Detection of Vascular Conditions Using Arterial Pressure Waveform Data

• US 20100204590A1
• Filed: 01/26/2010
• Published: 08/12/2010
• Est. Priority Date: 02/09/2009
• Status: Abandoned Application

• Current Assignee: Edwards Lifesciences Corporation
• Original Assignee: Edwards Lifesciences Corporation
• Inventors: Roteliuk, Luchy, Hatib, Feras
• Application Number:
  US12/694,205
• Publication Number:
  US 20100204590A1
• Time in Patent Office: Days
• US Class Current: 600/485
• CPC Class Codes: A61B 5/02028 Determining haemodynamic parameters not otherwise provided for, e.g. cardiac contractility or left ventricular ejection fraction
• CPC Class Codes: A61B 5/021 Measuring pressure in heart or blood vessels A61B5/0205 takes precedence
• CPC Class Codes: A61B 5/02108 from analysis of pulse wave characteristics
• CPC Class Codes: A61B 5/412 Detecting or monitoring sepsis
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 50/50 for simulation or modelling of medical disorders
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients

Abstract

Multivariate statistical models for the detection of vascular conditions, methods for creating such multivariate statistical models, and methods for the detection of vascular condition in a subject using the multivariate statistical models are described. The models are created based on arterial pressure waveform data from a first group of subjects that were experiencing a particular vascular condition and a second group of subjects that were not experiencing the same vascular condition. The multivariate statistical models are set up to provide different output values for each set of input data. Thus, when data from a subject under observation is input into the model, the relationship of the model output value to the established output values for the two groups upon which the model was established will indicate whether the subject is experiencing the vascular condition.

Claims

1. A model for use in detecting a vascular condition in a subject comprising:

a multivariate statistical model based on arterial pressure waveform data from a first group of subjects that were experiencing the vascular condition and a second group of subjects that were not experiencing the vascular condition, the multivariate statistical model providing a model output value that corresponds to a first established value for the arterial pressure waveforms of the first set of arterial pressure waveform data and a second established value for the arterial pressure waveforms of the second set of arterial pressure waveform data.

2. The model of claim 1, wherein the multivariate statistical model is based on a set of factors including one or more parameters affected by the vascular condition.

3. The model of claim 2, wherein the one or more parameters affected by the vascular condition are selected from the group consisting of (a) a parameter based on the pulse beats standard deviation of a set of arterial pressure waveform data, (b) a parameter based on the R-to-R interval of a set of arterial pressure waveform data, (c) a parameter based on the area under the systolic portion of a set of arterial pressure waveform data, (d) a parameter based on the duration of systole of a set of arterial pressure waveform data, (e) a parameter based on the duration of the diastole of a set of arterial pressure waveform data, (f) a parameter based on the mean arterial pressure of a set of arterial pressure waveform data, (g) a parameter based on the pressure weighted standard deviation of a set of arterial pressure waveform data, (h) a parameter based on the pressure weighted mean of a set of arterial pressure waveform data, (i) a parameter based on the arterial pulse beats skewness values of a set of arterial pressure waveform data, (j) a parameter based on the arterial pulse beats kurtosis values of a set of arterial pressure waveform data, (k) a parameter based on the pressure weighted skewness of a set of arterial pressure waveform data, (l) a parameter based on the pressure weighted kurtosis of a set of arterial pressure waveform data, and (m) a parameter based on the pressure dependent Windkessel compliance of a set of arterial pressure waveform data.

4. The method of claim 2, wherein the one or more parameters affected by the vascular condition include (a) a parameter based on the pulse beats standard deviation of a set of arterial pressure waveform data, (b) a parameter based on the R-to-R interval of a set of arterial pressure waveform data, (c) a parameter based on the area under the systolic portion of a set of arterial pressure waveform data, (d) a parameter based on the duration of systole of a set of arterial pressure waveform data, (e) a parameter based on the duration of the diastole of a set of arterial pressure waveform data, (f) a parameter based on the mean arterial pressure of a set of arterial pressure waveform data, (g) a parameter based on the pressure weighted standard deviation of a set of arterial pressure waveform data, (h) a parameter based on the pressure weighted mean of a set of arterial pressure waveform data, (i) a parameter based on the arterial pulse beats skewness values of a set of arterial pressure waveform data, (j) a parameter based on the arterial pulse beats kurtosis values of a set of arterial pressure waveform data, (k) a parameter based on the pressure weighted skewness of a set of arterial pressure waveform data, (l) a parameter based on the pressure weighted kurtosis of a set of arterial pressure waveform data, and (m) a parameter based on the pressure dependent Windkessel compliance of a set of arterial pressure waveform data.

