Method of and apparatus for analyzing a signal

• US 6,216,094 B1
• Filed: 02/13/1998
• Issued: 04/10/2001
• Est. Priority Date: 02/24/1995
• Status: Expired due to Term

• Current Assignee: Lidco Group, PLC (Masimo Corporation)
• Original Assignee: Monitoring Technology Corporation
• Inventors: Fox Linton, Nicholas William, Fox Linton, Robert Anthony, Band, David Marston
• Primary Examiner(s): SHAH, KAMINI S
• Application Number:
  US08/894,571
• Time in Patent Office: 1,152 Days
• Field of Search: 702/100, 702/65, 702/131, 600/481, 600/505, 600/526, 600/549, 128/898
• US Class Current: 702/100
• CPC Class Codes: A61B 5/0275 using tracers, e.g. dye dilution
• CPC Class Codes: A61B 5/029 Measuring or recording blood output from the heart, e.g. minute volume A61B8/065 takes precedence

Abstract

A method of analyzing a signal representing a physical parameter to obtain information from said signal and to extrapolate information contained in said signal and/or to ignore an interfering component in part of said signal, said signal representing either only a first portion of a skew distribution, or a first portion of a skew distribution and a second portion containing an interfering component, in which the area of a skew distribution, preferably a lognormal distribution is measured. The method is particularly applicable to the measurement of cardiac output. Apparatus for carrying out the method is also disclosed.

Claims

1. A method of analyzing a signal representing a physical parameter for at least one of the purposes:

(i) to obtain information from said signal, (ii) to extrapolate information contained in said signal and (iii) to ignore an interfering component in part of said signal, said signal representing one of the group consisting of only a first portion of a skew distribution, and a first portion of a skew distribution, and a second portion containing said interfering component, said method comprising the steps of;

a) selecting two fractions of the maximum value of said signal in said first portion and selecting whether said fractions are to represent positions on said skew distribution having a positive or negative slope, said fractions being used to delineate between first and second parts of said skew distribution, said first and second parts being delineated to lie within said first portion of said skew distribution,

b) determining the integrals of said first and second parts, and

c) determining the integrals of said complete skew distribution using said determined integrals of said first and second parts and a predetermined relationship between the integral of a complete skew distribution and functions of integrals of said first and second parts for said selected fractions.

2. A method as claimed in claim 1 wherein said skew distribution is a chi squared distribution.

3. A method as claimed in claim 1 wherein said skew distribution can be approximated as a lognormal distribution.

4. A method as claimed in claim 1 including the step of filtering said signal to remove noise before the step of determining the integrals of said first and second parts.

5. A method as claimed in claim 1 wherein said fractions are selected such that the integrals of said first and second parts are substantially equal.

6. A method as claimed in claim 1 wherein said fractions are less than 1.

7. A method as claimed in claim 1 wherein said signal represents a temporal skew distribution for said physical parameter, said step of determining the integral of said complete skew distribution providing a measure of the quantum of said physical parameter.

8. A method as claimed in claim 1 wherein said signal represents the concentration of a substance in a flowing fluid, and said determined integral of said complete skew distribution being used to calculate the flow rate of said fluid.

9. A method as claimed in claim 1 wherein said signal represents current and said determined integral of said complete skew distribution represents charge.

10. A method as claimed in claim 1 wherein said relationship is determined by equating said integral of said complete skew distribution divided by the sum of the integral of said first part and the integral of said second part with a function of the ratio of the integral of said first part to the integral of said second part.

11. A method as claimed in claim 1 wherein said skew distribution is approximated as a lognormal distribution and said relationship is determined from the equations $A=\alpha$

×

φ



{-σ

±

2

ln

(FA-1)}$B=\mathrm{\alpha <br><br>}\times <br><br>\mathrm{\phi <br><br>}<br><br>\left\{-\mathrm{\sigma <br><br>}\pm <br><br>\sqrt{2<br><br>\ln <br><br>\left(F_B^{-1}\right)}\right\}$$\mathrm{and}$$\mathrm{\phi <br><br>}<br><br>\left(x\right)=\frac{1}{\sqrt{2<br><br>\mathrm{\pi <br><br>}}}<br><br>{\int <br><br>}_{-\mathrm{\alpha <br><br>}}^X<br><br>{\mathrm{<br><br>}}^{-0.5<br><br>t^2}<br><br><br><br>t$

where

A is the integral of said first part

B is the integral of said second part

α

is the integral of said complete skew distribution

σ

is the standard deviation of the normal distribution from which the logarithmic transformation was obtained and which indicates the degree to which said distribution is skew

