Blood sugar level prediction device, measurement device, blood sugar level prediction method and computer-readable storage medium

• US 10,123,749 B2
• Filed: 10/30/2013
• Issued: 11/13/2018
• Est. Priority Date: 12/04/2012
• Status: Expired due to Fees

• Current Assignee: NEC Solution Innovators, Ltd (NEC Corporation)
• Original Assignee: NEC Solution Innovators, Ltd (NEC Corporation)
• Inventors: Kamimura, Ippei
• Primary Examiner(s): Winakur, Eric
• Assistant Examiner(s): Fardanesh, Marjan
• Application Number:
  US14/648,339
• Publication Number:
  US 20150297144A1
• Time in Patent Office: 1,840 Days
• Field of Search: None
• CPC Class Codes: A61B 2560/0238 Means for recording calibration data
• CPC Class Codes: A61B 5/14532 for measuring glucose, e.g. by tissue impedance measurement
• CPC Class Codes: A61B 5/1455 using optical sensors, e.g. spectral photometrical oximeters
• CPC Class Codes: A61B 5/4866 Evaluating metabolism A61B5/083 takes precedence
• CPC Class Codes: 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
• CPC Class Codes: A61B 5/7278 Artificial waveform generation or derivation, e.g. synthesising signals from measured signals A61B5/327 takes precedence
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

A blood sugar level prediction device (10) is for predicting the blood sugar level of a user (30). The blood sugar level prediction device (10) is provided with a pattern selection unit (11) that selects one pattern from a plurality of preset blood sugar level time series variation patterns, based on the blood sugar level of the user (30) at the time of fasting, and a correction processing unit (12) that corrects the selected pattern using a plurality of measured values obtained by executing optical measurement of blood sugar levels over a number of measurement dates, targeting the user (30).

Claims

1. A blood sugar level prediction device for predicting a blood sugar level of a user, the blood sugar level prediction device comprising a computer processor that realizes:

a pattern selection unit that selects one pattern from a plurality of preset blood sugar level time series variation patterns, based on a blood sugar level at a fasting time of the user; and

a correction processing unit that corrects the selected pattern using a plurality of measured values obtained by executing optical measurement of blood sugar levels over a number of measurement dates, targeting the user, wherein the computer processor further realizes;

a set condition determination unit that determines, after a set time period has elapsed since selection of the pattern, whether the measured values obtained by the optical measurement of blood sugar levels that targeted the blood of the user meet a set condition,

wherein if it is determined by the set condition determination unit that the obtained measured values do not meet the set condition, the pattern selection unit executes selection of the pattern again and the correction processing unit corrects the newly selected pattern.

2. The blood sugar level prediction device according to claim 1,

wherein the pattern selection unit selects the pattern using a blood sugar level after eating in addition to the blood sugar level at the fasting time of the user.

3. The blood sugar level prediction device according to claim 1,

wherein the plurality of patterns each include a set of functions that are set for every timeslot whose reference time is the fasting time of the user and that specify a relationship between elapsed time from the reference time and blood sugar level, and

the correction processing unit converts a measurement time of the optical measurement into an elapsed time from the reference time, derives, for each of the plurality of measured values, a square root of a residual error between the measured value and a value calculated by applying the converted elapsed time to the functions of the selected pattern, and corrects each of the functions of the selected time series variation pattern such that when the square roots are summed, the sum is minimized.

4. A measurement device that measures a blood sugar level of a user through optical measurement, comprising:

a measurement unit that executes the optical measurement;

a standard curve database that stores a standard curve for converting a measured value that is obtained by executing the optical measurement into a blood sugar level, for each size of corresponding blood sugar levels;

a control processing unit that calculates a blood sugar level by selecting one of the standard curves and applying the measured value obtained by executing the optical measurement to the selected standard curve; and

a blood sugar level prediction unit that predicts the blood sugar level of the user by selecting one pattern from a plurality of preset blood sugar level time series variation patterns, based on a blood sugar level at a fasting time of the user, and correcting the selected pattern using a plurality of measured values obtained by executing the optical measurement over a number of measurement dates, targeting the user,

wherein the control processing unit, in a case where selection and correction of the pattern have been executed by the blood sugar level prediction unit, predicts the blood sugar level using the corrected pattern and selects the standard curve based on the predicted blood sugar level,

and wherein a set condition determination unit determines, after a set time period has elapsed since selection of the pattern, whether the measured values obtained by the optical measurement of blood sugar levels that targeted the blood of the user meet a set condition,

wherein if it is determined by the set condition determination unit that the obtained measured values do not meet the set condition, the blood sugar level prediction unit executes selection of the pattern again and correction of the newly selected pattern.

5. A blood sugar level prediction method for predicting a blood sugar level of a user, comprising the steps of:

(a) selecting, by a computer, one pattern from a plurality of preset blood sugar level time series variation patterns, based on a blood sugar level at a fasting time of the user;

(b) correcting, by the computing, the selected pattern using a plurality of measured values obtained by executing optical measurement of blood sugar levels over a number of measurement dates, targeting the user, and

(c) determining, by the computer, after a set time period has elapsed since selection of the pattern, whether the measured values obtained by the optical measurement of blood sugar levels that targeted the blood of the user meet a set condition,

wherein if it is determined in the step (c) that the obtained measured values do not meet the set condition, the step (a) and the step (b) are executed again.

6. The blood sugar level prediction method according to claim 5,

wherein the step (a) comprises selecting the pattern using a blood sugar level after eating in addition to the blood sugar level at the fasting time of the user.

7. The blood sugar level prediction method according to claim 5,

wherein the plurality of patterns each include a set of functions that are set for every timeslot whose reference time is the fasting time of the user and that specify a relationship between elapsed time from the reference time and blood sugar level, and

the step (b) comprises converting, by the computer, a measurement time of the optical measurement into an elapsed time from the reference time, deriving, by the computer, for each of the plurality of measured values, a square root of a residual error between the measured value and a value calculated by applying the converted elapsed time to the functions of the selected pattern, and correcting, by the computer, each of the functions of the selected time series variation pattern such that when the square roots are summed, the sum is minimized.

8. A non-transitory computer-readable storage medium storing a computer program executed by a processor for predicting a blood sugar level of a user by computer, the processor executing the program to perform the steps of:

(a) selecting one pattern from a plurality of preset blood sugar level time series variation patterns, based on a blood sugar level at a fasting time of the user;

(b) correcting the selected pattern using a plurality of measured values obtained by executing optical measurement of blood sugar levels over a number of measurement dates, targeting the user; and

(c) determining, after a set time period has elapsed since selection of the pattern, whether the measured values obtained by the optical measurement of blood sugar levels that targeted the blood of the user meet a set condition, and

if it is determined in the step (c) that the obtained measured values do not meet the set condition, the step (a) and the step (b) are executed again.

9. The non-transitory computer-readable storage medium according to claim 8,

wherein the step (a) comprises selecting the pattern using a blood sugar level after eating in addition to the blood sugar level at the fasting time of the user.

10. The non-transitory computer-readable storage medium according to claim 8,

wherein the plurality of patterns each include a set of functions that are set for every timeslot whose reference time is the fasting time of the user and that specify a relationship between elapsed time from the reference time and blood sugar level, and

the step (b) comprises converting a measurement time of the optical measurement into an elapsed time from the reference time, deriving, for each of the plurality of measured values, a square root of a residual error between the measured value and a value calculated by applying the converted elapsed time to the functions of the selected pattern, and correcting each of the functions of the selected time series variation pattern such that when the square roots are summed, the sum is minimized.