5. The method of claim 2, further comprising detecting the vascular condition additionally using one or more of (n) a parameter based on the shape of the beat-to-beat arterial blood pressure signal and at least one statistical moment of the arterial blood pressure signal having an order of one or greater, (o) a parameter corresponding to the heart rate, and (p) a set of anthropometric parameters of the subject.

6. The method of claim 1, wherein the first established value is greater than the second established value.

7. The method of claim 1, wherein the first established value is a positive number and the second established value is a negative number.

8. The method of claim 1, wherein the first established value is +100 and the second established value is −

100.

9. A method for creating a model for use in detecting a vascular condition in a subject comprising:

providing a first set of arterial pressure waveform data from a first group of subjects that were experiencing the vascular condition;

providing a second set of arterial pressure waveform data from a second group of subjects that were not experiencing the vascular condition; and

building a multivariate statistical model based on the first and second sets of arterial pressure waveform data, the multivariate statistical model providing a model output value that corresponds to a first established value for the arterial pressure waveforms of the first set of arterial pressure waveform data and a second established value for the arterial pressure waveform of the second set of arterial pressure waveform data.

10. The method of claim 9, wherein the multivariate statistical model is based on a set of factors including one or more parameters affected by the vascular condition.

11. The method of claim 10, wherein the one or more parameters affected by the vascular condition are selected from the group consisting of (a) a parameter based on the pulse beats standard deviation of a set of arterial pressure waveform data, (b) a parameter based on the R-to-R interval of a set of arterial pressure waveform data, (c) a parameter based on the area under the systolic portion of a set of arterial pressure waveform data, (d) a parameter based on the duration of systole of a set of arterial pressure waveform data, (e) a parameter based on the duration of the diastole of a set of arterial pressure waveform data, (f) a parameter based on the mean arterial pressure of a set of arterial pressure waveform data, (g) a parameter based on the pressure weighted standard deviation of a set of arterial pressure waveform data, (h) a parameter based on the pressure weighted mean of a set of arterial pressure waveform data, (i) a parameter based on the arterial pulse beats skewness values of a set of arterial pressure waveform data, (j) a parameter based on the arterial pulse beats kurtosis values of a set of arterial pressure waveform data, (k) a parameter based on the pressure weighted skewness of a set of arterial pressure waveform data, (l) a parameter based on the pressure weighted kurtosis of a set of arterial pressure waveform data, and (m) a parameter based on the pressure dependent Windkessel compliance of a set of arterial pressure waveform data.

12. The method of claim 10, wherein the one or more parameters affected by the vascular condition include (a) a parameter based on the pulse beats standard deviation of a set of arterial pressure waveform data, (b) a parameter based on the R-to-R interval of a set of arterial pressure waveform data, (c) a parameter based on the area under the systolic portion of a set of arterial pressure waveform data, (d) a parameter based on the duration of systole of a set of arterial pressure waveform data, (e) a parameter based on the duration of the diastole of a set of arterial pressure waveform data, (f) a parameter based on the mean arterial pressure of a set of arterial pressure waveform data, (g) a parameter based on the pressure weighted standard deviation of a set of arterial pressure waveform data, (h) a parameter based on the pressure weighted mean of a set of arterial pressure waveform data, (i) a parameter based on the arterial pulse beats skewness values of a set of arterial pressure waveform data, (j) a parameter based on the arterial pulse beats kurtosis values of a set of arterial pressure waveform data, (k) a parameter based on the pressure weighted skewness of a set of arterial pressure waveform data, (l) a parameter based on the pressure weighted kurtosis of a set of arterial pressure waveform data, and (m) a parameter based on the pressure dependent Windkessel compliance of a set of arterial pressure waveform data.

13. The method of claim 10, further comprising detecting the vascular condition additionally using one or more of (d) a parameter based on the shape of the beat-to-beat arterial blood pressure signal and at least one statistical moment of the arterial blood pressure signal having an order of one or greater, (e) a parameter corresponding to the heart rate, and (f) a set of anthropometric parameters of the subject.

14. The method of claim 9, wherein the first established value is greater than the second established value.

15. The method of claim 9, wherein the first established value is a positive number and the second established value is a negative number.

16. The method of claim 9, wherein the first established value is +100 and the second established value is −

100.

