# US 2021/0330268 A1 · Method and device for calibrating physiological readings, electronic device using method, and non-transitory storage medium

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0330268 A1
- Document ID: US20210330268A1
- Status: Published application
- Kind: A1
- Application number: 17/029,343
- Filed: 2020-09-23
- Published: 2021-10-28
- Earliest claimed: 2020-04-23
- Assignees: Jiangyu Kangjian Innovation Medical Technology(Chengdu) Co., Ltd
- Applicants: Jiangyu Kangjian Innovation Medical Technology(Chengdu) Co., Ltd
- Inventors: CHENG CHEN; PING-HAO LIU; NENG-DE XIANG; MING-SHUN HU
- Web page: https://patentgrep.com/patent/US20210330268A1

## CPC

- A61B 5/7267: {involving training the classification device}
- A61B 5/021: Measuring pressure in heart or blood vessels
- A61B 5/14532: {for measuring glucose, e.g. by tissue impedance measurement}
- A61B 5/1455: using optical sensors, e.g. spectral photometrical oximeters
- A61B 5/681: {Wristwatch-type devices}
- A61B 5/0205: Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
- A61B 5/14542: {for measuring blood gases (A61B5/14551 takes precedence)}
- A61B 5/7221: {Determining signal validity, reliability or quality (preventing, reducing or removing noise induced by motion artefacts A61B5/7207; noise originating from a therapeutic or surgical apparatus A61B5/7217)}
- A61B 5/7235: {Details of waveform analysis (detecting specific parameters of the electrocardiograph cycle A61B5/349)}
- A61B 2560/0223: of calibration, e.g. protocols for calibrating sensors
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)

## Abstract

A method for calibrating measurements of human physiological state applied on an electronic device couples the device to at least one uncalibrated measurement device, worn or placed. Readings of a physiological state are collected by the uncalibrated measurement device and the readings are input into a model for calibration. A sub-module corresponding to the type of reading, for blood pressure or heart rate for example, is called up from the calibration model and a regression algorithm is applied to the calibration model to generate a calibrated reading, each sub-module in the model corresponding to one type of reading. The method also outputs calibrated physiological measurements by the called-up sub-module. A related electronic device and a non-transitory storage medium are also disclosed.

## Claims

### Claim 1

A method for calibrating measurements of physiology applied on an electronic device, the electronic device being communicationally coupled to at least one uncalibrated measurement device, the method comprising:
obtaining physiological reading collected by an uncalibrated measurement device;
inputting the reading into a model for calibration to generate a calibrated physiological measurement whereby a sub-module corresponding to the reading is called from the model for calibration and the sub-module is for calibrating the reading via a regression algorithm, where each sub-module in the model corresponds to one physiological reading;
outputting calibrated physiological measurement by the called sub-module.

### Claim 2 (depends on 1)

The method according to claim 1, wherein the method further comprises:
comparing the calibrated physiological measurement with a preset threshold range;
generating a prompt if the calibrated physiological measurement is outside of the preset threshold range.

### Claim 3 (depends on 1)

The method according to claim 1, wherein the method further comprises:
selecting at least two regression algorithms selected from preset sorting algorithms, to train and to create at least two modules for a same kind of reading of the physiological state;
averaging the at least two created modules to generate the sub-module in the model for calibration.

### Claim 4 (depends on 3)

The method according to claim 3, wherein selecting at least two regression algorithms selected from preset sorting algorithms, to train and to create at least two modules for a same kind of reading of the physiological state comprises:
obtaining a plurality of sets of sample data, the sample data comprises physiological measurements of a user and standard physiological information, where the physiological measurements are taken by the at least one uncalibrated measurement device, and the standard physiological information is contained within or taken by the at least one standard measurement device;
dividing the sample data into a training set and a validation set;
establishing an analysis model based on a regression algorithm;
training the analysis model via the training set, where an input of the analysis model is the physiological measurements in the training set and an output of the analysis model is the standard physiological information in the training set;
validating the trained analysis model via the validation set to generate a predicted accuracy of the trained analysis model;
determining whether the predicted accuracy of the trained analysis model is less than a preset threshold;
determining the trained analysis model as being the module if the predicted accuracy of the trained analysis model is not less than the preset threshold.

