Method and apparatus for estimating systolic and mean pulmonary artery pressures of a patient

• US 6,368,283 B1
• Filed: 09/08/2000
• Issued: 04/09/2002
• Est. Priority Date: 09/08/2000
• Status: Expired due to Term

• Current Assignee: Institut De Recherches Cliniques De Montreal, Universite Laval
• Original Assignee: Institut De Recherches Cliniques De Montreal, Universite Laval
• Inventors: Xu, Jingping, Pibarot, Philippe, Durand, Louis-Gilles
• Primary Examiner(s): Nasser, Robert L.
• Application Number:
  US09/658,631
• Time in Patent Office: 578 Days
• Field of Search: 600/481, 600/485, 600/500, 600/586, 600/528
• US Class Current: 600/485
• CPC Class Codes: A61B 5/0205 Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
• CPC Class Codes: A61B 5/021 Measuring pressure in heart or blood vessels A61B5/0205 takes precedence
• CPC Class Codes: A61B 5/024 Detecting, measuring or recording pulse rate or heart rate A61B5/0205, A61B5/021 take precedence
• CPC Class Codes: A61B 5/0245 by using sensing means generating electric signals, i.e. ECG signals
• CPC Class Codes: A61B 5/7257 using Fourier transforms
• CPC Class Codes: A61B 5/7275 Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
• CPC Class Codes: A61B 7/04 Electric stethoscopes microphones, acoustic transducers therefor H04R

Abstract

The method of estimating systolic and mean pulmonary artery pressures of a patient, comprising the steps of (a) producing an electric signal x_s(t) representative of heart sounds of the patient; (b) extracting second heart sound S₂(t) from the signal produced in step (a); (c) extracting pulmonary and aortic components P₂(t) and A₂(t) from S₂(t); (d) extracting a signal representative of mean cardiac interval; (e) correlating the pulmonary and aortic components P₂(t) and A₂(t) to obtain a cross correlation function; (f) measuring a splitting interval of the cross correlation function obtained in step (e); (g) producing a normalized splitting interval; and (h) estimating the systolic and mean pulmonary artery pressures by means of predetermined regressive functions. The present invention also relates to an apparatus for estimating systolic and mean pulmonary artery pressures of a patient.

Claims

1. Method of estimating systolic and mean pulmonary artery pressures of a patient, comprising the steps of:

(a) producing an electric signal x_s(t) representative of heart sounds of the patient;

(b) extracting second heart sound S₂(t) from the signal produced in step (a);

(c) extracting pulmonary and aortic components P₂(t) and A₂(t) from S₂(t);

(d) extracting a signal representative of mean cardiac interval from the signal produced in step (a);

(e) correlating the pulmonary and aortic components P₂(t) and A₂(t) to obtain a cross correlation function;

(f) measuring a splitting interval as the time of occurrence of the maximal value of the cross correlation function obtained in step (e);

(g) producing a normalized splitting interval by dividing the splitting interval obtained in step (f) by the mean cardiac interval obtained in step (d); and

(h) estimating the systolic and mean pulmonary artery pressures using of predetermined regressive functions, said predetermined regressive functions describing relationships between the normalized splitting interval and the systolic and mean pulmonary artery pressures.

2. Method according to claim 1, wherein the extracting of the pulmonary and aortic components P₂(t) and A₂(t) from S₂(t) in step (c) comprises the steps of:

(i) determining a Wigner-Ville distribution W_s(t,f) in view of time t and frequency f of the signal S₂(t) produced in step (b) using the following function;

$W_S<br><br>\left(t,f\right)={\int <br><br>}_{-\mathrm{\infty <br><br>}}^{\mathrm{\infty <br><br>}}<br><br>S_2<br><br>\left(t+\frac{\mathrm{\tau <br><br>}}{2}\right)·<br><br>S_2^{*}<br><br>\left(t-\frac{\mathrm{\tau <br><br>}}{2}\right)·<br><br>{\mathrm{<br><br>}}^{-\mathrm{j2\pi <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>f<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\tau <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br><br><br>\mathrm{\tau <br><br>};<br><br>$

(ii) filtering W_s(t,f) obtained in step (i) by means of the following function to obtain a masked time frequency representation m_A(t,f) of the aortic component A₂(t);

3. Method according to claim 2 wherein the low-pass filtering in step (v) (C) has a cut-off frequency selected within the range of 16 to 64 Hz and the low-pass filtering in step (xii) (C) has a cut-off frequency selected within the range of 16 to 64 Hz.

