Device and method for estimating biological information

Abstract

A device for estimating biological information according to an embodiment of the present disclosure includes a processor configured to obtain a spectrum from an object, obtain a component generated based on a change in pressure applied to the object, correct the spectrum based on the obtained component generated based on the pressure change, and estimate the biological information of the object based on the corrected spectrum.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application is based on and claims the benefit of priority from Korean Patent Application No. 10-2019-0101666, filed on August 20, 2019, in the Korean Intellectual Property Office, the disclosure of which is incorporated herein in its entirety by reference. Technical Field

[0003] The following description relates generally to an apparatus and method for estimating biological information, and more particularly to a technique for correcting a component in a spectrum obtained from a subject that is generated due to a change in pressure applied by the subject to a spectrum sensor. Background Art

[0004] Recently, methods for non-invasively measuring bioinformation such as blood sugar by using Raman spectroscopy or near-infrared spectroscopy have been studied. Bioinformation measuring devices using spectral technology generally include a light source for emitting light to an object and a detector for detecting an optical signal reflected by the object. The bioinformation measuring device reconstructs the spectrum by using the optical signal detected by the detector, and measures biological components in the body, such as carotenoids, blood sugar, calories, etc. by analyzing the reconstructed spectrum. The light absorption of hemoglobin in the blood affects the entire skin spectrum. Therefore, in order to minimize the absorption of light by hemoglobin in the blood, some optical sensors apply a pressure greater than or equal to a predetermined value to the skin when measuring the spectrum of human skin. These sensors may include pressure sensors, force sensors, etc. to improve accuracy. However, the sensor may not be manufactured in a compact size. In addition, the use of such sensors may increase the measurement time. Summary of the invention

[0005] According to one aspect of the present disclosure, an apparatus for estimating biological information includes a processor configured to obtain a spectrum from an object, obtain a component generated based on a change in pressure applied to the object, correct the spectrum based on the obtained component generated based on the pressure change, and estimate the biological information of the object based on the corrected spectrum.

[0006] The processor may perform principal component analysis (PCA) on the obtained spectrum, and the processor may obtain a component generated based on the pressure change based on a result of performing the PCA.

[0007] The processor may obtain the component generated based on the pressure change based on the shape of the component analyzed by using PCA.

[0008] The processor may obtain, as the component generated based on the pressure change, a component having a shape similar to the absorption peak of hemoglobin among the components analyzed by using PCA.

[0009] The processor may obtain, among the components analyzed by the PCA, a component having a spectral shape similar to a component defined in the component database as a component generated based on the pressure change.

[0010] The processor may obtain, from the composition database, a composition generated based on another pressure change applied by the object to the spectral sensor.

[0011] The processor may collect spectra obtained from a plurality of subjects as training data, perform principal component analysis (PCA) on the collected training data, and store spectral shapes and principal component scores of principal components in a component database based on performing the PCA.

[0012] The processor may perform PCA on the spectrum obtained from the object, and update the component database by using the components analyzed using the PCA.

[0013] The processor may determine whether to update the component database based on at least one of a PCA result of a spectrum obtained from the object, a generation date and an update date of the component database, and a collection environment of the training data.

[0014] The apparatus may include a communication interface that receives the ingredient database from an external device in response to determining to update the ingredient database.

[0015] The processor may correct the spectral changes caused by the pressure changes by removing the components based on the pressure changes from the acquired spectrum.

[0016] The processor may remove the obtained component from the obtained spectrum by using a least squares method.

[0017] The processor may obtain the composition based on the pressure change based on the spectrum obtained during the initial pressure stage over the entire time period in which the spectrum is obtained.

[0018] The processor may obtain a noise-related component of the obtained spectrum, and correct the obtained spectrum based on the component generated based on the pressure change and the noise-related component.

[0019] The processor may obtain the noise-related component based on a correlation between a principal component score of a principal component analyzed by performing principal component analysis on at least one of the obtained spectra and an estimated bio-information value estimated from the at least one spectrum.

[0020] The biological information of the subject may include at least one of blood sugar, triglyceride, cholesterol, calorie, protein, carotenoid, lactate, and uric acid.

[0021] A method for estimating biological information of an object includes: obtaining a spectrum from the object; obtaining a component generated based on a change in pressure applied to the object; correcting the spectrum according to the obtained component generated based on the pressure change; and estimating the biological information of the object based on the corrected spectrum.

[0022] The method may include performing principal component analysis (PCA) on the obtained spectrum, and obtaining a component generated based on a pressure change based on a result of the PCA.

[0023] The obtaining of the component generated based on the pressure change may include obtaining the component generated based on the pressure change based on a shape of the component analyzed by using PCA.

[0024] Obtaining the component generated based on the pressure change may include obtaining, as the component generated based on the pressure change, a component having a shape similar to an absorption peak of hemoglobin among the components analyzed by using PCA.

[0025] Obtaining the component generated based on the pressure change may include obtaining, among the components analyzed by PCA, a component having a spectral shape similar to a component defined in a component database as the component generated based on the pressure change.

