Apparatus and method for estimating bio-information

A wearable device for estimating bio-information, the wearable device comprising:

• a main body having a front surface and a rear surface,
• wherein the main body includes:
• a memory storing a bio-information estimation model,
• a communication interface,
• an output interface,
• a pulse wave sensor, and
• a processor configured to:
• control the pulse wave sensor to measure a first pulse wave signal from a skin of a user in response to a request for estimating the bio-information while the rear surface of the main body faces the skin of the user, wherein the first pulse wave signal represents a biological condition of the user and the request for estimating the bio-information is automatically generated in the device according to a predetermined factor including a change in an environment external to the device;
• perform band-pass filtering on the measured first pulse wave signal in a frequency range to remove noise from the measured first pulse wave signal;
• obtain two or more parameters among a maximum point, an onset point, and a point between the maximum point and the onset point, from a waveform of the measured first pulse wave signal;
• combine the obtained two or more parameters to generate a first pulse wave analysis result;
• and
• estimate the bio-information for assessing health of the user based on the generated first pulse wave analysis result and the bio-information estimation model stored in the memory,
• wherein the bio-information estimation model is a predefined model based on neural network or deep learning,
• wherein the pulse wave sensor comprises a photoplethysmogram (PPG) sensor including a light source for emitting light onto the skin facing the rear surface of the main body and a detector for detecting light reflected from the skin facing the rear surface of the main body and measures the first pulse wave signal based on the detected light, and
• wherein the processor is further configured to, based on the two or more parameters being obtained, generate, as the first pulse wave analysis result, a variation in the obtained two or more parameters, based on a reference parameter obtained at a time of calibration of the device,
• wherein the processor is further configured to:
• perform the calibration of the device in response to analysis of the estimated bio-information,
• wherein the performing of the calibration comprises:
• providing the user with a guidance for touching the pulse wave sensor and providing the user a guidance on connection of the wearable device to an external device via the communication interface,
• when the wearable device is connected to the external device, controlling the communication interface to receive reference bio-information from the external device,
• when the wearable device is not connected to the external device, providing a user interface via the output interface and receiving the reference bio-information from the user,
• controlling the pulse wave sensor to measure a second pulse wave signal for the calibration from the skin facing the rear surface of the main body, obtaining reference parameters corresponding to the two or more parameters from the measured second pulse wave signal for the calibration, combining the obtained reference parameters to generate a second plus wave analysis result, and storing the second plus wave analysis result in the memory as reference information for estimating the bio-information.

A method of estimating bio-information by a wearable device for estimating the bio-information, the wearable device including a main body having a front surface and a rear surface, the method comprising:

• measuring, by a pulse wave sensor of the wearable device for estimating the bio-information, a first pulse wave signal from a skin of a user in response to a request for estimating the bio-information while the rear surface of the main body faces the skin of the user, wherein the first pulse wave signal represents a biological condition of the user and the request for estimating the bio-information is automatically generated in the device according to a predetermined factor including a change in an environment external to the device;
• performing band-pass filtering on the measured first pulse wave signal in a frequency range to remove noise from the measured first pulse wave signal;
• obtaining, by the device, two or more parameters among a maximum point, an onset point, and a point between the maximum point and the onset point, from a waveform of the measured first pulse wave signal;
• combining, by the device, the obtained two or more parameters to generate a first pulse wave analysis result;
• estimating, by the device, the bio-information for assessing health of the user based on the generated first pulse wave analysis result and a bio-information estimation model stored in a memory of the device,
• wherein the bio-information estimation model is a predefined model based on neural network or deep learning,
• wherein the measuring of the first pulse wave signal comprises emitting light onto the skin facing the rear surface of the main body by a light source of a photoplethysmogram (PPG) sensor of the device, detecting light reflected from the skin facing the rear surface of the main body by a detector of the PPG sensor of the device, and measuring the first pulse wave signal based on the detected light, and
• wherein the obtaining of the two or more parameters comprises:
• obtaining a plurality of one-period signals from the measured first pulse wave signal;
• obtaining any one of the obtained plurality of one-period signals as a representative waveform, or superpose two or more of the obtained plurality of one-period signals to obtain the representative waveform;
• and
• obtaining the two or more parameters from the obtained representative waveform,
• wherein the method further comprises:
• performing calibration of the device in response to analysis of the estimated bio-information,
• wherein the performing of the calibration comprises:
• providing the user with a guidance for touching the pulse wave sensor and providing the user with a guidance on connection of the wearable device to an external device via a communication interface of the device,
• when the wearable device is connected to the external device, receiving reference bio-information from the external device via the communication interface,
• when the wearable device is not connected to the external device, providing a user interface via an output interface of the device and receiving the reference bio-information from the user, and
• measuring, by the pulse wave sensor, a second pulse wave signal for the calibration from the skin facing the rear surface of the main body, obtaining reference parameters corresponding to the two or more parameters from the measured second pulse wave signal for the calibration, combining the obtained reference parameters to generate a second plus wave analysis result, and storing the second plus wave analysis result in the memory as reference information for estimating the bio-information.

A method of estimating bio-information by a wearable device for estimating the bio-information, the wearable device including a main body having a front surface and a rear surface, the method comprising:

• measuring, by the device for estimating the bio-information, a first pulse wave signal from a skin of a user in response to a request for estimating the bio-information while the rear surface of the main body faces the skin of the user, wherein the first pulse wave signal represents a biological condition of the user and the request for estimating the bio-information is automatically generated in the device according to a predetermined factor including a change in an environment external to the device;
• performing band-pass filtering on the measured first pulse wave signal in a frequency range to remove noise from the measured first pulse wave signal;
• obtaining, by the device, two or more parameters among a maximum point, an onset point, and a point between the maximum point and the onset point, from a waveform of the measured first pulse wave signal;
• combining, by the device, the obtained two or more parameters to generate a first pulse wave analysis result;
• estimating, by the device, the bio-information for assessing health of the user based on the generated first pulse wave analysis result and a bio-information estimation model stored in a memory of the device,
• wherein the bio-information estimation model is a predefined model based on neural network or deep learning,
• wherein the measuring of the first pulse wave signal comprises emitting light onto the skin facing the rear surface of the main body by a light source of a photoplethysmogram (PPG) sensor of the device, detecting light reflected from the skin facing the rear surface of the main body by a detector of the PPG sensor of the device, and measuring the first pulse wave signal based on the detected light, and
• wherein the generating of the first pulse wave analysis result comprises, based on the two or more parameters being obtained, generating, as the first pulse wave analysis result, a variation in the obtained two or more parameters, based on a reference parameter obtained at a time of calibration of the device,
• wherein the method further comprises:
• performing the calibration of the device in response to analysis of the estimated bio-information,
• wherein the performing of the calibration comprises:
• providing the user with a guidance for touching the PPG sensor and providing the user with a guidance on connection of the wearable device to an external device via a communication interface of the device,
• when the wearable device is connected to the external device, receiving reference bio-information from the external device via the communication interface,
• when the wearable device is connected to the external device, providing a user interface via an output interface of the device and receiving the reference bio-information from the user, and
• measuring, by the device, a second pulse wave signal for the calibration from the skin facing the rear surface of the main body, obtaining reference parameters corresponding to the two or more parameters from the measured second pulse wave signal for the calibration, combining the obtained reference parameters to generate a second plus wave analysis result, and storing the second plus wave analysis result in the memory as reference information for estimating the bio-information.