F_A is a first of said preselected fractions delineating said first and second parts

F_B is a second of said preselected fractions delineating said second part and any further part of said skew distribution

φ

is the cumulative normal distribution function

t is time

x is the signal value, and

±

determines whether said preselected fractions F_A and F_B are on a positive slope (−

) or on a negative slope (+).

12. A method as claimed in claim 1 wherein values for the integral of complete skew distributions for different ratios of the integral of said first and second parts, for different fractions selected in said selecting step, and for different degrees of skewness are prestored in a look-up table, said step of determining the integral of said complete skew distribution comprising the step of calculating the ratio of the integral of said first and second parts, and looking-up the integral of said complete skew distribution using said selected fractions and the ratio calculated in said calculating steep.

13. A method as claimed in claim 1 wherein said signal represents a spatial skew distribution for said physical parameter, said step of calculating the integral of said complete skew distribution providing a measure of the quantum of the spatially distributed said physical parameter.

14. A method of measuring flow rate of a fluid comprising the steps of

a) adding a known quantity of a substance into said fluid at a point in time,

b) measuring the quantity of said substance in said fluid at a point downstream to generate a signal, and

c) analyzing said signal in accordance with the method of claim 1.

15. A method as claimed in claim 14 wherein said substance is soluble in said fluid and the concentration of said substance in said fluid is measured at said downstream point.

16. A method of determining the flow rate of a fluid comprising the steps of:

a) adding a known quantity of a substance to said fluid at a point in time,

b) measuring the quantity of said substance in said fluid at a point downstream to generate a distribution with respect to time, said distribution approximating either only a first portion of a skew distribution, or a first portion of a skew distribution and a second portion containing extraneous components,

c) assuming said distribution can be approximated by a skew distribution and selecting two fractions of the maximum value of said signal in said first portion and selecting whether said fractions are to represent positions on said skew distribution having a positive or a negative slope, said fractions being used to delineate between first and second parts of said skew distribution, said first and second parts being delineated to lie within said first portion of said skew distribution,

d) calculating the integrals of said first and second parts,

e) determining the integral of the complete skew distribution using said calculated integrals of said first and second parts and a predetermined relationship between functions of the integrals of said first and second parts and the integral of said complete skew distribution for said selected fractions, and

f) calculating the flow rate of said fluid from said known quantity of said substance and said integral of said complete skew distribution.

17. A method as claimed in claim 16 including the step of digitizing the measurement of the quantity of said substance, and storing said digitized measurement for analysis.

18. A method as claimed in claim 16 wherein said step of calculating the flow rate comprises the step of dividing the determined integral of said complete skew distribution into said known quantity of said substance.

19. A method as claimed in claim 16 wherein said skew distribution can be approximated to a lognormal distribution.

20. A method as claimed in claim 16 including one of the steps from the group consisting of filtering and integrating said distribution to remove noise before the step of calculating the integral of said first and second parts.

21. A method as claimed in claim 16 wherein said fractions are selected such that the integrals of said first and second parts are substantially equal.

22. A method as claimed in claim 16 wherein said fractions are less than 1.

23. A method as claimed in claim 16 wherein said relationship is determined by equating said integral of said complete skew distribution divided by the sum of the integral of said first part and the integral of said second part with a function of the ratio of the integral of said first part to the integral of said second part.