[0026] Obtaining the component generated based on the pressure change may include: obtaining the component generated based on the pressure change applied to the spectral sensor by the object from a component database.

[0027] Correcting the spectrum may include correcting a change in the spectrum caused by a pressure change by removing a component generated based on the pressure change from the acquired spectrum.

[0028] Obtaining the component due to the pressure change may include obtaining the component due to the pressure change based on a spectrum obtained during an initial pressure stage of the entire time period for obtaining the spectrum.

[0029] The method may include obtaining a noise-related component of the obtained spectrum, and correcting the spectrum may include correcting the obtained spectrum according to a component generated based on a pressure change and the noise-related component.

[0030] Obtaining the noise-related component may include obtaining the noise-related component based on a correlation between a principal component score of a principal component analyzed by performing principal component analysis on at least one of the obtained spectra and an estimated bio-information value estimated from the at least one spectrum. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The above and other aspects, features and advantages of certain embodiments of the present disclosure will become more apparent from the following description in conjunction with the accompanying drawings, in which:

[0032] Figure 1 is a block diagram showing an apparatus for estimating biological information according to an embodiment;

[0033] Figure 2 is a diagram schematically showing an example of the structure of a spectrum sensor according to an embodiment;

[0034] FIG. 3A to FIG. 3C is a diagram explaining an example of correcting a spectrum by using principal component analysis according to an embodiment;

[0035] FIG. 4A to FIG. 4D is a diagram explaining another example of correcting a spectrum by using principal component analysis according to the embodiment;

[0036] Figure 5 is a block diagram showing an apparatus for estimating bio-information according to another embodiment;

[0037] Figure 6 is a block diagram showing an apparatus for estimating biological information according to yet another embodiment;

[0038] Figure 7 is a flowchart illustrating a method of estimating biological information according to an embodiment;

[0039] Figure 8 is a flowchart illustrating a method of estimating biological information according to another embodiment; and

[0040] Fig. 9 is a diagram illustrating an example of a wearable device to which an embodiment of the apparatus for estimating biological information can be applied.

[0041] Throughout the drawings and detailed description, unless otherwise described, the same reference numerals may refer to the same elements, features, and structures. The relative sizes and depictions of these elements may be exaggerated for clarity, illustration, and convenience. DETAILED DESCRIPTION

[0042] Details of other embodiments are included in the following detailed description and drawings. According to the embodiments described in detail below with reference to the drawings, the advantages and features of the present disclosure and their implementation methods will be more clearly understood. Throughout the drawings and detailed description, unless otherwise described, the same reference numerals may refer to the same elements, features and structures.

[0043] It should be understood that although the terms "first", "second", etc. can be used in this article to describe various elements, these elements should not be limited by these terms. These terms can be used to distinguish one element from another element. Unless otherwise clearly stated, any reference to the singular form of the term can include the plural form of the term. In addition, unless explicitly described to the contrary, expressions such as "including" or "comprising" can mean including the elements described, but do not exclude any other elements. In addition, terms such as "part", "module", etc. can refer to a unit that performs at least one function or operation, and can be embodied as hardware, software or a combination thereof.

[0044] Hereinafter, embodiments of an apparatus and method for estimating bio-information will be described in detail with reference to the accompanying drawings.

[0045] Figure 1 is a block diagram illustrating an apparatus for estimating biological information according to an embodiment of the present disclosure.

[0046] refer to Figure 1 , an apparatus 100 for estimating biological information includes a spectral sensor 110 and a processor 120 .

[0047] The spectral sensor 110 can continuously obtain spectral data from the object within a predetermined time period. In this case, the object can be, for example, skin tissue of a human body, such as a wrist, a finger, etc., where veins or capillaries are present, or the object can be an area on the wrist adjacent to a radial artery. However, the object is not limited thereto. The spectral sensor 110 can measure a spectrum by using diffuse reflectance spectroscopy, absorption spectroscopy, Raman spectroscopy, near infrared spectroscopy, or mid-infrared spectroscopy.

[0048] The spectral sensor 110 may include one or more light sources configured to emit light toward an object and one or more detectors configured to detect light scattered by or reflected from an object. The light source may include a light emitting diode (LED), a laser diode, a phosphor, etc. A plurality of light sources may emit light of different wavelengths. In this case, a color filter for transmitting or blocking light in a specific wavelength range may be disposed on top of at least one light source.

[0049] The detector may include one pixel or an array of two or more pixels, each of which may include a photodiode or a phototransistor. Based on the detected light, the detector may convert the detected light signal into an electrical signal. A collimator such as a microlens that improves light collection efficiency may be disposed on top of each pixel.

[0050] Upon receiving the request to measure the spectrum, the processor 120 may control the spectrum sensor 110. Upon receiving the request to measure the spectrum, the processor 120 may output information guiding the user to touch the spectrum sensor 110 with an object and change the pressure applied to the spectrum sensor 110 within a predetermined period of time.

[0051] Based on the user touching the spectrum sensor 110 with an object and changing the contact pressure within a predetermined period of time, the spectrum sensor 110 may emit light toward the object within a predetermined period of time and may detect the light reflected by the object.

[0052] Based on receiving the spectrum data from the spectrum sensor 110 , the processor 120 may restore the spectrum based on the received spectrum data.

[0053] Figure 2 is a diagram schematically showing an example of the structure of the spectral sensor 110 . Figure 2 The structure of the spectrum sensor 110 shown in FIG. 1 is merely an example, and thus the spectrum sensor 110 is not limited thereto and may have various other structures.

[0054] refer to Figure 2 The spectrum sensor 110 according to the embodiment includes an LED array LA having n LED light sources arranged on a circular frame. Here, the shape of the frame is not limited to a circular shape, but may be modified according to the shape of the apparatus 100 for estimating biological information.

[0055] Each LED light source may have at least some peak wavelengths in different wavelength bands. For example, each LED light source may emit light toward the object OBJ simultaneously or sequentially. The peak wavelength of each LED light source may be preset and may be set based on the spectrum measurement part, the target component to be analyzed, etc. After each LED light source emits light toward the object, the emitted light is absorbed into the object, or reflected or scattered from the object, depending on the tissue characteristics of the object. In this case, the light reaction characteristics of the object may vary according to the type of the object and the wavelength of the light, and the degree of absorption, reflection, transmission or scattering of the light by the object may vary according to the light reaction characteristics of the object. In addition, the spectral sensor 110 may include a detector CS, which is disposed at the center of the circular frame and detects the scattered or reflected light L2 when the light L1 is emitted by the LED light source LA toward the object and scattered or reflected from the object. For example, the detector CS may be a sensor based on a complementary metal oxide semiconductor (CMOS) image sensor (CIS), and a spectral filter for detecting light of various wavelengths may be provided on the CIS, but the detector CS is not limited thereto.

[0056] In addition, the spectrum sensor 110 may include a light blocking portion (LB) that prevents light emitted by the LED light source LA from being emitted directly toward the detector CS without first being emitted toward the object, and which guides light scattered by or reflected from the object toward the detector CS.

[0057] Based on the spectrum obtained from the spectrum sensor 110, the processor 120 can process the spectrum to analyze biological information from the object, such as body surface components or body components, and can estimate the biological information by using the processed spectrum. In this case, the biological information may include, for example, blood sugar, calories, ethanol, triglycerides, proteins, cholesterol, uric acid, carotenoids, etc., but is not limited thereto.

[0058] When light is emitted toward the subject by the light source of the spectrum sensor 110 and the light is absorbed into body tissue or scattered by body tissue or reflected from body tissue, the light absorption by hemoglobin in the blood can significantly affect the entire skin spectrum. Generally, when measuring the spectrum from the subject, a pressure greater than or equal to a predetermined value can be applied to the subject to minimize the absorption of light by hemoglobin in the blood. However, when pressure is applied to the subject, the spectrum may change, and the spectrum changes dynamically according to the intensity of the pressure applied to the subject, the time period for which the pressure is applied, etc.

[0059] In a spectrum continuously obtained by the spectral sensor 110 during a predetermined period of time (hereinafter referred to as a “first spectrum”), the processor 120 may correct a spectrum change caused by a change in pressure applied to the object when the object presses the spectral sensor 110 .

[0060] For example, the processor 120 may extract a component (hereinafter referred to as a "pressure component") generated due to a change in pressure of the object from the first spectrum, and may correct the first spectrum based on the extracted pressure component. For example, the processor 120 may extract the pressure component from the entire first spectrum or a portion of the first spectrum (hereinafter referred to as a "second spectrum") by using principal component analysis. In this case, the second spectrum may be a spectrum obtained in an initial pressure stage when pressure begins to be applied after the object contacts the spectral sensor 110. The initial pressure stage may be predetermined and may be appropriately adjusted by considering the spectral processing speed and / or the bio-information estimation speed.

[0061] Furthermore, based on the extracted pressure component, the processor 120 may remove the pressure component from the entire first spectrum or from the spectrum obtained after the initial pressure stage, thereby obtaining a spectrum that is less affected by the pressure change of the object.

[0062] FIG. 3A to FIG. 3C is a diagram explaining an example of correcting a spectrum by using principal component analysis.

[0063] Figure 3A is a diagram showing a first spectrum obtained by the spectrum sensor 110 within a predetermined period of time, Figure 3B is a graph showing the results of principal component analysis of the second spectrum obtained in the initial pressure stage. Figure 3A and Figure 3B As shown, the processor 120 can extract the pressure component PC1 generated due to the pressure change of the object from the result of the principal component analysis. The processor 120 can extract the pressure component PC1 based on the spectral shape of the principal component. By comparing the spectral shape of the principal component analyzed using the principal component analysis with the hemoglobin absorption peak HbAP, the processor 120 can extract the principal component having a shape similar to the hemoglobin absorption peak as the pressure component PC1 of the object. In this case, the preset information can be used as the hemoglobin absorption peak HbAP.

[0064] Figure 3C is a diagram showing a corrected spectrum. Based on extracting the pressure component by using principal component analysis, the processor 120 can remove the pressure component PCI from the first spectrum by using, for example, a least square method. Figure 3B and Figure 3C , it can be seen that a predetermined wavelength range (a range of approximately 430 nm to 570 nm) in the first spectrum, ie, a wavelength range affected by a pressure change of an object, is corrected by principal component analysis.