4. Method according to claim 3 where determining the cut-off frequency of the low-pass filter in step (v) (C) comprises the steps of:

(1) transforming the S₂(t) signal into frequency domain using a Discrete Fourier Transform to obtain a transform;

(2) determining a power spectrum of the transform obtained in step (1);

(3) determining a main-low frequency energy lobe of the power spectrum;

(4) determining the cut-off frequency as a frequency corresponding to 5% of energy of the main-low frequency energy lobe;

(5) determining frequency bins of the S₂(t) signal using a Discrete Fourier Transform;

(6) multiplying real and imaginary parts of frequency bins which are below the cut-off frequency by 1.00;

(7) multiplying real and imaginary parts of the of a frequency bin corresponding to the cut-off frequency by 0.70;

(8) multiplying real and imaginary parts of the of a first frequency bin above the cut-off frequency by 0.20;

(9) multiplying real and imaginary parts of the of a second frequency bin above the cut-off frequency by 0.02;

(10) multiplying real and imaginary parts of all other of frequency bins which are above the cut-off frequency of the low pass filter by 0.00; and

(11) applying inverse Discrete Fourier Transform to the results of steps 6, 7, 8, 9 and 10 to obtain representations in the time domain.

5. Method according to claim 3 where determining the cut-off frequency of the low-pass filter in step (xii) (C) comprises the steps of:

(1) transforming the difference signal X_D(t) into frequency domain using a Discrete Fourier Transform to obtain a transform;

(2) determining a power spectrum of the transform obtained in step (1);

(3) determining a main-low frequency energy lobe of the power spectrum;

(4) determining the cut-off frequency as a frequency corresponding to 5% of energy of the main-low frequency energy lobe;

(5) determining frequency bins of the difference signal X_D(t) using a Discrete Fourier Transform;

(6) multiplying real and imaginary parts of frequency bins which are below the cut-off frequency by 1.00;

(7) multiplying real and imaginary parts of a frequency bin corresponding to the cut-off frequency by 0.70;

(8) multiplying real and imaginary parts of the of a first frequency bin above the cut-off frequency by 0.20;

(9) multiplying real and imaginary parts of the of a second frequency bin above the cut-off frequency by 0.02;

(10) multiplying the real and imaginary parts of all other of the frequency bins which are above the cut-off frequency of the low pass filter by 0.00; and

(11) applying inverse Discrete Fourier Transform to the results of steps 6, 7, 8, 9 and 10 to obtain representations in the time domain.

6. Method according to claim 1 where the predetermined regressive functions have the form:

7. Apparatus for estimating systolic and mean pulmonary artery pressures of a patient, comprising:

first producing means for producing an electric signal x_S(t) representative of heart sounds of the patient;

first extracting means for extracting second heart sound S₂(t) from the signal produced by the first producing means;

second extracting means for extracting pulmonary and aortic components P₂(t) and A₂(t) from S₂(t) extracted by the first extracting means;

third extracting means for extracting a signal representative of mean cardiac interval from the signal produced by the first producing means;

correlating means for correlating the pulmonary and aortic components P₂(t) and A₂(t) to obtain a cross correlation function;

measuring means for measuring a splitting interval as the time of occurrence of the maximal value of the cross correlation function obtained by the correlating means;

second producing means for producing a normalized splitting interval by dividing the splitting interval obtained from the measuring means by the mean cardiac interval obtained from the third extracting means; and

estimating means for estimating the systolic and mean pulmonary artery pressures by means of predetermined regressive functions said predetermined regressive functions describing relationships between the normalized splitting interval and the systolic and mean pulmonary artery pressures.

8. Apparatus according to claim 7, wherein the second extracting means for extracting the pulmonary and aortic components P₂(t) and A₂(t) from S₂(t) comprises:

first determining means for determining a Wigner-Ville distribution W_S(t,f) in view of time t and frequency f of the signal S₂(t) extracted by the first extracting means using the following function;

$W_S<br><br>\left(t,f\right)={\int <br><br>}_{-\mathrm{\infty <br><br>}}^{\mathrm{\infty <br><br>}}<br><br>S_2<br><br>\left(t+\frac{\mathrm{\tau <br><br>}}{2}\right)·<br><br>S_2^{*}<br><br>\left(t-\frac{\mathrm{\tau <br><br>}}{2}\right)·<br><br>{\mathrm{<br><br>}}^{-\mathrm{j2}<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\pi <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>f<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{\tau <br><br>}}<br><br><br><br>\mathrm{\tau <br><br>};<br><br>$

first filtering means for filtering W_S(t,f) obtained from the first determining means using the following function to obtain a masked time frequency representation m_A(t,f) of the aortic component A₂(t);

9. Apparatus according to claim 8 where the predetermined regressive functions have the form: