Snore detecting method and apparatus

• US 20030066529A1
• Filed: 10/07/2002
• Published: 04/10/2003
• Est. Priority Date: 10/10/2001
• Status: Active Grant

• Current Assignee: Philips RS North America LLC (Koninklijke Philips N.V.)
• Original Assignee: RIC INVESTMENTS LLC (Koninklijke Philips N.V.)
• Inventors: Truschel, William A., Duff, Winslow K., Shissler, Andrew, Wheeler, Robert
• Granted Patent: US 7,246,619 B2
• Time in Patent Office: Days
• US Class Current: 128/204.18
• CPC Class Codes: A61B 5/087 Measuring breath flow
• CPC Class Codes: A61B 5/4818 Sleep apnoea
• CPC Class Codes: A61B 7/003 Detecting lung or respiration noise
• CPC Class Codes: A61M 16/0063 Compressors
• CPC Class Codes: A61M 16/024 including calculation means, e.g. using a processor
• CPC Class Codes: A61M 2016/0021 with a proportional output signal, e.g. from a thermistor
• CPC Class Codes: A61M 2016/0036 in the breathing tube and used in both inspiratory and expiratory phase
• CPC Class Codes: A61M 2205/3375 Acoustical, e.g. ultrasonic, measuring means

Abstract

Method and apparatus for detecting disordered breathing in a patient and, in particular, to a method and apparatus for detecting patient snoring and/or for dynamically determining a snore detection threshold, and to a pressure support system and method of providing pressure support that uses this unique snore detection and/or snore detection dynamic adjustment technique to control the pressure provided to at patient.

Claims

1. A method of detecting a snore in a patient comprising the steps of:

(a) providing a pressure support system including a pressure sensor;

(b) monitoring a pressure within the pressure support system indicative of a pressure at an airway of a patient and outputting a pressure signal indicative of the pressure;

(c) filtering the pressure signal to obtain a filtered pressure signal that contains frequencies in a specified frequency range;

(d) generating a threshold value;

(e) comparing the filtered pressure signal to the threshold value;

(f) detecting a first vibration responsive to the filtered pressure signal crossing the threshold value;

(g) determining a benchmark period for the first vibration;

,

(h) detecting a second vibration when a second, subsequent vibration in the filtered pressure signal exceeds the threshold value;

(i) determining a second period for the second, subsequent vibration;

(j) comparing the second period to the benchmark period to determine if the periods are consistent; and

(k) declaring a snore responsive to the second period being consistent with the benchmark period.

2. The method according to claim 1, further comprising:

(l) detecting a third vibration subsequent to the second vibration when the filtered digital signal exceeds the threshold value;

(m) determining a third period for the third vibration;

(n) comparing the third period to the benchmark period to determine if the third period and the benchmark period are consistent;

(o) detecting a fourth vibration responsive to the filtered digital signal exceeding the threshold value;

(p) determining a fourth period for the fourth vibration; and

(q) comparing the fourth period to the benchmark period to determine if the fourth period is consistent with the benchmark period, and wherein a snore is declared in step (k) responsive to the periods of the second, third, and fourth vibrations being consistent with the benchmark period.

3. The method according to claim 2, further comprising:

(r) determining a benchmark frequency for a first vibration that exceeds the threshold;

(s) comparing subsequent frequencies of corresponding subsequent vibrations with the benchmark frequency; and

(t) declaring a snore if the subsequent frequencies are each consistent with the benchmark frequency.

4. The method according to claim 1, further comprising the steps of:

(l) determining a benchmark frequency for a first vibration that exceeds the threshold;

(m) comparing subsequent frequencies of corresponding subsequent vibrations with the benchmark frequency; and

(n) declaring a snore if the subsequent frequencies are each consistent with the benchmark frequency.

5. The method according to claim 1, wherein generating a threshold value comprises the steps of:

(1) processing the pressure signal to provide an approximation of average peak values for the pressure signal; and

(2) adding an offset to the approximation of average peak values to produce the threshold value.