[0065] In addition, the processor 120 can remove noise from the first spectrum, which causes spectral changes other than those caused by pressure changes. For example, the noise can include various factors that affect spectral accuracy, such as temperature, humidity, motion noise, detector noise, light source noise, etc., but is not limited thereto.

[0066] For example, the processor 120 may extract the noise component based on the correlation between the principal component scores of the principal components analyzed by using principal component analysis and the estimated bio-information value estimated from the spectrum before correction. In this case, the correlation may include at least one of the Euclidean distance, the Pearson correlation coefficient, the Spearman correlation coefficient, and the cosine similarity, but is not limited thereto.

[0067] FIG. 4A to FIG. 4C is a diagram explaining another example of correcting a spectrum by using principal component analysis.

[0068] Figure 4A : is a graph showing the characteristics of three principal components PC1, PC2 and PC3 for each wavelength, which are obtained by principal component analysis of 100 second spectra obtained in the initial pressure stage. Here, it is assumed that the first component PC1 is a pressure component. Figure 4BShows Figure 4A Here, the X-axis represents the number of times the principal component is analyzed, for example, it indicates that the principal component is analyzed 100 times by using each of the 100 spectra; and the Y-axis represents the principal component score of each analyzed principal component.

[0069] As described above, the processor 120 can analyze the principal component by using each of the 100 second spectra in the initial pressure stage, and can estimate carotenoids. In particular, various noises such as contact failure may exist in the initial pressure stage, so that the accuracy of the second spectrum may be reduced. If carotenoids are estimated by using each second spectrum including a noise component, the estimated value does not fall within a predetermined range and may fluctuate significantly. Therefore, in the case where there is a high correlation between the fluctuation shape of the estimated carotenoid value and the principal component score, the processor 120 can determine the principal component as a factor causing the fluctuation of the estimated carotenoid value.

[0070] Figure 4C : is a graph showing the correlation between the principal components PC2 and PC3 excluding the pressure component PCI among the principal components PCI, PC2 and PC3 and the estimated carotenoid value. Figure 4C , it can be seen that the second principal component PC2 has a relatively low correlation with the estimated carotenoid value, and the third principal component PC3 has a relatively high correlation with the estimated carotenoid value. In this case, the processor 120 may determine the third principal component PC3 having a correlation greater than or equal to a predetermined threshold as a noise component. By removing the determined noise component PC3 from the entire first spectrum or from the spectrum after the initial pressure stage, the processor 120 may obtain a spectrum that is less affected by pressure changes and noise effects, such as Figure 4D shown.

[0071] Figure 5 is a block diagram illustrating an apparatus for estimating biological information according to another embodiment of the present disclosure.

[0072] refer to Figure 5 , the apparatus 500 for estimating biological information includes a spectral sensor 510 , a processor 520 , and a storage device 530 .

[0073] The spectrum sensor 510 may include a light source and a detector as described above, and may continuously obtain a light absorption spectrum within a predetermined period of time.

[0074] The processor 520 may obtain a first spectrum based on the obtained spectrum data, and by referring to a component database (DB) 531 of the storage device 530 , the processor 520 may remove a component due to a pressure change of the object and / or a noise component from the first spectrum.

[0075] For example, the processor 520 may perform principal component analysis by using the second spectrum obtained in the initial pressure stage. In addition, the processor 520 may obtain a component having a shape similar to the pressure component obtained from the component DB 531 among the principal components obtained by using the principal component analysis as a pressure component generated due to the pressure change of the object. By removing the pressure component from the entire first spectrum or the spectrum after the initial pressure stage, the processor 520 may obtain a spectrum that is less affected by the pressure change of the object.

[0076] In another example, the processor 520 may obtain a pressure component representing a spectrum change caused by a pressure change of the object from the component DB 531 by referring to the storage device 530. The processor 520 may obtain a corrected spectrum by removing the pressure component obtained from the component DB 531 from the first spectrum.

[0077] In addition, the processor 520 may obtain a noise-related component from the component DB 531, and may remove the noise-related component from the first spectrum or the spectrum after the initial stress stage. Alternatively, the processor 520 may obtain a component similar to the noise component stored in the component DB 531 from the result of the principal component analysis of the second spectrum, and may remove the obtained component from the first spectrum or the spectrum after the initial stress stage.

[0078] In addition, the processor 520 may determine whether to use the component DB 531 based on the generation date and update date of the component DB 531, the training data collection environment, or the surrounding environment of the current user, etc. In this case, the training data may refer to spectral data obtained from a plurality of subjects. In addition, the training data collection environment or the surrounding environment of the current user may include the age, gender, and health status of the subject from which the spectrum is obtained, the surrounding environment (e.g., temperature, humidity, etc.) when the spectrum is obtained, etc.

