# US 2018/0020958 A1 · Apparatus and method for estimating blood sugar based on heterogeneous spectrums > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2018/0020958 A1 - Document ID: US20180020958A1 - Status: Published application - Kind: A1 - Application number: 15/417,408 - Filed: 2017-01-27 - Published: 2018-01-25 - Earliest claimed: 2016-07-19 - Assignees: Samsung Electronics Co., Ltd.; Korea University Research and Business Foundation - Applicants: Samsung Electronics Co., Ltd.; Korea University Research and Business Foundation - Inventors: Woo Chang LEE; Young Hoon KIM; Jin Young PARK; Eui Seok SHIN; Woo Sik CHOI; Seoung Bum KIM - Web page: https://patentgrep.com/patent/US20180020958A1 ## CPC - A61B 5/14532: {for measuring glucose, e.g. by tissue impedance measurement} - A61B 5/0075: {by spectroscopy, i.e. measuring spectra, e.g. Raman spectroscopy, infrared absorption spectroscopy (A61B5/0071 takes precedence)} - A61B 5/1455: using optical sensors, e.g. spectral photometrical oximeters - A61B 5/7267: {involving training the classification device} - A61B 5/7275: {Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor} - A61B 5/7278: {Artificial waveform generation or derivation, e.g. synthesizing signals from measured signals} - G16H 50/50: for simulation or modelling of medical disorders - A61B 5/443: {Evaluating skin constituents, e.g. elastin, melanin, water} - A61B 5/4884: {inducing physiological or psychological stress, e.g. applications for stress testing} - A61B 2560/0475: Special features of memory means, e.g. removable memory cards - A61B 2562/0238: Optical sensor arrangements for performing transmission measurements on body tissue - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA ## Abstract An apparatus and a method for determining blood sugar based on heterogeneous spectrums are provided. The blood sugar model generation apparatus includes a data obtainer configured to obtain blood sugar profile data of a user based on a first type spectrum-blood sugar profile relationship model, a second type spectrum measurer configured to measure training second type spectrum data for the skin of the user, and a processor configured to generate an individualized blood sugar model based on the obtained blood sugar profile data and the measured training second type spectrum data. ## Claims ### Claim 1 A blood sugar model generation apparatus comprising: a data obtainer configured to calculate blood sugar profile data of a user based on a first type spectrum-blood sugar profile relationship model; a second type spectrum measurer configured to measure training second type spectrum data for a skin of the user; a processor configured to generate an individualized blood sugar model based on the calculated blood sugar profile data and the measured training second type spectrum data; and an outputter configured to output the generated individualized blood sugar model. ### Claim 2 (depends on 1) The blood sugar model generation apparatus of claim 1, wherein a first type spectrum is a Raman spectrum, and a second type spectrum is a near infrared (NIR) spectrum. ### Claim 3 (depends on 1) The blood sugar model generation apparatus of claim 1, further comprising a first type spectrum-blood sugar profile relationship model generator configured to generate the first type spectrum-blood sugar profile relationship model through machine learning based on training first type spectrum data and training the blood sugar profile data. ### Claim 4 (depends on 3) The blood sugar model generation apparatus of claim 3, wherein the training blood sugar profile data is obtained by performing an oral glucose tolerance test (OGTT) on a subject whose said training first type spectrum data has been measured. ### Claim 5 (depends on 3) The blood sugar model generation apparatus of claim 3, wherein the first type spectrum-blood sugar profile relationship model generator is configured to generate the first type spectrum-blood sugar profile relationship model through machine learning with the training of first type spectrum data as an input and the training blood sugar profile data as a target. ### Claim 6 (depends on 3) The blood sugar model generation apparatus of claim 3, wherein the first type spectrum-blood sugar profile relationship model generator is configured to generate the first type spectrum-blood sugar profile relationship model through machine learning with the training first type spectrum data as an input and an area under the curve (AUC) value calculated from the training blood sugar profile data as a target. ### Claim 7 (depends on 3) The blood sugar model generation apparatus of claim 3, wherein the machine learning comprises a machine learning algorithm selected from among partial least squares regression, linear regression, neural network, decision tree, genetic algorithm, genetic programming, K-nearest neighbor, radial basis function network, random, forest, support vector machine, and deep-learning. ### Claim 8 (depends on 1) The blood sugar model generation apparatus of claim 1, wherein the second type spectrum measurer comprises: a light source configured to emit light to the skin of the user; and a spectroscope configured to detect absorbed, scattered, or reflected light from the skin of the user and measure the training second type spectrum data based on the detected light. ### Claim 9 (depends on 1) The blood sugar model generation apparatus of claim 1, wherein the processor is further configured to generate the individualized blood sugar model through machine learning. ### Claim 10 (depends on 9) The blood sugar model generation apparatus of claim 9, wherein the processor is further configured to generate the individualized blood sugar model by calculating a blood sugar value corresponding to the measured training second type spectrum data from the calculated blood sugar profile data and through the machine learning with the training second type spectrum data as an input and the calculated blood sugar value as a target. ### Claim 11 A blood sugar profile providing apparatus comprising: a memory configured to store a first type spectrum-blood sugar profile relationship model; a first type spectrum measurer configured to measure first type spectrum data for a skin of a user; a processor configured to calculate a blood sugar profile of the user based on the first type spectrum-blood sugar profile relationship model and the measured first type spectrum data; and an outputter configured to output the calculated blood sugar profile of the user. ### Claim 12 (depends on 11) The blood sugar profile providing apparatus of claim 11, wherein a first type spectrum is