MUSCLE-ACTIVITY DIAGNOSIS APPARATUS, METHOD, AND PROGRAM

• US 20110313310A1
• Filed: 06/03/2011
• Published: 12/22/2011
• Est. Priority Date: 06/16/2010
• Status: Active Grant

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Tomita, Takashi, Sano, Akane, Oba, Haruo
• Granted Patent: US 8,639,320 B2
• Time in Patent Office: Days
• US Class Current: 600/546
• CPC Class Codes: A61B 5/389 Electromyography [EMG]
• CPC Class Codes: A61B 5/4519 Muscles A61B5/389, A61B5/224 take precedence
• CPC Class Codes: A61B 5/7253 characterised by using transforms
• CPC Class Codes: A61B 5/7257 using Fourier transforms

Abstract

A muscle-activity diagnosis apparatus includes: an acquiring section acquiring a myoelectric signal from a test subject; using the myoelectric signal as an original signal, a transformed-signal generating section generating a transformed signal by performing Hilbert transformation and inverse Fourier transformation on the original signal; a phase-velocity calculation section calculating a phase velocity of the myoelectric signal on the basis of phases of the original signal and the transformed signal; and on the basis of a plurality of feature quantities of a waveform of the myoelectric signal in a unit time and the plurality of feature quantities including at least a size of amplitude of the myoelectric signal and the calculated phase velocity, a state-identifying section identifying an activity state of a muscle of a predetermined part of a body of the test subject for each of the unit time.

Claims

1. A muscle-activity diagnosis apparatus comprising:

an acquiring section acquiring a myoelectric signal from a test subject;

using the myoelectric signal as an original signal, a transformed-signal generating section generating a transformed signal by performing Hilbert transformation and inverse Fourier transformation on the original signal;

a phase-velocity calculating section calculating a phase velocity of the myoelectric signal on the basis of phases of the original signal and of the transformed signal; and

on the basis of a plurality of feature quantities of a waveform of the myoelectric signal in a unit time and the plurality of feature quantities including at least a size of amplitude of the myoelectric signal and the calculated phase velocity, a state-identifying section identifying an activity state of a muscle of a predetermined part of a body of the test subject for each of the unit time.

2. The muscle-activity diagnosis apparatus according to claim 1,

wherein the acquiring section differentially detects an electronic signal obtained from an earphone sensor inserted into an external auditory canal of the test subject, and digitize the signal so as to obtain the myoelectric signal.

3. The muscle-activity diagnosis apparatus according to claim 2,

wherein the state-identifying section identifies activity states of muscles of a plurality of parts of a face of the test subject, and outputs a signal indicating a facial expression of the test subject estimated on the basis of the identified activity states of the muscles of the plurality of parts.

4. The muscle-activity diagnosis apparatus according to claim 2,

wherein the earphone sensor is provided with

a myoelectric electrode for detecting an electronic signal generated by muscle activity of the test subject;

a voltage detection electrode for detecting a reference voltage being a difference between a potential in an ear of the test subject and a potential of a circuit in the earphone sensor; and

a feedback electrode for supplying a feedback voltage generated by inversely amplifying the reference voltage.

5. The muscle-activity diagnosis apparatus according to claim 1,

wherein using the plurality of feature quantities of the waveform of the myoelectric signal in the unit time provided in advance as parameters, the state-identifying section performs machine learning on the basis of a sample having a corresponding relationship between the parameter and the activity state of the muscle of the test subject, and

identifies the plurality of feature quantities of the myoelectric signal on the basis of a result of the machine learning so as to identify the activity state of the muscle of the predetermined part of the body of the test subject.

6. The muscle-activity diagnosis apparatus according to claim 5,

wherein AdaBoost is used as the identification method by the machine learning.

7. A method of diagnosing muscle activity, comprising:

an acquiring section acquiring a myoelectric signal from a test subject;

using the myoelectric signal as an original signal, a transformed-signal generating section generating a transformed signal by performing Hilbert transformation and inverse Fourier transformation on the original signal;

a phase-velocity calculating section calculating a phase velocity of the myoelectric signal on the basis of phases of the original signal and the transformed signal; and

on the basis of a plurality of feature quantities of a waveform of the myoelectric signal in a unit time and the plurality of feature quantities including at least a size of amplitude of the myoelectric signal and the calculated phase velocity, a state-identifying section identifying an activity state of a muscle of a predetermined part of a body of the test subject for each of the unit time.

8. A program for causing a computer to function as a muscle-activity diagnosis apparatus, the apparatus comprising:

an acquiring section acquiring a myoelectric signal from a test subject;

using the myoelectric signal as an original signal, a transformed-signal generating section generating a transformed signal by performing Hilbert transformation and inverse Fourier transformation on the original signal;

a phase-velocity calculating section calculating a phase velocity of the myoelectric signal on the basis of phases of the original signal and the transformed signal; and

on the basis of a plurality of feature quantities of a waveform of the myoelectric signal in a unit time and the plurality of feature quantities including at least a size of amplitude of the myoelectric signal and the calculated phase velocity, a state-identifying section identifying an activity state of a muscle of a predetermined part of a body of the test subject for each of the unit time.