17. The method of claim 9, wherein building a multivariate statistical model comprises:

determining an approximating function relating a set of clinically determined reference measurements of a model output value that corresponds to a first established value for the arterial pressure waveforms of the first set of arterial pressure waveform data and a second established value for the arterial pressure waveform of the second set of arterial pressure waveform data, the output values representing clinical measurements of the cardiovascular parameter from both subjects not experiencing the vascular condition and subjects experiencing the vascular condition, the approximating function being a function of one or more of the following parameters(a) a parameter based on the pulse beats standard deviation of a set of arterial pressure waveform data, (b) a parameter based on the R-to-R interval of a set of arterial pressure waveform data, (c) a parameter based on the area under the systolic portion of a set of arterial pressure waveform data, (d) a parameter based on the duration of systole of a set of arterial pressure waveform data, (e) a parameter based on the duration of the diastole of a set of arterial pressure waveform data, (f) a parameter based on the mean arterial pressure of a set of arterial pressure waveform data, (g) a parameter based on the pressure weighted standard deviation of a set of arterial pressure waveform data, (h) a parameter based on the pressure weighted mean of a set of arterial pressure waveform data, (i) a parameter based on the arterial pulse beats skewness values of a set of arterial pressure waveform data, (j) a parameter based on the arterial pulse beats kurtosis values of a set of arterial pressure waveform data, (k) a parameter based on the pressure weighted skewness of a set of arterial pressure waveform data, (l) a parameter based on the pressure weighted kurtosis of a set of arterial pressure waveform data, and (m) a parameter based on the pressure dependent Windkessel compliance of a set of arterial pressure waveform data;

determining a set of arterial blood pressure parameters from the arterial blood pressure waveform data, the set of arterial blood pressure parameters including one or more of the following parameters(a) a parameter based on the pulse beats standard deviation of a set of arterial pressure waveform data, (b) a parameter based on the R-to-R interval of a set of arterial pressure waveform data, (c) a parameter based on the area under the systolic portion of a set of arterial pressure waveform data, (d) a parameter based on the duration of systole of a set of arterial pressure waveform data, (e) a parameter based on the duration of the diastole of a set of arterial pressure waveform data, (f) a parameter based on the mean arterial pressure of a set of arterial pressure waveform data, (g) a parameter based on the pressure weighted standard deviation of a set of arterial pressure waveform data, (h) a parameter based on the pressure weighted mean of a set of arterial pressure waveform data, (i) a parameter based on the arterial pulse beats skewness values of a set of arterial pressure waveform data, (j) a parameter based on the arterial pulse beats kurtosis values of a set of arterial pressure waveform data, (k) a parameter based on the pressure weighted skewness of a set of arterial pressure waveform data, (l) a parameter based on the pressure weighted kurtosis of a set of arterial pressure waveform data, and (m) a parameter based on the pressure dependent Windkessel compliance of a set of arterial pressure waveform data; and

estimating the model output value by evaluating the approximating function with the set of the parameters.

18. A method for detecting a vascular condition in a subject comprising:

providing arterial pressure waveform data from the subject;

applying a multivariate statistical model to the arterial pressure waveform data to determine a cardiovascular parameter, the multivariate statistical model being prepared from a first set of arterial pressure waveform data from a first group of subjects that were experiencing the vascular condition and a second set of arterial pressure waveform data from a second group of subjects that were not experiencing the vascular condition, the multivariate statistical model providing a cardiovascular parameter that corresponds to a first established value for the arterial pressure waveforms of the first set of arterial pressure waveform data and a second established value for the arterial pressure waveform of the second set of arterial pressure waveform data; and

comparing the cardiovascular parameter to a threshold value,

wherein a cardiovascular parameter equal to or greater than the threshold value indicates the subject is experiencing the vascular condition, and a cardiovascular parameter less than the threshold value indicates the subject is not experiencing the vascular condition.

19. The method of claim 18, wherein the multivariate statistical model is based on a set of factors including one or more parameters affected by the vascular condition.

20. The method of claim 19, wherein the one or more parameters affected by the vascular condition are selected from the group consisting of (a) a parameter based on the pulse beats standard deviation of a set of arterial pressure waveform data, (b) a parameter based on the R-to-R interval of a set of arterial pressure waveform data, (c) a parameter based on the area under the systolic portion of a set of arterial pressure waveform data, (d) a parameter based on the duration of systole of a set of arterial pressure waveform data, (e) a parameter based on the duration of the diastole of a set of arterial pressure waveform data, (f) a parameter based on the mean arterial pressure of a set of arterial pressure waveform data, (g) a parameter based on the pressure weighted standard deviation of a set of arterial pressure waveform data, (h) a parameter based on the pressure weighted mean of a set of arterial pressure waveform data, (i) a parameter based on the arterial pulse beats skewness values of a set of arterial pressure waveform data, (j) a parameter based on the arterial pulse beats kurtosis values of a set of arterial pressure waveform data, (k) a parameter based on the pressure weighted skewness of a set of arterial pressure waveform data, (l) a parameter based on the pressure weighted kurtosis of a set of arterial pressure waveform data, and (m) a parameter based on the pressure dependent Windkessel compliance of a set of arterial pressure waveform data.