### Claim 5 (depends on 4)

The method according to claim 4, wherein the method further comprises:
adjusting the regression algorithm of the analysis model, or the number of the samples in the training set, or the regression algorithm of the analysis model and the number of the samples in the training set, to resume training the analysis model via the training set if the predicted accuracy of the trained analysis model is less than the preset threshold;
validating the resumed trained analysis model via the validation set to resume generating a predicted accuracy of the resumed trained analysis model;
determining whether the resumed predicted accuracy of the resumed trained analysis model is less than the preset threshold;
determining the resumed trained analysis model as being the module if the resumed predicted accuracy of the resumed trained analysis model is not less than the preset threshold;
continuing to adjust the regression algorithm of the analysis model, the number of the samples in the training set, or the regression algorithm of the analysis model and the number of the samples in the training set, to resume training the analysis model via the training set, and to validate the resumed trained analysis model via the validation set to resume generating a predicted accuracy of the resumed trained analysis model until the predicted accuracy is not less than the preset threshold if the resumed predicted accuracy of the resumed trained analysis model is less than the preset threshold.

### Claim 6 (depends on 4)

The method according to claim 4, wherein obtaining a plurality of sets of sample data comprises:
obtaining a plurality of sets of physiological measurements taken by the at least two uncalibrated measurement devices and a plurality of sets of standard physiological information measured by the at least two standard measurement devices;
combining the sets of physiological measurements and the sets of standard physiological information according to codes of the uncalibrated measurement devices and codes of the standard measurement devices to obtain a combination of each physiological measurement and all the standard physiological information, or a combination of each standard physiological information and all the physiological measurement.

### Claim 7 (depends on 1)

The method according to claim 1, wherein the to-be-calibrated measurement devices comprises one or more of a group of a smart bracelet, smartwatch, a portable oximeter, and a portable scale.

### Claim 8

An electronic device communicationally coupled to at least one to-be-calibrated measurement device, the electronic device comprising:
a storage device;
at least one processor; and
the storage device storing one or more programs, which when executed by the at least one processor, cause the at least one processor to:
obtain physiological reading collected by an uncalibrated measurement device;
input the reading into a model for calibration to generate a calibrated physiological measurement whereby a sub-module corresponding to the reading is called from the model for calibration and the sub-module is for calibrating the reading via a regression algorithm, where each sub-module in the model corresponds to one physiological reading;
output calibrated physiological measurement by the called sub-module.

### Claim 9 (depends on 8)

The electronic device according to claim 8, further causing the at least one processor to:
compare the calibrated physiological measurement with a preset threshold range;
generate a prompt if the calibrated physiological measurement is outside of the preset threshold range.

### Claim 10 (depends on 8)

The electronic device according to claim 8, further causing the at least one processor to:
select at least two regression algorithms selected from preset sorting algorithms, to train and to create at least two modules for a same kind of reading of the physiological state;
average the at least two created modules to generate the sub-module in the model for calibration.

### Claim 11 (depends on 10)

The electronic device according to claim 10, further causing the at least one processor to:
obtain a plurality of sets of sample data, the sample data comprises physiological measurements of a user and standard physiological information, where the physiological measurements are taken by the at least one uncalibrated measurement device, and the standard physiological information is contained within or taken by the at least one standard measurement device;
divide the sample data into a training set and a validation set;
establish an analysis model based on a regression algorithm;
train the analysis model via the training set, where an input of the analysis model is the physiological measurements in the training set and an output of the analysis model is the standard physiological information in the training set;
validate the trained analysis model via the validation set to generate a predicted accuracy of the trained analysis model;
determine whether the predicted accuracy of the trained analysis model is less than a preset threshold;
determine the trained analysis model as being the module if the predicted accuracy of the trained analysis model is not less than the preset threshold.

### Claim 12 (depends on 11)

The electronic device according to claim 11, further causing the at least one processor to:
adjust the regression algorithm of the analysis model, or the number of the samples in the training set, or the regression algorithm of the analysis model and the number of the samples in the training set, to resume training the analysis model via the training set if the predicted accuracy of the trained analysis model is less than the preset threshold;
validate the resumed trained analysis model via the validation set to resume generating a predicted accuracy of the resumed trained analysis model;
determine whether the resumed predicted accuracy of the resumed trained analysis model is less than the preset threshold;
determine the resumed trained analysis model as being the module if the resumed predicted accuracy of the resumed trained analysis model is not less than the preset threshold;
continue to adjust the regression algorithm of the analysis model, or the number of the samples in the training set, or the regression algorithm of the analysis model and the number of the samples in the training set, to resume training the analysis model via the training set, and to validate the resumed trained analysis model via the validation set to resume generating a predicted accuracy of the resumed trained analysis model until the predicted accuracy is not less than the preset threshold if the resumed predicted accuracy of the resumed trained analysis model is less than the preset threshold.