a Raman spectrum. ### Claim 13 (depends on 11) The blood sugar profile providing apparatus of claim 11, further comprising a first type spectrum-blood sugar profile relationship model generator configured to generate the first type spectrum-blood sugar profile relationship model through machine learning based on training the first type spectrum data and training blood sugar profile data. ### Claim 14 (depends on 13) The blood sugar profile providing apparatus of claim 13, wherein the training blood sugar profile data is obtained by performing an oral glucose tolerance test (OGTT) on a subject whose said training of the first type spectrum data has been measured. ### Claim 15 (depends on 13) The blood sugar profile providing apparatus of claim 13, wherein the first type spectrum-blood sugar profile relationship model generator is further configured to generate the first type spectrum-blood sugar profile relationship model through machine learning with the training first type spectrum data as an input and the training blood sugar profile data as a target. ### Claim 16 (depends on 13) The blood sugar profile providing apparatus of claim 13, wherein the first type spectrum-blood sugar profile relationship model generator is further configured to generate the first type spectrum-blood sugar profile relationship model through machine learning with the training first type spectrum data as an input and an area under the curve (AUC) value calculated from the training blood sugar profile data as a target. ### Claim 17 (depends on 13) The blood sugar profile providing apparatus of claim 13, wherein the machine learning comprises a machine learning algorithm selected from among partial least squares regression, linear regression, neural network, decision tree, genetic algorithm, genetic programming, K-nearest neighbor, radial basis function network, random, forest, support vector machine, and deep-learning. ### Claim 18 (depends on 11) The blood sugar profile providing apparatus of claim 11, wherein the first type spectrum measurer comprises: a light source configured to emit light to the skin of the user; and a spectroscope configured to detect absorbed, scattered, or reflected light from the skin of the user and to measure the first type spectrum data based on the detected light. ### Claim 19 (depends on 11) The blood sugar profile providing apparatus of claim 11, further comprising: a first type spectrum-blood sugar profile relationship model generator configured to generate the first type spectrum-blood sugar profile relationship model through machine learning with training the first type spectrum data as an input and training blood sugar profile data as a target, wherein the processor is further configured to calculate the blood sugar profile of the user by a blood profile output by inputting the measured first type spectrum data into the first type spectrum-blood sugar profile relationship model. ### Claim 20 (depends on 11) The blood sugar profile providing apparatus of claim 11, further comprising a first type spectrum-blood sugar profile relationship model generator configured to generate the first type spectrum-blood sugar profile relationship model through machine learning with training first type spectrum data as an input and an area under the curve (AUC) value calculated from training blood sugar profile data as a target, and wherein the processor calculates the blood sugar profile of the user based on the AUC value output by inputting the measured first type spectrum data into the first type spectrum-blood sugar profile relationship model. ### Claim 21 (depends on 20) The blood sugar profile providing apparatus of claim 20, wherein the processor adjusts a reference blood sugar profile to allow the AUC value of the reference blood sugar profile to be the output AUC value and calculates a blood sugar profile data of the user by the data of the adjusted reference blood sugar profile. ### Claim 22 An apparatus comprising: a memory configured to store an individualized blood sugar estimation model; a second type spectrum measurer configured to measure second type spectrum data for a skin of a user; and a processor configured to calculate blood sugar of the user based on the measured second type spectrum data and the individualized blood sugar model, wherein the blood sugar model is generated based on blood sugar profile data of the user calculated based on a first type spectrum-blood sugar profile relationship model and training second type spectrum data for the skin of the user. ### Claim 23 (depends on 22) The apparatus of claim 22, wherein a first type spectrum is a Raman spectrum, and a second type spectrum is a near infrared (NIR) spectrum. ### Claim 24 (depends on 22) The apparatus of claim 22, wherein the first type spectrum-blood sugar profile relationship model is generated through machine learning based on training first type spectrum data and training the blood sugar profile data. ### Claim 25 (depends on 24) The apparatus of claim 24, wherein the training blood sugar profile data is obtained by performing an oral glucose tolerance test (OGTT) on a subject whose training the first type spectrum data has been measured. ### Claim 26 (depends on 24) The apparatus of claim 24, wherein the first type spectrum-blood sugar profile relationship model is generated through machine learning with the training first type spectrum data as an input and the training blood sugar profile data as a target. ### Claim 27 (depends on 24) The apparatus of claim 24, wherein the first type spectrum-blood sugar profile relationship model is generated through machine learning with the training first type spectrum data as an input and an area under curve (AUC) value calculated from the training blood sugar profile data as a target. ### Claim 28 (depends on 24) The apparatus of claim 24, wherein the machine learning comprises a machine learning algorithm selected from among partial least squares regression, linear regression, neural network, decision tree, genetic algorithm, genetic programming, K-nearest neighbor, radial basis function network, random, forest, support vector machine, and deep-learning. ### Claim 29 (depends on 21) The apparatus of claim 21, wherein the second type spectrum measurer comprises: a light source configured to emit light to the skin of the user; and a spectroscope configured to detect absorbed, scattered, or reflected light from the skin of the user and measures the second type spectrum data based on the detected light. ### Claim 30 (depends on 21) The apparatus of claim 21, wherein the individualized blood sugar model is generated through machine learning.