21. The method of claim 19, wherein the one or more parameters affected by the vascular condition include (a) a parameter based on the pulse beats standard deviation of a set of arterial pressure waveform data, (b) a parameter based on the R-to-R interval of a set of arterial pressure waveform data, (c) a parameter based on the area under the systolic portion of a set of arterial pressure waveform data, (d) a parameter based on the duration of systole of a set of arterial pressure waveform data, (e) a parameter based on the duration of the diastole of a set of arterial pressure waveform data, (f) a parameter based on the mean arterial pressure of a set of arterial pressure waveform data, (g) a parameter based on the pressure weighted standard deviation of a set of arterial pressure waveform data, (h) a parameter based on the pressure weighted mean of a set of arterial pressure waveform data, (i) a parameter based on the arterial pulse beats skewness values of a set of arterial pressure waveform data, (j) a parameter based on the arterial pulse beats kurtosis values of a set of arterial pressure waveform data, (k) a parameter based on the pressure weighted skewness of a set of arterial pressure waveform data, (l) a parameter based on the pressure weighted kurtosis of a set of arterial pressure waveform data, and (m) a parameter based on the pressure dependent Windkessel compliance of a set of arterial pressure waveform data.

22. The method of claim 18, wherein the first established value is greater than the second established value.

23. The method of claim 18, wherein the first established value is a positive number and the second established value is a negative number.

24. The method of claim 18, wherein the first established value is +100 and the second established value is −

100.

25. The method of claim 18, wherein the threshold value is 0.

26. The method of claim 18, wherein the threshold value is the mean of the first established value and the second established value.

27. The method of claim 18, wherein the threshold value is a threshold range, and if the cardiovascular parameter is within the threshold range the result is indeterminate and the method is repeated using additional arterial pressure waveform data from the subject.

28. The method of claim 18, wherein the multivariate statistical model is created using the following steps:

determining an approximating function relating a set of clinically determined reference measurements of a model output value that corresponds to a first established value for the arterial pressure waveforms of the first set of arterial pressure waveform data and a second established value for the arterial pressure waveform of the second set of arterial pressure waveform data, the output values representing clinical measurements of the cardiovascular parameter from both subjects not experiencing the vascular condition and subjects experiencing the vascular condition, the approximating function being a function of one or more of the following parameters (a) a parameter based on the pulse beats standard deviation of a set of arterial pressure waveform data, (b) a parameter based on the R-to-R interval of a set of arterial pressure waveform data, (c) a parameter based on the area under the systolic portion of a set of arterial pressure waveform data, (d) a parameter based on the duration of systole of a set of arterial pressure waveform data, (e) a parameter based on the duration of the diastole of a set of arterial pressure waveform data, (f) a parameter based on the mean arterial pressure of a set of arterial pressure waveform data, (g) a parameter based on the pressure weighted standard deviation of a set of arterial pressure waveform data, (h) a parameter based on the pressure weighted mean of a set of arterial pressure waveform data, (i) a parameter based on the arterial pulse beats skewness values of a set of arterial pressure waveform data, (j) a parameter based on the arterial pulse beats kurtosis values of a set of arterial pressure waveform data, (k) a parameter based on the pressure weighted skewness of a set of arterial pressure waveform data, (l) a parameter based on the pressure weighted kurtosis of a set of arterial pressure waveform data, and (m) a parameter based on the pressure dependent Windkessel compliance of a set of arterial pressure waveform data;

determining a set of arterial blood pressure parameters from the arterial blood pressure waveform data, the set of arterial blood pressure parameters including one or more of the following parameters(a) a parameter based on the pulse beats standard deviation of a set of arterial pressure waveform data, (b) a parameter based on the R-to-R interval of a set of arterial pressure waveform data, (c) a parameter based on the area under the systolic portion of a set of arterial pressure waveform data, (d) a parameter based on the duration of systole of a set of arterial pressure waveform data, (e) a parameter based on the duration of the diastole of a set of arterial pressure waveform data, (f) a parameter based on the mean arterial pressure of a set of arterial pressure waveform data, (g) a parameter based on the pressure weighted standard deviation of a set of arterial pressure waveform data, (h) a parameter based on the pressure weighted mean of a set of arterial pressure waveform data, (i) a parameter based on the arterial pulse beats skewness values of a set of arterial pressure waveform data, (j) a parameter based on the arterial pulse beats kurtosis values of a set of arterial pressure waveform data, (k) a parameter based on the pressure weighted skewness of a set of arterial pressure waveform data, (l) a parameter based on the pressure weighted kurtosis of a set of arterial pressure waveform data, and (m) a parameter based on the pressure dependent Windkessel compliance of a set of arterial pressure waveform data; and

estimating the model output value by evaluating the approximating function with the set of the parameters.

29. The method of claim 18, wherein the arterial pressure waveform data from the subject is continuously provided and continuously analyzed.

30. The method of claim 18, further comprising alerting a user when the vascular condition is detected.

31. The method of claim 30, wherein the user is alerted by publishing a notice on a graphical user interface.

32. The method of claim 30, wherein a user is alerted by emitting a sound