[0079] For example, if the update date of the component DB 531 is within a predetermined time period from the current time, the processor 520 may correct the spectrum by immediately using the pressure component obtained from the component DB 531. In another example, even when the update date is within the predetermined time period, if the training data collection environment is substantially different from the surrounding environment of the current user, the processor 520 may correct the spectrum by comparing the result of the principal component analysis of the spectrum obtained from the current user with the spectral shape of the pressure component obtained from the component DB 531. In yet another example, in the case where it is determined that the data of the component DB 531 is unreliable, such as in the case where the update date is within the predetermined time period, the processor 520 may correct the spectrum by obtaining the pressure component based on the result of the principal component analysis of the spectrum and the shape of the hemoglobin absorption peak as described above. These examples are intended to help understand the present disclosure and should not be construed as limiting the scope thereof.

[0080] The storage device 530 may store a component DB 531 including spectral shapes and / or principal component scores of principal components analyzed by principal component analysis of a plurality of training data. In this case, the training data may be spectra obtained from a plurality of subjects; and the component DB 531 may manage the results of principal component analysis of a plurality of groups classified according to health conditions, age, and / or gender of the plurality of subjects. By considering the age and gender of the user, when extracting a stress component from the component DB 531, the processor 520 may obtain a stress component and / or a noise component of a corresponding group.

[0081] The storage device 530 may store various reference information related to the spectrum processing in addition to the component DB 531. For example, the reference information may include a criterion for determining whether to use the component DB 531, an initial stress stage for principal component analysis, a bio-information estimation algorithm and / or a user's health status, age, and gender, etc.

[0082] The storage device 530 may include at least one of the following storage media: flash memory, hard disk memory, multimedia card micro memory, card-type memory (for example, secure digital (SD) memory, extreme digital (XD) memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk and optical disk, etc., but not limited to these.

[0083] In addition, as described above, the processor 520 may determine whether to update the component DB 531 based on the generation date and update date of the component DB 531, the training data collection environment, or the surrounding environment of the current user, etc. For example, if the last update date of the component DB 531 is within a predetermined time period, the processor 520 may determine to update the component DB 531.

[0084] For example, when determining to update component DB 531, processor 520 may perform principal component analysis on the spectrum obtained from the user's object, and may update component DB 531 based on the result of the principal component analysis. Processor 520 may control spectrum sensor 510 multiple times at predetermined intervals, and may collect spectra obtained from the user as training data multiple times.

[0085] The processor 520 may replace the existing data of the component DB 531 with data including a pressure component obtained based on the principal component analysis result of the spectrum and the shape of the hemoglobin absorption peak and / or a noise component, etc., wherein the pressure component is obtained based on the principal component analysis result and the shape of the hemoglobin absorption peak, and the noise component is determined based on the correlation between the principal component analysis result and the estimated biological information value. Alternatively, the processor 520 may update the existing data of the component DB 531 by using data generated by combining the data obtained through the principal component analysis with the existing data of the component DB 531.

[0086] In another example, the processor 520 may obtain spectra obtained from a plurality of subjects as training data. When the spectra are obtained from a plurality of subjects, the processor 520 may analyze the principal component analysis results of each group by considering the health conditions, age and / or gender of the plurality of subjects, and may update the component DB 531 by using the principal component analysis results of each group.

[0087] The processor 520 can provide the spectral processing result to the user by using an output component. In this case, the output component may include a visual output component (e.g., a display, etc.), an audio output component (e.g., a speaker, etc.), a tactile component configured to provide vibration, tactile sensation, etc., but is not limited thereto. For example, the processor 520 can output the pre-corrected spectrum, the principal component analysis result, and / or the corrected spectrum obtained from the user's object to the display, and can divide the display into a plurality of regions to output the pre-corrected spectrum and the corrected spectrum in various regions, so that the user can easily visually compare the spectra.

[0088] In addition, based on receiving a request for the corrected spectrum from the external device, the processor 520 may send the corrected spectrum to the external device via the communication interface. Alternatively, if it is determined to update the component DB 531, the processor 520 may receive the principal component analysis result from an external device dedicated to performing principal component analysis on a plurality of subjects. In this case, examples of the external device may include information processing devices such as smart phones, tablet personal computers (PCs), desktop computers, laptop computers, etc., and may be devices having a function of estimating bio-information by using a spectrum.

[0089] The communication interface can communicate with the external device by using the following communication technologies: for example, Bluetooth communication, Bluetooth low energy (BLE) communication, near field communication (NFC), wireless local area network (WLAN) communication, Zigbee communication, infrared data association (IrDA) communication, wireless fidelity (Wi-Fi) direct (WFD) communication, ultra-wideband (UWB) communication, Ant+ communication, Wi-Fi communication, radio frequency identification (RFID) communication, third generation (3G) communication, fourth generation (4G) communication, fifth generation (5G) communication, etc. However, the foregoing examples are exemplary and non-limiting.

[0090] Figure 6 is a block diagram illustrating an apparatus 600 for estimating biological information according to yet another embodiment of the present disclosure.

[0091] refer to Figure 6 The apparatus 600 for estimating biological information includes a spectral sensor 610 , a processor 620 , a storage device 630 , an output interface 640 , and a communication interface 650 .

