Waveform generation identification method and computer-readable medium

Patent metadata

Field	Value
Publication number	US-11694117-B2
Application number	US-202017083350-A
Country	US
Kind code	B2
Filing date	Oct 29, 2020
Priority date	Oct 30, 2019
Publication date	Jul 4, 2023
Grant date	Jul 4, 2023

Abstract

A waveform generation identification method includes: comparing individual waveform data obtained by a plurality of sensors, with at least one piece of characteristic waveform information; determining appearance probability of characteristic waveform information in at least a certain section of the waveform data, based on a degree of correlation between a peak section of the waveform data and the characteristic waveform information; and identifying a time when a section matching with the characteristic waveform information appears and a concerned sensor, based on the appearance probability.

First claim

What is claimed is:

1. A waveform generation identification method comprising:
comparing individual magneto-encephalograph waveform data obtained by each of a plurality of sensors, with at least one piece of characteristic waveform information;
determining appearance probability of characteristic waveform information in at least a certain section of the waveform data, based on a degree of correlation between a section of the waveform data and the characteristic waveform information; and
identifying a time when the certain section matching with the characteristic waveform information appears and identifying a concerned sensor from among the plurality of sensors, based on the appearance probability.

2. The waveform generation identification method according to claim 1 , wherein the determining includes displaying a two-dimensional map representing the appearance probability of the characteristic waveform information in at least the certain section of the waveform data.

3. The waveform generation identification method according to claim 2 , wherein the determining includes displaying the two-dimensional map to be superimposed on the individual waveform data obtained by the plurality of sensors.

4. The waveform generation identification method according to claim 1 , wherein the determining includes calculating a probability map of the characteristic waveform information using a machine learning model having learned in advance.

5. The waveform generation identification method according to claim 4 , wherein the determining includes causing learning to be performed using, as correct answer data, information about a point of time of characteristic waveform information and one of the plurality of sensors that are ideal in an equivalent current dipole method, in machine learning.

6. The waveform generation identification method according to claim 4 , wherein the identifying includes defining groups of sensors in advance, and expanding a number of sensors having probability equal to or greater than a predetermined value in the probability map into all sensors of a group to which the number of sensors belong, to narrow down the plurality of sensors.

7. The waveform generation identification method according to claim 2 , wherein the determining includes performing color display of the two-dimensional map in which color changes according to the appearance probability of the characteristic waveform information.

8. A non-transitory computer-readable medium including programmed instructions that cause a computer to execute:
comparing individual magneto-encephalograph waveform data obtained by each of a plurality of sensors, with at least one piece of characteristic waveform information;
determining, appearance probability of characteristic waveform information in at least a certain section of the waveform data, based on a degree of correlation between a section of the waveform data and the characteristic waveform information; and
identifying a time when the certain section matching with the characteristic waveform information appears, and identifying a concerned sensor from among the plurality of sensors, based on the appearance probability.

9. The computer-readable medium according to claim 8 , wherein the determining includes displaying a two-dimensional map representing the appearance probability of the characteristic waveform information in at least the certain section of the waveform data.

10. The computer-readable medium according to claim 9 , wherein the determining includes displaying the two-dimensional map to be superimposed on individual waveform data obtained by the plurality of sensors.

11. The computer-readable medium according to claim 8 , wherein the determining includes calculating a probability map of the characteristic waveform information using a machine learning model having learned in advance.

12. The computer-readable medium according to claim 11 , wherein the determining includes causing learning to be performed using, as correct answer data, information about a point of time of characteristic waveform information and one of the plurality of sensors that are ideal in an equivalent current dipole method, in machine learning.

13. The computer-readable medium according to claim 11 , wherein the identifying includes defining groups of sensors in advance, and expanding a number of sensors having probability equal to or greater than a predetermined value in the probability map into all sensors of a group to which the number of sensors belong, to narrow down the plurality of sensors.

14. The computer-readable medium according to claim 9 , wherein the determining includes performing color display of the two-dimensional map in which color changes according to the appearance probability of the characteristic waveform information.

Assignees

• Hirano Ryoji
• Hirata Masayuki
• Nakata Otoichi
• Ricoh Co Ltd

Inventors

• Hirano Ryoji
• Hirata Masayuki
• Nakata Otoichi

Classifications

• G06N20/00 Primary

  Machine learning · CPC title

• G06F2218/14

  by matching peak patterns · CPC title

• G06F2218/12

  Classification; Matching · CPC title

• A61B5/4094 Primary

  Diagnosing or monitoring seizure diseases, e.g. epilepsy · CPC title

• A61B5/72

  Signal processing specially adapted for physiological signals or for diagnostic purposes · CPC title