Method and apparatus for predicting in vivo analyte concentration using learning and a net analyte signal
A method for predicting a concentration of an in vivo analyte includes obtaining a plurality of in vivo spectra of the in vivo analyte, determining a learning section for a concentration predicting algorithm for the analyte based on an unchanged section, during which the concentration of the analyte is not substantially changed, and a plurality of the in vivo spectra, and predicting the concentration of the in vivo analyte by using the concentration predicting algorithm based on a learned result of the learning section and an intrinsic spectrum of the in vivo analyte.
1. A method for predicting a concentration of an in vivo analyte, the method comprising:
obtaining a plurality of in vivo spectra of the in vivo analyte;
determining a learning section of a concentration predicting algorithm for the in vivo analyte based on an unchanged section, during which a concentration of the in vivo analyte is not substantially changed, and the in vivo spectra; and
predicting the concentration of the in vivo analyte by using the concentration predicting algorithm based on a learned result of the learning section and an intrinsic spectrum of the in vivo analyte.
2. The method of claim 1 , wherein the in vivo analyte is at least one of glucose, urea, lactate, triglyceride, protein, cholesterol, and ethanol.
3. The method of claim 1 , wherein,
the in vivo analyte is glucose, and
the unchanged section, during which the concentration of the in vivo analyte is not substantially changed, is a fasting section.
4. The method of claim 1 , wherein the in vivo spectrum comprise at least one of an absorption spectrum or a reflection spectrum of an infra-red ray.
5. The method of claim 1 , wherein the in vivo spectrum comprise a dispersion spectrum of a single wavelength electromagnetic wave.
6. The method of claim 1 , wherein the obtaining the in vivo spectra of the in vivo analyte comprises obtaining the in vivo spectra continually at a predetermined time interval.
7. The method of claim 1 , wherein the concentration predicting algorithm comprises a net analyte signal algorithm.
8. The method of claim 1 , wherein the determining the learning section comprises:
calculating a similarity between the in vivo spectra;
determining a section having a high similarity as a similar section; and
determining a section, during which the unchanged section and the similar section overlap each other, as the learning section.
9. The method of claim 8 , wherein the calculating the similarity between the in vivo spectra comprises:
aligning baselines of at least two spectra for calculating similarities thereof among the in vivo spectra; and
calculating a difference between the at least two in vivo spectra, the baselines of which are aligned.
10. The method of claim 1 , wherein the predicting the concentration of the in vivo analyte comprises predicting the concentration of the in vivo analyte in the similar section including the learning section when a length of the learning section is longer than a predetermined section length.
11. The method of claim 1 , wherein the predicting the concentration of the in vivo analyte comprises re-determining the learning section in the similar section when a length of the learning section is shorter than a predetermined length.
12. The method of claim 1 , wherein the predicting the concentration of the in vivo analyte comprises displaying a message to inform a user that a concentration prediction is unavailable when a length of the learning section is shorter than a predetermined length.
13. The method of claim 1 , wherein the in vivo analyte is included in a human body, an animal, a mammal, a non-mammal or a microorganism.
14. An apparatus for predicting a concentration of an in vivo analyte, the apparatus comprising:
a processor; and
a memory,
wherein the processor executes a program stored in the memory to perform:
obtaining a plurality of in vivo spectra of the in vivo analyte;
determining a learning section for predicting the concentration of the in vivo analyte using a net analyte signal based on an unchanged section, and based on the in vivo spectra, wherein the unchanged section is a time duration during which a concentration of the analyte is not substantially changed, and wherein the determining the learning section comprises:
calculating similarities between the in vivo spectra;
determining a section, which has a predetermined similarity, based on the calculated similarities between the in vivo spectra as a similar section; and
determining a section, during which the unchanged section and the similar section overlap each other, as the learning section; and
predicting the concentration of the in vivo analyte by using the net analyte signal and a spectrum changing factor, which factor is not a function of a change in the concentration of the in vivo analyte, during the learning section and by using an intrinsic spectrum of the in vivo analyte, wherein the intrinsic spectrum is a spectrum of the analyte itself obtained from a sample comprised of the analyte dissolved in an aqueous buffer solution.
15. The apparatus of claim 14 , wherein the in vivo analyte is at least one of glucose, urea, lactate, triglyceride, protein, cholesterol, and ethanol.
16. The apparatus of claim 14 , wherein
the in vivo analyte is glucose, and
the unchanged section, during which the concentration of the in vivo analyte is not substantially changed, is a fasting section.
17. The apparatus of claim 14 , wherein each of the in vivo spectra comprises at least one of an absorption spectrum and a reflection spectrum of an infra-red ray.
18. The apparatus of claim 14 , wherein each of the in vivo spectra is obtained by a Raman spectroscopy.
19. The apparatus of claim 14 , wherein when the processor performs the obtaining the in vivo spectra of the in vivo analyte, the processor performs:
obtaining the in vivo spectra repeatedly at a predetermined time interval.
20. The apparatus of claim 14 , wherein, when the processor performs the calculating the similarities between the in vivo spectra, the processor performs:
aligning baselines of at least two in vivo spectra for calculating similarities thereof among the in vivo spectra;
calculating a difference between the at least two in vivo spectra by aligning baselines of the at least two in vivo spectra; and
calculating the similarities between the at least two in vivo spectra based on the difference between the at least two in vivo spectra.
21. The apparatus of claim 14 , wherein, when the processor performs the predicting the concentration of the in vivo analyte, the processor performs:
predicting the concentration of the in vivo analyte in the similar section including the learning section, in a case that a length of the learning section is longer than a predetermined section length.
22. The apparatus of claim 14 , wherein, when the processor performs the predicting the concentration of the in vivo analyte, the processor performs:
re-determining the learning section in the similar section, in a case that a length of the learning section is shorter than a predetermined length.
23. The apparatus of claim 14 , wherein, when the processor performs the predicting the concentration of the in vivo analyte, the processor performs:
displaying a message to inform a user that a concentration prediction is unavailable, when a length of the learning section is shorter than a predetermined length.
24. The apparatus of claim 14 , wherein the in vivo analyte is included in a human body, an animal, a mammal, a non-mammal, or a microorganism.
25. The apparatus of claim 14 further comprising:
a communicator which receives the in vivo spectra from an infra-red sensor or a laser sensor through a wired or wireless network.
26. The apparatus of claim 14 further comprising:
an infra-red sensor which generates the in vivo spectra by radiating an infra-red ray to a human body.
27. The apparatus of claim 14 further comprising: a laser sensor which generates the in vivo spectra by radiating laser light to a human body.