[0092] The spectrum sensor 610 may obtain a continuous spectrum from an object of a user. The spectrum sensor 610 may include a light source and a detector. Under the control of the processor 620, the light source may emit light toward the object; and the detector may detect light reflected by the object after the light is absorbed into the object or scattered by the object or reflected from the object, may convert the detected light into an electrical signal indicating the light intensity, and may send the signal to the processor 620.

[0093] The processor 620 may be electrically connected to the spectral sensor 610 and various other components 630 , 640 , and 650 to control operations thereof.

[0094] In response to a user's input or at a predetermined bio-information estimation interval, the processor 620 may control the spectrum sensor 610 , and may restore a spectrum by receiving spectrum data obtained by the spectrum sensor 610 .

[0095] For example, based on the restored spectrum, the processor 620 can perform principal component analysis by using a set of spectra obtained in the initial pressure stage; and by using the principal component analysis results, the processor 620 can extract the pressure component of the object, that is, the component caused by the spectral change due to the pressure change of the object when the object contacts the spectral sensor and applies pressure to it.

[0096] In another example, when the component DB is included in the storage device 630, the processor 620 may obtain the pressure component by referring to the component DB. In this case, the processor 620 may use the pressure component obtained from the component DB, and may obtain the pressure component of the object from the principal component analysis result based on comparing the principal component analysis result of the spectrum obtained by the spectrum sensor 610 with the pressure component obtained from the component DB. In this case, the criteria for determining whether to use the component DB may be predetermined based on, for example, the generation date and update date of the component DB, the training data collection environment, or the surrounding environment of the user.

[0097] In addition, the processor 620 may determine whether to update the component DB based on the generation date and update date of the component DB, the training data collection environment, or the user's surrounding environment, etc. For example, based on determining that the component DB is to be updated, the processor 620 may update the component DB based on the principal component analysis result of the spectrum obtained from the user. In another example, the processor 620 may receive the component DB from an external device via the communication interface 650.

[0098] Furthermore, by removing the obtained pressure component from the spectrum obtained by the spectrum sensor 610 , the processor 620 may correct the spectrum change caused by the pressure change of the object.

[0099] In addition, the processor 620 can extract the noise-related components of the object other than the pressure component, and can obtain a more accurate spectrum by further removing the extracted components. In this case, the noise-related components can be obtained by principal component analysis of the spectrum or based on the component DB, which is described in detail above.

[0100] By using the corrected spectrum, the processor 620 can estimate biological information, such as blood sugar, triglycerides, cholesterol, calories, proteins, carotenoids, lactate, uric acid, etc. However, the estimation of biological information is not limited to this. For example, an estimation model that represents the correlation between the corrected spectrum and the biological information to be obtained can be predefined. The processor 620 can estimate the biological information based on the corrected spectrum and the estimation model.

[0101] The storage device 630 may store the component DB and various reference information related to estimating the bio-information. For example, the reference information may include a criterion for determining whether to use the component DB, an initial stress stage for principal component analysis, a bio-information estimation algorithm, and / or a user's health status, age, and gender, etc., but is not limited thereto.

[0102] The output interface 640 can provide the user with the processing result of the processor 620. In this case, the output interface 640 may include a visual output interface (e.g., a display, etc.), an audio output interface (e.g., a speaker, etc.), a tactile output interface configured to provide vibration, tactile sensation, etc., but is not limited thereto.

[0103] The communication interface 650 can communicate with an external device to send and receive data related to estimated bio-information. For example, the communication interface 650 can send the pre-corrected and post-corrected spectra, the principal component analysis results, the bio-information estimation results, etc. to the external device, and can receive the component DB, etc. from the external device. In this case, examples of the external device may include information processing devices such as smart phones, tablet PCs, desktop computers, laptop computers, etc., but are not limited thereto.

[0104] Figure 7 is a flowchart illustrating a method of estimating biological information according to an embodiment of the present disclosure.

[0105] Figure 7 This is an example of a method of estimating biological information performed by the apparatus for estimating biological information according to the above-described embodiment described in detail above, and thus will be briefly described below.

[0106] The apparatus for estimating bio-information may obtain a spectrum from the object in operation 710. The apparatus for estimating bio-information 100 may control the spectrum sensor to continuously obtain spectrum data from the object within a predetermined period of time.

[0107] Then, in operation 720, the apparatus for estimating bio-information may obtain a component generated based on a pressure change of the object based on the obtained spectrum and / or the component DB.

[0108] For example, as referenced above Figure 1As described above, based on the obtained spectrum, the apparatus 100 for estimating biological information can perform principal component analysis by using a certain set of spectrums obtained in the initial pressure stage; and by using the principal component analysis result, the apparatus 100 for estimating biological information can obtain a pressure component generated based on the pressure change of the object. For example, the shape of the hemoglobin absorption peak changes as the object applies pressure, so that the apparatus 100 for estimating biological information can obtain a component having a spectral shape similar to the known hemoglobin absorption peak among the spectral shapes of the principal components of the spectrum as the pressure component.