24. A method as claimed in claim 16 wherein said skew distribution is approximated as a lognormal distribution and said relationship is determined from the equations $A=\alpha$

×

φ



{-σ

±

2

ln

(FA-1)}$B=\mathrm{\alpha <br><br>}\times <br><br>\mathrm{\phi <br><br>}<br><br>\left\{-\mathrm{\sigma <br><br>}\pm <br><br>\sqrt{2<br><br>\ln <br><br>\left(F_B^{-1}\right)}\right\}$$\mathrm{and}$$\mathrm{\phi <br><br>}<br><br>\left(X\right)=\frac{1}{\sqrt{2<br><br>\mathrm{\pi <br><br>}}}<br><br>{\int <br><br>}_{-\mathrm{\alpha <br><br>}}^X<br><br>{\mathrm{<br><br>}}^{-0.5<br><br>t^2}<br><br><br><br>t$

where

A is the integral of said first part

B is the integral of said second part

α

is the integral of said complete skew distribution

σ

is the standard deviation of the normal distribution from which the logarithmic transformation was obtained and which indicates the degree to which said distribution is skew

F_A is a first of said preselected fractions delineating said first and second parts

F_B is a second of said preselected fractions delineating said second part and any further part of said skew distribution

φ

is the cumulative normal distribution function

t is time

x is the signal value, and

±

determines whether said preselected fractions F_A and F_B are on a positive slope (−

) or on a negative slope (+).

25. A method as claimed in claim 16 wherein values for the integrals of complete skew distributions for different ratios of the integrals of said first and second parts, for different said selected fractions, and for different degrees of skewness are prestored in a look-up table, said step of determining the integral of said complete skew distribution comprising the step of calculating the ratio of the integrals of said first and second parts, and looking-up the integral of said complete skew distribution using said selected fractions and said calculated ratio.

26. Apparatus for analyzing a signal representing a physical parameter for at least one of the purposes:

(i) to obtain information from said signal, (ii) to extrapolate information contained in said signal and (iii) to ignore an interfering component in part of said signal, said signal representing one of a group consisting of only a first portion of a skew distribution and a first portion of a skew distribution, and a second portion containing said interfering component, said system comprising;

a) means for selecting two fractions of the maximum value of said signal in said first portion and selecting whether said fractions are to represent positions on said skew distribution having a positive or a negative slope, said fractions being used to delineate between first and second parts of said skew distribution, said first and second parts being delineated to lie within said first portion of said skew distribution,

b) means for determining the integrals of said first and second parts, and

c) means for determining the integrals of said complete skew distribution using said determined integrals of said first and second parts and a predetermined relationship between the integral of a complete skew distribution and functions of integrals of said first and second parts for said selected fractions.

27. Apparatus as claimed in claim 26 wherein said means for determining the integral of said complete skew distribution uses a relationship between the integral of a complete lognormal distribution and integrals of said first and second parts for said selected fractions.

28. Apparatus as claimed in claim 26 including filter means for filtering said signal to remove noise before the integrals of said first and second parts are determined.

29. Apparatus as claimed in claim 26 including means to determine said relationship by equating said integral of said complete skew distribution divided by the sum of the integral of said first part and the integral of said second part with a function of the ratio of the integral of said first part to the integral of said second part.

30. Apparatus as claimed in claim 26 wherein said skew distribution is approximated as a lognormal distribution including means to determine said relationship using the equations:

$A=\mathrm{\alpha <br><br>}\times <br><br>\mathrm{\phi <br><br>}<br><br>\left\{-\mathrm{\sigma <br><br>}\pm <br><br>\sqrt{2<br><br>\ln <br><br>\left(F_A^{-1}\right)}\right\}$$B=\mathrm{\alpha <br><br>}\times <br><br>\mathrm{\phi <br><br>}<br><br>\left\{-\mathrm{\sigma <br><br>}\pm <br><br>\sqrt{2<br><br>\ln <br><br>\left(F_B^{-1}\right)}\right\}$$\mathrm{and}$$\mathrm{\phi <br><br>}<br><br>\left(X\right)=\frac{1}{\sqrt{2<br><br>\mathrm{\pi <br><br>}}}<br><br>{\int <br><br>}_{-\mathrm{\alpha <br><br>}}^X<br><br>{\mathrm{<br><br>}}^{-0.5<br><br>t^2}<br><br><br><br>t$

where

A is the integral of said first part

B is the integral of said second part

α

is the integral of said complete skew distribution

σ

is the standard deviation of the normal distribution from which the logarithmic transformation was obtained and which indicates the degree to which said distribution is skew

F_A is a first of said preselected fractions delineating said first and second parts

F_B is a second of said preselected fractions delineating said second part and any further part of said skew distribution

φ

is the cumulative normal distribution function

t is time

x is the signal value, and

±

determines whether said preselected fractions F_A and F_B are on a positive slope (−

) or on a negative slope (+).

31. Apparatus as claimed in claim 26 including a look-up table containing values for the integrals of complete skew distributions for different ratios of the integrals of said first and second parts, for different said selected fractions, and for different degrees of skewness, said means for determining the integral of said complete distribution comprising means for calculating the ratios of the integrals of said first and second parts, and means for looking-up the integral of said complete skew distribution using said selected fractions and said calculated ratio.

32. Apparatus for determining the flow rate of a fluid comprising:

a) sensor means for measuring the quantity of a substance in said fluid at a point downstream of where a known quantity said substance has been added to said fluid, the measurement generating a distribution with respect to time which approximates either only a first portion of a skew distribution, or a first portion of a skew distribution and a second portion containing extraneous components,

b) means for selecting two fractions of the maximum value of said signal in said first portion and selecting whether said fractions are to represent positions on said skew distribution having a positive or a negative slope, said fractions being used to delineate between first and second parts of said skew distribution, said first and second parts being delineated to lie within said first portion of said skew distribution,

c) means for calculating the integrals of said first and second parts,

d) means for determining the integral of the complete skew distribution using said calculated integrals of said first and second parts and a predetermined relationship between functions of integrals of said first and second parts and the integral of said complete skew distribution for said selected fractions, and

e) means for calculating the flow rate of said fluid from said known quantity of said substance and said integral of said complete skew distribution.

33. Apparatus as claimed in claim 32 including digitizing means for digitizing said measurement, and storage means for storing the digitized measurement for analysis.

34. Apparatus as claimed in claim 32 wherein said means for calculating the flow rate comprises means for dividing the determined integral of said complete skew distribution into said known quantity of said substance.

35. Apparatus as claimed in claim 32 including filter means to filter said measurement to remove noise before calculating the integrals of said first and second parts.

36. Apparatus as claimed in claim 32 including means for predetermining the relationship by equating said integral of said complete skew distribution divided by the sum of the integral of said first part and the integral of said second part with a function of the ratio of the integral of said first part to the integral of said second part.

37. Apparatus as claimed in claim 32 including means for predetermining the relationship by assuming said skew distribution approximates a lognormal distribution and using the equations:

$A=\mathrm{\alpha <br><br>}\times <br><br>\mathrm{\phi <br><br>}<br><br>\left\{-\mathrm{\sigma <br><br>}\pm <br><br>\sqrt{2<br><br>\ln <br><br>\left(F_A^{-1}\right)}\right\}$$B=\mathrm{\alpha <br><br>}\times <br><br>\mathrm{\phi <br><br>}<br><br>\left\{-\mathrm{\sigma <br><br>}\pm <br><br>\sqrt{2<br><br>\ln <br><br>\left(F_B^{-1}\right)}\right\}$$\mathrm{and}$$\mathrm{\phi <br><br>}<br><br>\left(X\right)=\frac{1}{\sqrt{2<br><br>\mathrm{\pi <br><br>}}}<br><br>{\int <br><br>}_{-\mathrm{\alpha <br><br>}}^X<br><br>{\mathrm{<br><br>}}^{-0.5<br><br>t^2}<br><br><br><br>t$

where

A is the integral of said first part

B is the integral of said second part

α

is the integral of said complete skew distribution

σ

is the standard deviation of the normal distribution from which the logarithmic transformation was obtained and which indicates the degree to which said distribution is skew

F_A is a first of said preselected fractions delineating said first and second parts

F_B is a second of said preselected fractions delineating said second part and any further part of said skew distribution

φ

is the cumulative normal distribution function

t is time

x is the signal value, and

±

determines whether said preselected fractions F_A and F_B are on a positive slope (−

) or on a negative slope (+).

38. Apparatus as claimed in claim 32 including a look-up table containing values for the integrals of complete skew distributions for different ratios of the integrals of said first and second parts, for different said selected fractions, and for different degrees of skewness, said means for determining the integral of said complete distribution comprising means for calculating the ratios of the integrals of said first and second parts, and means for looking-up the integral of said complete skew distribution using said selected fractions and said calculated ratio.