SYSTEM AND METHOD FOR DETECTING AN EPILEPTIC SEIZURE IN A PRONE EPILEPTIC PERSON

• US 20120165705A1
• Filed: 08/19/2009
• Published: 06/28/2012
• Est. Priority Date: 08/19/2009
• Status: Active Grant

• Current Assignee: Commissariat a L'Energie Atomique (French Alternative Energies and Atomic Energy Commission), Movea, Move Incorporated (News Corporation)
• Original Assignee: Movea
• Inventors: Jallon, Pierre, Bonnet, Stphane
• Granted Patent: US 9,161,711 B2
• Time in Patent Office: Days
• US Class Current: 600/595
• CPC Class Codes: A61B 2562/0219 Inertial sensors, e.g. accelerometers, gyroscopes, tilt switches
• CPC Class Codes: A61B 5/1118 Determining activity level
• CPC Class Codes: A61B 5/4094 Diagnosing or monitoring seizure diseases, e.g. epilepsy
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: G06F 18/295 Markov models or related models, e.g. semi-Markov models; Markov random fields; Networks embedding Markov models
• CPC Class Codes: G06F 2218/12 Classification; Matching

Abstract

A system for detecting an epileptic seizure in a prone person, comprising:

• • at least one motion sensor with at least one measurement axis having fastening means for securing said motion sensor to said person;
  • first means for determining a first probability of at least a first state transition diagram of the nocturnal activity of a prone person with respect to data representing the measurement signals of the motion sensor, said first diagram comprising predetermined probabilities of oriented transitions between two different or identical states, the probabilities of the states of said first diagram being predetermined; and
  • second means for determining a second probability of at least a second state transition diagram for an epileptic seizure with respect to data representing the measurement signals of the motion sensor, said second diagram comprising predetermined probabilities of oriented transitions between two different or identical states, the probabilities of the states of said second diagram being predetermined.

Claims

1. A system for detecting an epileptic seizure in a prone person comprising:

at least one motion sensor sensibly fastened to said person with at least one measurement axis;

a first determination module for determining a first probability of at least a first state transition diagram of the nocturnal activity of a prone person with respect to data representing the measurement signals of the motion sensor, said first diagram comprising predetermined probabilities of oriented transitions between two different or identical states, the probabilities of the states of said first diagram being predetermined;

a second determination module for determining a second probability of at least a second state transition diagram for an epileptic seizure with respect to data representing the measurement signals of the motion sensor, said second diagram comprising predetermined probabilities of oriented transitions between two different or identical states, the probabilities of the states of said second diagram being predetermined;

an association module for associating a state of said person as a function of the probabilities of the measurement signals of the motion sensor;

a first for calculation module for calculating the relations (φ

₁, φ

₁) between the first probability and the second probability; and

a detection module for detecting an epileptic seizure when at least one of said calculated relations (φ

₁, φ

₁) is below a threshold (λ

₁, λ

₂).

2. The system of claim 1, wherein at least one of said state transition diagrams is adapted to use a hidden Markov model.

3. The system of claim 2, further comprising:

a filter for selecting, for each measurement axis of the motion sensor, high frequencies above a first threshold, and low frequencies below a second threshold which is lower than or equal to said first threshold;

a second calculation module for calculating a first value x equal to a linear combination of the respective variations along each measurement axis, between two successive time intervals, of said low frequencies per time interval n;

a third calculation module for calculating a second value y equal to the mean of the energies, along each measurement axis, of said high frequencies;

a third determination module for determining the probability P_LF(x) of said first value x defined by a normal centered Gaussian distribution; and

a fourth determination module for determining the probability P_HF(x) of said second value y defined by a Chi 2 distribution with a degree of freedom equal to the number of measurement axes of the motion sensor taken into consideration;

said of association module being adapted to use the probabilities of said low and high frequencies.

4. The system of claim 3, wherein the probability density P(x(n),y(n)) of obtaining a pair of values (x(n), y(n)) for the low frequency component and the high frequency component being equal to the product of the probability density P_LF(x) of obtaining the value x(n) for the low frequency component and the probability density P_HF(x) of obtaining the value y(n) for the high frequency component, said probability densities P_LF(x) and P_HF(x) being defined by the following expressions:

5. The system of claim 2, wherein said hidden Markov model comprises not more than five states chosen from a group comprising a rest activity, a slight agitation activity, a tremor activity, an agitation activity, and a strong agitation activity.

6. The system of claim 2, wherein the probabilities of the states of said state transition diagrams are defined by the following relation:

7. The system of claim 1, wherein said second determination module is adapted to determine a second probability of a second state transition diagram for an epileptic seizure with clonic manifestations and of a second state transition diagram for an epileptic seizure with tonic manifestations.

8. The system of claim 1, wherein the states and the probabilities of the states are identical for the first state transition diagram and for the second state transition diagrams.

9. The system as claimed of claim 6, wherein there are 18 of said pairs (σ

_x⁽ⁱ⁾, o_y⁽ⁱ⁾), which are obtained by the combination of the following values;

σ

_x[0]=5×

10<sup>−

3</sup>, σ

_x[1]=1.8×

10<sup>−

2</sup>, σ

_x[2]=3.5×

10<sup>−

2</sup>, σ

_x[3]=5×

0.510<sup>−

2</sup>, σ

_x[4]=8×

10<sup>−

2</sup>, σ

_x[5]=1×

10<sup>−

1</sup>, and σ

_y[0]=1×

10<sup>−

2</sup>, σ

_y[1]=3×

10<sup>−

2</sup>, σ

_y[2]=8×

10<sup>−

2</sup>.

10. The system of claim 9, wherein said coefficients α

_i,p are defined as follows, for a pair (σ

_x^(p)[m], σ

_y^(p)[n]), where p is an index such that p=m+6n, varying from 0 to 17 for the 18 pairs (σ

_x^(p), σ

_y^(p));

11. The system of claim 10, wherein a first state transition diagram of general nocturnal activity is defined by the following matrix:

12. The system of claim 10, wherein a second state transition diagram for an epileptic seizure with clonic manifestations is defined by the following matrix:

13. The system of claim 10, wherein a second state transition diagram for an epileptic seizure with tonic manifestations is defined by the following matrix:

14. The system of claim 1, further comprising an alerting module adapted for providing a warning of the detection of an epileptic seizure.

15. The system of claim 1, wherein said motion sensor comprises an accelerometer, a magnetometer or a gyrometer.

16. A method for detecting an epileptic seizure in a prone person, wherein:

a first probability of at least a first state transition diagram of the nocturnal activity of a prone person is determined, with respect to the measurement signals of a motion sensor sensibly fastened to said person with at least one measurement axis, said first diagram comprising predetermined probabilities of oriented transitions between two different or identical states, the probabilities of the states of said first diagram being predetermined;

a second probability of at least a second state transition diagram for an epileptic seizure is determined, with respect to the measurement signals of the motion sensor, where said second diagram comprises predetermined probabilities of oriented transitions between two different or identical states, the probabilities of the states of said second diagram being predetermined;

a state of said person is associated as a function of the probabilities of the measurement signals of the motion sensor;

relations (φ

₁, φ

₁) of the first probability and the second probability are calculated; and

an epileptic seizure is detected when at least one of said calculated relations (φ

₁, φ

₁) is below a threshold (λ

₁, λ

₂).