[0109] In another example, as mentioned above Figure 5 As described above, based on the obtained spectrum, the apparatus 500 for estimating biological information can obtain the pressure component from the pre-generated component DB. Alternatively, the apparatus 500 for estimating biological information can perform principal component analysis on the spectrum obtained in operation 710, and can obtain a component having a spectral shape similar to the shape of the pressure component stored in the component DB among the spectral shapes of the principal components as the pressure component generated based on the pressure change of the object.

[0110] Then, the apparatus 500 for estimating bio-information may correct the spectrum based on the obtained pressure component in operation 730. For example, by removing the pressure component obtained in operation 720 from the spectrum obtained in operation 710 using a technique such as a least square method, the apparatus 500 for estimating bio-information may correct the spectrum change caused by the pressure change of the subject.

[0111] Subsequently, in operation 740 , the apparatus 500 for estimating bio-information may estimate bio-information based on the corrected spectrum.

[0112] Figure 8 is a flowchart illustrating a method of estimating biological information according to another embodiment of the present disclosure.

[0113] Figure 8 This is an example of a method of estimating biological information performed by the apparatus for estimating biological information according to the above-described embodiment described in detail above, and thus will be briefly described below.

[0114] The apparatus for estimating bio-information may obtain a spectrum from the object in operation 810. The apparatus for estimating bio-information 100 may control the spectrum sensor to continuously obtain spectrum data from the object within a predetermined period of time.

[0115] Then, in operation 820, the apparatus for estimating bio-information may obtain a component generated based on a pressure change of the object based on the obtained spectrum and / or the component DB.

[0116] Subsequently, in operation 830, the apparatus for estimating bio-information may obtain noise-related components other than components generated based on pressure changes of the object based on the obtained spectrum and / or component DB. For example, the apparatus for estimating bio-information may obtain components having a high correlation between the principal component scores of the principal components and the estimated bio-information values ​​as noise-related components. In another example, the apparatus for estimating bio-information may obtain noise-related components from the component DB. Alternatively, the apparatus for estimating bio-information may perform principal component analysis on the spectrum obtained in operation 810, and may obtain components having a spectral shape similar to the shape of the noise-related components stored in the component DB among the spectral shapes of the principal components as noise-related components.

[0117] Next, the apparatus for estimating bio-information may correct the spectrum based on the obtained pressure component in operation 840. For example, the apparatus for estimating bio-information may correct the spectrum change by removing the pressure component obtained in operation 820 and the noise-related component obtained in operation 830 from the spectrum obtained in operation 810 using a technique such as a least square method.

[0118] Then, in operation 850, the means for estimating bio-information may estimate the bio-information based on the corrected spectrum.

[0119] Fig. 9 is a diagram illustrating an example of a wearable device to which an embodiment of the apparatus for estimating biological information is applied.

[0120] The apparatuses 100, 500, and 600 for estimating bio-information according to the above-described embodiments may be installed in the wearable device 900. Fig. 9 A smart watch type wearable device 900 is shown, but the wearable device is not limited thereto and may be modified into various information processing devices such as a smartphone, a tablet PC, and the like.

[0121] refer to Fig. 9 , the wearable device 900 includes a main body 910 and a band 930 , and various modules of the aforementioned apparatuses 100 , 500 , and 600 for estimating bio-information may be installed in the main body 910 .

[0122] The strap 930 may be made of a flexible material and may be connected to the body 910 .

[0123] The band 930 may be bent to be wrapped around the user's wrist, or may be bent in a manner that allows the band 930 to be removed from the wrist. In this case, a battery may be embedded in the body 910 or the band 930 to power the wearable device 900.

[0124] like Fig. 9As shown, the spectrum sensor 920 may be mounted on the rear surface of the body 910 at a position contacting the user's wrist. For example, the spectrum sensor 920 may include a linear variable filter (LVF) having a spectrum characteristic that changes linearly over the entire length.

[0125] The processor, the storage device, and the communication interface may be installed in the body 910 of the wearable device 900 .

[0126] The processor may correct the spectrum obtained by the spectrum sensor 920 through principal component analysis, and may estimate biological information based on the corrected spectrum.

[0127] The display of the output interface may be mounted on the front surface of the body 910 and may output various information to the user. In addition, the display may include a touch screen for receiving a user's touch input, and may receive the touch input and send the touch input to the processor.

[0128] In addition, the body 910 of the wearable device 900 may include an input component 940, which is used to operate the function of estimating biometric information and various other functions of the wearable device 900 (e.g., clock application, music application, data video application, text message application, etc.). The input component 940 can receive the user's input and can send the input to the processor. In addition, the input component 940 may include a power button to turn on / off the wearable device 900.

[0129] The present disclosure may be implemented as computer-readable codes stored on a non-transitory computer-readable medium. The computer-readable medium may be any type of recording device that stores data in a computer-readable manner.

[0130] Examples of computer-readable media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage, and carrier wave (e.g., data transmission via the Internet). Computer-readable media can be distributed on multiple computer systems connected to the network so that computer-readable codes are written therein and executed therefrom in a decentralized manner. Those of ordinary skill in the art can derive functional programs, codes, and code segments for implementing the present disclosure.