6. The method according to claim 5, wherein processing the pressure signal to provide the approximation of average peak values for the pressure signal, comprises:

(i) calculating, during a current processing cycle n, a current approximation of average peak value C[n] as;

C[n]=C[n−

1]+W(Y[n]−

C[n−

1]),

responsive to an approximation of the average peak value in a previous processing cycle C[n−

1] being greater than the pressure signal during a current processing cycle Y[n], and

(ii) calculating, during the current processing cycle, the current approximation of average peak value C[n] as;

$C<br><br>\left[n\right]=C<br><br>\left[n-1\right]-C<br><br>\left[n-1\right]<br><br>\frac{W}{K_3},$

responsive to the approximation of the average peak value in a previous processing cycle C[n−

1] not being greater than the pressure signal during a current processing cycle Y[n], wherein W is charge constant for an RC circuit, and wherein K₃ is a ratio of a charge to discharge rate for the RC circuit.

7. The method according to claim 6, wherein adding the offset to the approximation of average peak values to produce the threshold value T[n] during a current processing cycle comprises processing the current approximation of average peak value C[n] as follows:

T[n]=K₁C[n]+K_y,

where K₁ and K_y are constants.

8. The method according to claim 7, wherein a value of K_y is selected based on the pressure signal.

9. The method according to claim 1, further comprising the step of adjusting the pressure provided by the pressure support system based on whether a snore is declared.

22. The snore detector according to claim 1, further comprising means for adjusting the pressure provided by the pressure support system based on whether a snore is declared.

10. A method of determining a threshold value for a snore detector comprising the steps of:

(a) generating a pressure signal indicative of a pressure at an airway of a patient; and

(b) processing the pressure signal to provide an approximation of average peak values for the pressure signal; and

(c) adding an offset to the approximation of average peak values to produce the threshold value.

11. The method according to claim 10, wherein processing the pressure signal to provide the approximation of average peak values for the pressure signal, comprises:

(1) calculating, during a current processing cycle n, a current approximation of average peak value C[n] as;

C[n]=C[n−

1]+W(Y[n]−

C[n−

1]),

responsive to an approximation of the average peak value in a previous processing cycle C[n−

1] being greater than the pressure signal during a current processing cycle Y[n], and

(2) calculating, during the current processing cycle, the current approximation of average peak value C[n] as;

$C<br><br>\left[n\right]=C<br><br>\left[n-1\right]-C<br><br>\left[n-1\right]<br><br>\frac{W}{K_3},$

responsive to the approximation of the average peak value in a previous processing cycle C[n−

1] not being greater than the pressure signal during a current processing cycle Y[n], wherein W is charge constant for an RC circuit, and wherein K₃ is a ratio of a charge to discharge rate for the RC circuit.

12. The method according to claim 11, wherein adding the offset to the approximation of average peak values to produce the threshold value T[n] during a current processing cycle comprises processing the current approximation of average peak value C[n] as follows:

T[n]=K₁C[n]+K_y,

where K₁ and K_y are constants.

13. The method according to claim 12, wherein a value of K_y is selected based on the pressure signal.

14. A snore detector comprising:

(a) pressure sensor adapted to detect a pressure indicative of a pressure at an airway of a patient and to output a pressure signal indicative thereof,

(b) means for filtering the pressure signal to obtain a filtered pressure signal that contains frequencies in a specified frequency range,

(c) means for providing a threshold value;

(d) means for comparing the filtered pressure signal to the threshold value;

(e) means for detecting a first vibration responsive to the filtered digital signal exceeding the threshold value;

(f) means for determining a benchmark period for the first vibration;

(g) means for detecting a second vibration responsive to a second vibration in the filtered pressure signal exceeding the threshold value;

(h) means for determining a second period for the second vibration;

(i) means for comparing the second period to the benchmark period to determine if the second periods and the benchmark period are consistent; and

(j) means for declaring a snore responsive to the second period being consistent with the benchmark period.

15. The snore detector of claim 14, further comprising:

(k) means for detecting a third vibration responsive to the filtered digital signal exceeding the threshold value;

(l) means for determining a third period for the third vibration;

(m) means for comparing the third period to the benchmark period to determine if the third period and the benchmark period are consistent;

(n) means for detecting a fourth vibration responsive to the filtered digital signal exceeding the threshold value;

(o) means for determining a fourth period for the fourth vibration; and

(p) means for comparing the fourth period to the benchmark period to determine if the fourth period and the benchmark period are consistent, and wherein the means for declaring a snore responsive to the periods of the second, third, and fourth vibrations being consistent with the benchmark period.

