measuring glucose, e.g. by tissue impedance measurement} - A61B 5/681: {Wristwatch-type devices} - A61B 5/7203: {for noise prevention, reduction or removal} - A61B 5/7267: {involving training the classification device} - A61B 2560/0443: Modular apparatus - A61B 2562/0238: Optical sensor arrangements for performing transmission measurements on body tissue - A61B 2562/0242: for varying or adjusting the optical path length in the tissue - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) ## Abstract A wearable device and a method for selecting and interpreting light intensity data values applicable thereto. The wearable device includes a light emitting unit and a spectrum sensing unit. The method includes steps of: controlling the light-emitting unit to simultaneously emit a mixed light including multiple spectrum frequency bands to enter inside skin of the user; controlling the spectrum sensing unit to sense the intensity of an outgoing light from inside skin of the user at a series of sampling time to generate a spectrum data set including a plurality of groups of frequency band-light intensity data values; and selecting at least one of a first group of frequency band-light intensity data values satisfying a signal quality index in the spectrum data set to perform a data interpretation at a first judgment time point. ## Claims ### Claim 1 A wearable device, applicable to a user, comprising: a main body, comprising a casing, which comprises a sensing surface configured to be close to skin of the user; a physiological data sensing module, disposed in the casing, comprising a light emitting unit and a spectrum sensing unit, wherein the light emitting unit is configured to emit a mixed light comprising multiple spectrum frequency bands, and the multiple spectrum frequency bands of the mixed light are configured to simultaneously enter the skin of the user through the sensing surface into different depths inside the skin; the spectrum sensing unit is configured to automatically sense an outgoing light generated from the different depths inside the skin of the user in response to the mixed light at each of a series of sampling time points, thereby obtaining multiple groups of frequency band-light intensity data values of the outgoing light respectively corresponding to the multiple spectrum frequency bands of the mixed light; and an arithmetic unit, in signal communication with the physiological data sensing module, configured to automatically compare a specified parameter of each of the multiple groups of frequency band-light intensity data values received from the physiological data sensing module with a specified index at each of a series of judgement time points, thereby determining respective signal qualities of the multiple groups of frequency band-light intensity data values, wherein at least one of the multiple groups of frequency band-light intensity data values is automatically retrieved for data interpretation by the arithmetic unit based on the respective signal qualities of the multiple groups of frequency band-light intensity data values at each of the series of judgement time points. ### Claim 2 (depends on 1) The wearable device according to claim 1, wherein the sensing surface is configured to be against the skin of a wrist of the user, the physiological data sensing module is a micro spectrometer comprising the light emitting unit and the spectrum sensing unit, the light emitting unit in the micro spectrometer comprises multiple light emitting diodes simultaneously emitting respective light with the multiple spectrum frequency bands to form the mixed light, the spectrum sensing unit in the micro spectrometer senses the outgoing light at each of the series of sampling time points to generate multiple spectrum frequency bands-intensity signal curves corresponding to the multiple groups of the frequency band-light intensity data values, respectively. ### Claim 3 (depends on 2) The wearable device according to claim 2, wherein the arithmetic unit determines the respective signal qualities of the multiple groups of frequency band-light intensity data values by comparing respective signal-to-noise ratio (SNR) values of the multiple spectrum frequency band-intensity signal curves with an SNR threshold. ### Claim 4 (depends on 3) The wearable device according to claim 3, wherein the arithmetic unit is configured to perform a photoplethysmography (PPG) data interpretation according to all of the groups of frequency band-light intensity data values automatically retrieved by the arithmetic unit at the series of judgement time points. ### Claim 5 (depends on 4) The wearable device according to claim 4, wherein the arithmetic unit is configured to automatically generate a white blood cell count, red blood cell count, hemoglobin, hematocrit, blood glucose, blood pressure, blood urea nitrogen, creatinine and/or alanine aminotransferase (ALT) of the user as a result of the PPG data interpretation. ### Claim 6 (depends on 4) The wearable device according to claim 4, wherein the arithmetic unit is configured to optimize the PPG data interpretation by using artificial intelligence learning. ### Claim 7 (depends on 4) The wearable device according to claim 4, wherein the SNR values are configured to include a signal part defined in a range from a lower limit frequency of 0.5 Hz to an upper limit frequency limit of 7 Hz, and a noise part beyond the range, and the arithmetic unit performs a maximum ratio combining of all of the retrieved groups of frequency band-light intensity data values at the series of judgement time points for weighting and combining, thereby obtaining an optimized intensity signal curve for the PPG data interpretation subsequently. ### Claim 8 (depends on 7) The wearable device according to claim 7, wherein the arithmetic unit is configured to automatically generate a white blood cell count, red blood cell count, hemoglobin, hematocrit, blood glucose, blood pressure, blood urea nitrogen, creatinine and/or alanine aminotransferase (ALT) of the user as a result of the PPG data