[0131] The present disclosure has been described herein with reference to various embodiments. However, it is apparent to those skilled in the art that various changes and modifications may be made without departing from the technical concept of the present disclosure. Therefore, it is clear that the above embodiments are illustrative in all aspects and are not intended to limit the present disclosure.

Claims

1. A device for estimating biological information, include: The processor is configured as: obtaining a spectrum from the object; obtaining a component generated based on a change in pressure applied to the object; correcting the spectrum based on the obtained pressure change-based component; as well as estimating biological information of the subject based on the corrected spectrum, Wherein, the processor is further configured to: Perform principal component analysis PCA on the obtained spectra, and wherein the processor is configured to obtain the component generated based on the pressure change based on the result of performing PCA, Wherein, the processor is configured to: The components generated based on the pressure change are obtained based on the shapes of the components analyzed by using PCA, Wherein, the processor is further configured to: obtaining a component having a shape similar to the absorption peak of hemoglobin among the components analyzed by using PCA as the component generated based on the pressure change, or obtaining, from among the components analyzed by PCA, a component having a spectral shape similar to the spectral shape of a component defined in a component database as the component generated based on the pressure change, Wherein, the processor is further configured to: Correcting the spectrum change caused by the pressure change by removing the component generated based on the pressure change from the acquired spectrum, Wherein, the processor is further configured to: The obtained components were removed from the obtained spectrum by using the least square method.

2. The device according to claim 1, in, The processor is further configured to: A component generated based on another pressure change applied to the spectral sensor by the object is obtained from the component database.

3. The device according to claim 2, in, The processor is further configured to: collecting spectra obtained from multiple subjects as training data; Perform principal component analysis (PCA) on the collected training data; as well as Based on performing PCA, the spectral shapes of the principal components and the principal component scores are stored in the component database.

4. The device according to claim 3, in, The processor is further configured to: performing PCA on spectra obtained from the subject; and The component database is updated by using the components analyzed using PCA.

5. The device according to claim 4, in, The processor is further configured to: Whether to update the component database is determined based on at least one of a PCA result of the spectrum obtained from the object, a generation date and an update date of the component database, and a collection environment of the training data.

6. The device according to claim 5, further comprising a communication interface, in, In response to determining to update the ingredient database, the communication interface receives an ingredient database from an external device.

7. The device according to claim 1, in, The processor is further configured to: The pressure change-based component is obtained based on the spectrum obtained during the initial pressure stage over the entire time period for obtaining the spectrum.

8. The device according to claim 1, in, The processor is further configured to: obtaining a noise-related component of the obtained spectrum; and The obtained spectrum is corrected based on the component generated based on the pressure change and the noise-related component.

9. The device according to claim 8, in, The processor is further configured to obtain the noise-related component based on a correlation between a principal component score of a principal component analyzed by performing principal component analysis on at least one of the obtained spectra and an estimated bio-information value estimated from the at least one spectrum.

10. The device according to claim 1, in, The biological information of the subject includes at least one of blood sugar, triglyceride, cholesterol, calories, protein, carotenoids, lactate, and uric acid.

11. A method for estimating biological information of an object, the method include: obtaining a spectrum from the object; obtaining a component generated based on a change in pressure applied to the object; correcting the spectrum based on the obtained pressure change-based component; as well as estimating biological information of the subject based on the corrected spectrum, Wherein, obtaining the component generated based on the pressure change comprises: performing principal component analysis (PCA) on the obtained spectra; and The components generated based on the pressure change are obtained based on the results of PCA, Wherein, obtaining the component generated based on the pressure change comprises: The components generated based on the pressure change are obtained based on the shapes of the components analyzed by using PCA, Wherein, obtaining the component generated based on the pressure change comprises: obtaining a component having a shape similar to the absorption peak of hemoglobin among the components analyzed by using PCA as the component generated based on the pressure change, or obtaining, from among the components analyzed by PCA, a component having a spectral shape similar to the spectral shape of a component defined in a component database as the component generated based on the pressure change, Wherein, correcting the spectrum comprises: Correcting the spectrum change caused by the pressure change by removing the component generated based on the pressure change from the acquired spectrum, Wherein, correcting the spectrum comprises: The obtained components were removed from the obtained spectrum by using the least square method.

12. The method according to claim 11, in, Obtaining the component generated based on the pressure change includes: obtaining, from the component database, the component generated based on the pressure change applied by the object to the spectral sensor.

13. The method according to claim 11, in, Obtaining the component generated based on the pressure change includes obtaining the component generated based on the pressure change based on the spectrum obtained during an initial pressure stage of the entire time period for obtaining the spectrum.

14. The method according to claim 11, further comprising: include: Obtain the noise-related component of the acquired spectrum, Wherein, correcting the spectrum includes: correcting the obtained spectrum according to the component generated based on the pressure change and the noise-related component.

15. The method according to claim 14, in, Obtaining the noise-related component includes obtaining the noise-related component based on a correlation between a principal component score of a principal component analyzed by performing principal component analysis on at least one of the obtained spectra and an estimated bio-information value estimated based on the at least one spectrum.

Images