### Claim 13 (depends on 11)

The electronic device according to claim 11, further causing the at least one processor to:
obtain a plurality of sets of physiological measurements taken by the at least two uncalibrated measurement devices and a plurality of sets of standard physiological information measured by the at least two standard measurement devices;
combine the sets of physiological measurements and the sets of standard physiological information according to codes of the uncalibrated measurement devices and codes of the standard measurement devices to obtain a combination of each physiological measurement and all the standard physiological information, or a combination of each standard physiological information and all the physiological measurement.

### Claim 14 (depends on 8)

The electronic device according to claim 8, wherein the to-be-calibrated measurement devices comprises one or more of a group of a smart bracelet, smartwatch, a portable oximeter, and a portable scale.

### Claim 15

A non-transitory storage medium storing a set of instructions, when the instructions being executed by at least one processor of an electronic device, causing the at least one processor to:
obtain physiological reading collected by an uncalibrated measurement device;
input the reading into a model for calibration to generate a calibrated physiological measurement whereby a sub-module corresponding to the reading is called from the model for calibration and the sub-module is for calibrating the reading via a regression algorithm, where each sub-module in the model corresponds to one physiological reading;
output calibrated physiological measurement by the called sub-module.

### Claim 16 (depends on 15)

The non-transitory storage medium according to claim 15, further causing the at least one processor to:
compare the calibrated physiological measurement with a preset threshold range;
generate a prompt if the calibrated physiological measurement is outside of the preset threshold range.

### Claim 17 (depends on 15)

The non-transitory storage medium according to claim 15, further causing the at least one processor to:
select at least two regression algorithms selected from preset sorting algorithms, to train and to create at least two modules for a same kind of reading of the physiological state;
average the at least two created modules to generate the sub-module in the model for calibration.

### Claim 18 (depends on 17)

The non-transitory storage medium according to claim 17, further causing the at least one processor to:
obtain a plurality of sets of sample data, the sample data comprises physiological measurements of a user and standard physiological information, where the physiological measurements are taken by the at least one uncalibrated measurement device, and the standard physiological information is contained within or taken by the at least one standard measurement device;
divide the sample data into a training set and a validation set;
establish an analysis model based on a regression algorithm;
train the analysis model via the training set, where an input of the analysis model is the physiological measurements in the training set and an output of the analysis model is the standard physiological information in the training set;
validate the trained analysis model via the validation set to generate a predicted accuracy of the trained analysis model;
determine whether the predicted accuracy of the trained analysis model is less than a preset threshold;
determine the trained analysis model as being the module if the predicted accuracy of the trained analysis model is not less than the preset threshold.

### Claim 19 (depends on 18)

The non-transitory storage medium according to claim 18, further causing the at least one processor to:
adjust the regression algorithm of the analysis model, or the number of the samples in the training set, or the regression algorithm of the analysis model and the number of the samples in the training set, to resume training the analysis model via the training set if the predicted accuracy of the trained analysis model is less than the preset threshold;
validate the resumed trained analysis model via the validation set to resume generating a predicted accuracy of the resumed trained analysis model;
determine whether the resumed predicted accuracy of the resumed trained analysis model is less than the preset threshold;
determine the resumed trained analysis model as being the module if the resumed predicted accuracy of the resumed trained analysis model is not less than the preset threshold;
continue to adjust the regression algorithm of the analysis model, or the number of the samples in the training set, or the regression algorithm of the analysis model and the number of the samples in the training set, to resume training the analysis model via the training set, and to validate the resumed trained analysis model via the validation set to resume generating a predicted accuracy of the resumed trained analysis model until the predicted accuracy is not less than the preset threshold if the resumed predicted accuracy of the resumed trained analysis model is less than the preset threshold.

### Claim 20 (depends on 18)

The non-transitory storage medium according to claim 18, further causing the at least one processor to:
obtain a plurality of sets of physiological measurements taken by the at least two uncalibrated measurement devices and a plurality of sets of standard physiological information measured by the at least two standard measurement devices;
combine the sets of physiological measurements and the sets of standard physiological information according to codes of the uncalibrated measurement devices and codes of the standard measurement devices to obtain a combination of each physiological measurement and all the standard physiological information, or a combination of each standard physiological information and all the physiological measurement.