16. The snore detector as claimed in claim 15, further comprising:

(q) means for determining a benchmark frequency for a, first vibration that exceeds the threshold; and

(r) means for comparing subsequent frequencies of corresponding subsequent vibrations with the benchmark frequency; and

(s) means for declaring a snore if the subsequent frequencies are each consistent with the benchmark frequency.

17. The snore detector according to claim 14, further comprising:

(k) means for determining a benchmark frequency for a first vibration that exceeds the threshold;

(l) means for comparing subsequent frequencies of corresponding subsequent vibrations with the benchmark frequency; and

(m) means for declaring a snore if the subsequent frequencies are each consistent with the benchmark frequency.

18. The snore detector according to claim 14, wherein the means for providing a threshold value comprises:

(1) means for processing the pressure signal to provide an approximation of average peak values for the pressure signal; and

(2) means for adding an offset to the approximation of average peak values to produce the threshold value.

19. The snore detector according to claim 18, wherein means for processing the pressure signal to provide an approximation of average peak values for the pressure signal:

(i) calculates, during a current processing cycle n, a current approximation of average peak value C[n] as;

C[n]=C[n−

1]+W(Y[n]−

C[n−

1]),

responsive to an approximation of the average peak value in a previous processing cycle C[n−

1] being greater than the pressure signal during a current processing cycle Y[n], and

(ii) calculates, during the current processing cycle, the current approximation of average peak value C[n] as;

$C<br><br>\left[n\right]=C<br><br>\left[n-1\right]-C<br><br>\left[n-1\right]<br><br>\frac{W}{K_3},$

responsive to the approximation of the average peak value in a previous processing cycle C[n−

1] not being greater than the pressure signal during a current processing cycle Y[n], wherein W is charge constant for an RC circuit, and wherein K₃ is a ratio of a charge to discharge rate for the RC circuit.

20. The snore detector according to claim 19, wherein the means for adding the offset to the approximation of average peak values to produce the threshold value T[n] during a current processing cycle add the offset by processing the current approximation of average peak value C[n] as follows:

T[n]=K₁C[n]+K_y,

where K₁ and K_y are constants.

21. The snore detector according to claim 20, wherein a value of K_y is selected based on the pressure signal.

23. A threshold detector that determines a threshold value for a snore detector comprising the steps of:

(a) a pressure sensor adapted to generate a pressure signal indicative of a pressure at an airway of a patient that includes frequencies in a qualifying range;

(b) means for processing the pressure signal to provide an approximation of average peak values for the pressure signal; and

(c) means for adding an offset to the approximation of average peak values to produce the threshold value.

24. The snore detector according to claim 23, wherein means for processing the pressure signal to provide the approximation of average peak values for the pressure signal also:

(1) calculates, during a current processing cycle n, a current approximation of average peak value C[n] as;

C[n]=C[n−

1]+W(Y[n]−

C[n−

1]),

responsive to an approximation of the average peak value in a previous processing cycle C[n−

1] being greater than the pressure signal during a current processing cycle Y[n], and

(2) calculates, during the current processing cycle, the current approximation of average peak value C[n] as;

$C<br><br>\left[n\right]=C<br><br>\left[n-1\right]-C<br><br>\left[n-1\right]<br><br>\frac{W}{K_3},$

responsive to the approximation of the average peak value in a previous processing cycle C[n−

1] not being greater than the pressure signal during a current processing cycle Y[n], wherein W is charge constant for an RC circuit, and wherein K₃ is a ratio of a charge to discharge rate for the RC circuit.

25. , The snore detector according to claim 24, wherein the means for adding the offset to the approximation of average peak values to produce the threshold value T[n] during a current processing cycle adds the offset by processing the current approximation of average peak value C[n] as follows:

T[n]=K₁C[n]+K_y,

where K₁ and K_y are constants.

26. The snore detector, according to claim 25, wherein a value of K_y is selected based on the pressure signal.