interpretation. ### Claim 9 (depends on 7) The wearable device according to claim 7, wherein the arithmetic unit is configured to optimize the PPG data interpretation by using artificial intelligence learning. ### Claim 10 (depends on 1) The wearable device according to claim 1, wherein the arithmetic unit is configured to convert the retrieved groups of frequency band-light intensity data values at the series of the judgement time points into a color image, which is presented with two-dimensional coordinates sampling time and spectrum frequency band, wherein a color value of a point in the color image corresponds to a light intensity value, and the arithmetic unit is configured to perform the data interpretation on the color image by using an image recognition technology of convolutional neural network. ### Claim 11 A wearable device, applicable to a user, comprising: a main body, comprising a casing, which comprises a sensing surface configured to be close to skin of the user; a physiological data sensing module, disposed in the casing, comprising a light emitting unit and a spectrum sensing unit, wherein the light emitting unit is configured to emit an incident light whose intensity is changing with time, and different intensities of the incident light are configured to sequentially enter the skin of the user and reach different depths under the sensing surface; the spectrum sensing unit is configured to automatically sense an outgoing light generated from one of the different depths inside the skin of the user at each of a series of sampling time points in response to a one of the different intensities of the incident light, thereby obtaining multiple groups of depth-light intensity data values of the outgoing signal corresponding to the one of the different intensities of the incident light; and an arithmetic unit, in signal communication with the physiological data sensing module, configured to automatically compare a specified parameter of each of the multiple groups of depth-light intensity data values received from the physiological data sensing module with a specified index at each of a series of judgement time points, thereby determining respective signal qualities of the multiple groups of depth-light intensity data values, wherein at least one of the multiple groups of depth-light intensity data values is automatically retrieved for data interpretation by the arithmetic unit based on the respective signal qualities of the multiple groups of depth-light intensity data values at each of the series of judgement time points. ### Claim 12 A method for physiological data interpretation of a wearable device, wherein the wearable device comprises a sensing surface configured to be close to skin of a user, and a light emitting unit and a spectrum sensing unit configured to obtain physiological data of the user through the skin, the method comprising: the light emitting unit emitting a mixed light comprising multiple spectrum frequency bands when the wearable device is activated for the physiological data interpretation, wherein the multiple spectrum frequency bands of the mixed light are configured to simultaneously enter the skin of the user through the sensing surface into different depths inside the skin; the spectrum sensing unit automatically sensing an outgoing light generated from the different depths inside the skin of the user at each of a series of sampling time points, thereby obtaining multiple groups of frequency band-light intensity data values of the output signal respectively corresponding to the multiple spectrum frequency bands of the mixed light; automatically comparing a specified parameter of each of the multiple groups of frequency band-light intensity data values with a specified index at each of a series of judgement time points, thereby determining respective signal qualities of the multiple groups of frequency band-light intensity data values, wherein at least one of the multiple groups of frequency band-light intensity data values is automatically retrieved based on the respective signal qualities of the multiple groups of frequency band-light intensity data values at each of the series of judgement time points; and performing the physiological data interpretation according to the retrieved groups of frequency band-light intensity data values at the series of judgement time points. ### Claim 13 A method for physiological data interpretation of a wearable device, wherein the wearable device comprises a sensing surface configured to be close to skin of a user, and a light emitting unit and a spectrum sensing unit configured to obtain physiological data of the user through the skin, the method comprising: the light emitting unit emitting an incident light whose intensity is changing with time, wherein different intensities of the incident light are configured to sequentially enter the skin of the user and reach different depths under the sensing surface; the spectrum sensing unit automatically sensing an outgoing light generated from one of the different depths inside the skin of the user at each of a series of sampling time points in response to one of the different intensities of the incident light, thereby obtaining multiple groups of depth-light intensity data values of the outgoing signal corresponding to the one of the different intensities of the incident light; and automatically comparing a specified parameter of each of the multiple groups of depth-light intensity data values with a specified index at each of a series of judgement time points, thereby determining respective signal qualities of the multiple groups of depth-light intensity data values, wherein at least one of the multiple groups of depth-light intensity data values is automatically retrieved based on the respective signal qualities of the multiple groups of depth-light intensity data values at each of the series of judgement time points; and performing the physiological data interpretation according to the retrieved groups of depth-light intensity data values at the series of judgement time points.