System and method for determining the posture of a person

• US 9,445,752 B2
• Filed: 04/26/2010
• Issued: 09/20/2016
• Est. Priority Date: 04/24/2009
• Status: Active Grant

• Current Assignee: Commissariat a L'Energie Atomique (French Alternative Energies and Atomic Energy Commission), Movea
• Original Assignee: Commissariat a L'Energie Atomique (French Alternative Energies and Atomic Energy Commission), Movea
• Inventors: Jallon, Pierre
• Primary Examiner(s): Dvorak, Linda
• Assistant Examiner(s): LLOYD, EMILY M
• Application Number:
  US13/264,965
• Publication Number:
  US 20120143094A1
• Time in Patent Office: 2,339 Days
• Field of Search: 600/595, 340/573.7
• US Class Current: 1/1
• CPC Class Codes: A61B 5/1116 Determining posture transitions
• CPC Class Codes: A61B 5/1118 Determining activity level
• CPC Class Codes: A61B 5/1123 Discriminating type of movement, e.g. walking or running A61B5/1116, A61B5/112 take precedence
• CPC Class Codes: A61B 5/4094 Diagnosing or monitoring seizure diseases, e.g. epilepsy
• CPC Class Codes: G06F 2218/00 Aspects of pattern recognition specially adapted for signal processing
• CPC Class Codes: G06V 40/20 Movements or behaviour, e.g. gesture recognition recognition of facial expressions G06V40/16

Abstract

A system for determining the posture of a person has a motion sensor (CM) with at least one axis of measurement, which is provided with fixing means (MF) for rigidly connecting said motion sensor (CM) to a user. Analysis means (AN) are also included for determining a posture of the user. The analysis means (AN) utilize: (A) joint densities of probabilities of a low-frequency component and a high-frequency component, these densities of probabilities being defined for each posture; and (B) probabilities of transition between two successive postures.

Claims

1. A system for determining a posture of a person, comprising:

at least one motion sensor configured to be affixed to the person, the at least one motion sensor configured to measure k measurement axes, k being greater than or equal to two;

at least one filter coupled to the at least one motion sensor and configured to generate output signals that include first signals with frequencies higher than a first threshold selected for each axis of measurement, and second signals with frequencies below a second threshold lower than or equal to said first threshold;

a calculation module configured to calculate a density of probability of a first variable representing said first signals, said first variable modeled by a Chi-2 law with a degree of freedom equal to k, and to calculate a density of probability of a second variable representing said second signals, said second variable being a one-dimensional variable representative of the second signals along the k axes of measurement and being modeled by a Gaussian law; and

an analysis module configured to determine a posture of the person out of a plurality of postures by combining;

joint densities of probabilities of said first and second variables, the joint densities of probabilities being defined for each posture of the plurality of postures; and

probabilities of transition between two successive postures of the plurality of postures.

2. The system of claim 1, wherein the calculation module is further configured to calculate the joint densities of probabilities (P(x(n),y(n))) defined at values of the second variable and the first variable by computing the product of a density of probability P_x,i of obtaining a value (x(n)) for the second variable and a density of probability P_y,i of obtaining a value (y(n)) for the first variable, the density of probability P_x,i of obtaining the value (x(n)) and the density of probability P_y,i of obtaining the value (y(n)) being defined for each state i by the following expressions:

3. The system of claim 2, wherein said analysis module is configured for determining a posture of the person as a function of time, by using the hidden Markov model with four states including (1) a standing or seated posture, (2) a walking posture, (3) a leaning posture and (4) a recumbent posture.

4. The system of claim 3, wherein the system is further configured to determine the posture of the person according to the following:

for the standing or seated posture, μ

_x,1 ε

[0.7;

1.3], σ

_x,1 ε

[0.05;

0 4], σ

_y,1 ε

[1×

10<sup>−

3</sup>;

5×

10<sup>−

1</sup>];

for the walking posture, μ

_x,2 ε

[0.7;

1.3], σ

_x,2 ε

[0.05;

0.4], σ

_y,2 ε

[1×

10<sup>−

2</sup>;

1];

for the leaning posture, μ

_x,3 ε

[0.7;

1.3], σ

_x,3 ε

[0.05;

0.4], σ

_y,3 ε

[1×

10<sup>−

3</sup>;

5×

10<sup>−

1</sup>]; and

for the recumbent posture μ

_x,4 ε

[−

0.3;

0.3], σ

_x,4 ε

[0.05;

0.4], σ

_y,4 ε

[1×

10<sup>−

3</sup>;

5×

10<sup>−

1</sup>].

5. The system of claim 1, further comprising a display.

6. The system of claim 1, wherein the at least one motion sensor comprises an accelerometer, a magnetometer, a gyrometer or combinations thereof.

7. The system of claim 1, wherein the at least one motion sensor comprises a first accelerometer with a measurement axis and fixing element configured for fixing the first accelerometer to a torso of the person in such a way that the measurement axis coincides substantially with a vertical axis of a body when the person is upright.

8. The system of claim 7, wherein the at least one motion sensor comprises a second accelerometer with an axis of measurement and a fixing element configured for fixing the second accelerometer to a thigh of the person in such a way that the axis of measurement of the second accelerometer coincides with the vertical axis of the body when the person is upright.

9. The system of claim 8, wherein said analysis module is configured for determining a posture of the person as a function of time, by using a hidden Markov model, the hidden Markov model including at least two states selected from four states including (1) a standing posture, (2) a seated posture, (3) a recumbent posture, and (4) a walking posture.

10. The system as claimed in claim 9, wherein x(n) represents a vector comprising a pair of values of respective second variables of said two accelerometers, and y(n) represents a value of the first variable of said second accelerometer, at sample index n, and the system is further configured to compute the density of probability of obtaining the values x(n) according to the following expression:

11. The system of claim 10, wherein the system is further configured to determine the posture of the person according to the following:

for the standing posture, μ

_x,1=[1 1]^T,

12. A method for determining a posture of a person, comprising:

processing output signals of at least one motion sensor to produce first signals with frequencies above a first threshold selected for each axis of measurement, and second signals with frequencies below a second threshold lower than or equal to said first threshold, wherein the at least one motion sensor is configured to measure k measurement axes, k being equal to or greater than two;

calculating a density of probability of a first variable representative of said first signals and a density of probability of a second variable representative of said second signals, said first variable being modeled by a Chi-2 law with a degree of freedom equal to k, and said second variable being a one-dimensional variable representative of the second signals along the k axes of measurement and being modeled by a Gaussian law; and

determining a posture of the person as a function of time, by using a hidden Markov model with N states corresponding respectively to N postures, the determination being carried out by utilizing;

joint densities of probabilities of said second and first variables, the joint densities of probabilities being defined for each posture of the N postures, and

probabilities of transition between two successive postures of the N postures.