Method of and system for signal separation during multivariate physiological monitoring

• US 7,572,231 B2
• Filed: 12/23/2004
• Issued: 08/11/2009
• Est. Priority Date: 01/31/2000
• Status: Active Grant

• Current Assignee: Justin D. Pearlman
• Original Assignee: Justin D. Pearlman
• Inventors: Pearlman, Justin D.
• Primary Examiner(s): EVANISKO, GEORGE ROBERT
• Application Number:
  US11/020,927
• Publication Number:
  US 20050197586A1
• Time in Patent Office: 1,692 Days
• Field of Search: 600/508, 600/509, 600/513
• US Class Current: 600/508
• CPC Class Codes: A61B 5/055 involving electronic [EMR] or nuclear [NMR] magnetic resonance, e.g. magnetic resonance imaging
• CPC Class Codes: A61B 5/327 Generation of artificial ECG signals based on measured signals, e.g. to compensate for missing leads
• CPC Class Codes: A61B 5/35 by template matching
• CPC Class Codes: A61B 5/7239 using differentiation including higher order derivatives
• CPC Class Codes: A61B 5/7267 involving training the classification device
• CPC Class Codes: A61B 5/7285 for synchronising or triggering a physiological measurement or image acquisition with a physiological event or waveform, e.g. an ECG signal A61B6/5205, A61B6/541, A61B8/5207, A61B8/543 take precedence
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients

Abstract

Multiple electrode contacts make electrical connections to the anterior and/or posterior chest for multivariate characterization of the electrical activation of the heart. A central processing unit derives synthetic composite electrographic signals as well as flag signals for specific purposes. A preferred embodiment uses this system to trigger or gate magnetic resonance imaging, eliminating or reducing problems from small or inverted R-waves, lead detachment, noise, flow signal, gradient changes, and rhythm changes, more reliably flagging the onset of electrical activation of the ventricles. Additional derived data are ST-segment shifts, filling times, and respiratory cycle. Filling times may be used for greatly improved imaging in the presence of rhythm disturbances, such as atrial fibrillation. Respiratory cycle may be used as a respiratory trigger to control for the effects of breathing on the heart position and image quality.

Claims

1. A method for determining a condition associated with a live body organ by separating desired physiologic signals produced by the live body organ from superimposed signals acquired during physiologic monitoring via a plurality of physiologic signal sensors at different positions in relation to the live body organ, comprising the steps of:

acquiring multivariate physiologic signals via the plurality signal sensors, wherein the physiologic signals acquired on each of the signal sensors reflects the respective sensitivity of the particular signal sensor to multiple signal sources, including a desired signal and one or more superimposed contaminant signals;

inputting the acquired physiologic signals into a data processor that performs the steps of;

converting the acquired physiologic signals into signal data upon which mathematical operations are performed by representing the acquired signal data as an observation vector O(t) whose values are representative of the physiologic signals acquired by the corresponding sensors;

accessing a separability operator S that includes a matrix of separation coefficients;

applying the separability operator S by dot product to the observation vector O(t) to produce an output;

extracting one or more output signals from the output to determine a condition associated with the live body organ; and

enabling the physiological monitoring of the condition associated with the live body organ;

wherein the matrix of separation coefficients collectively specify a model of the conditions of physiologic and external signal sources encountered during data acquisition.

2. The method of claim 1, wherein the live body organ is a heart.

3. The method of claim 1, further comprising training the separability operator S for a particular physiologic monitoring arrangement, through the steps of:

selecting a template signal T(t) associated with a training sensor as a measure of success;

acquiring a new observation O(t) from the training sensor;

initializing the separability operator S to be an i x j matrix, wherein the components of Sixj reflect the sensitivity of the i sensors to j signal sources, where i represents a template signal data coefficient associated with the i^th lead, and j represents the particular sensor;

redefining the coefficients of S by obtaining a result signal R(t) by dot product of S.O(t), and then adjusting the coefficients of S based on a minimization of an error function calculated as the difference between a channel of the result signal R(t) and the template signal T(t),

$E_1=\underset{t}{\sum <br><br>}<br><br>{\left(R_1<br><br>\left(t\right)-T_1<br><br>\left(t\right)\right)}^2.$

4. The method of claim 3, wherein the template signal T(t) is a resting multi-lead EKG signal.

5. The method of claim 3, wherein

the separability operator S comprises a 2×

2 matrix;

$O<br><br>\left(t\right)=\left(\begin{array}{c}a<br><br>\left(t\right)\\ p<br><br>\left(t\right)\end{array}\right)$

wherein a(t) is the observed signal at an anteriorly positioned sensor, and p(t) is the observed signal at a posteriorly positioned sensor;

the template signal T(t) comprises EKG v1 for the anteriorly positioned sensor, and a signal representing great vessel flow for the posteriorly positioned sensor; and

the error function comprises

$E_1=\underset{t}{\sum <br><br>}<br><br>{\left(R_1<br><br>\left(t\right)-\mathrm{EKGv}<br><br><br><br>1<br><br>\left(t\right)\right)}^2.$

6. The method of claim 3, wherein:

multiple template signals T(t) are selected; and

the error function comprises a sum of independent error terms,

$E=\underset{i}{\sum <br><br>}<br><br>E_i,$

 

where E_i is error function for component i.

7. The method of claim 3, further comprising the step of correcting for DC offsets in each sensor through the steps of:

expanding the operator S to support an additional degree of freedom; and

appending the value 1 to the observation vector O(t).

8. The method of claim 1, wherein the desired separate physiologic signals represent contributions to overall signal from different regions and the signals from other regions are treated as contaminants with respect to each desired signal.

9. The method of claim 1, wherein the deriving step further comprises the step of:

extracting from the acquired multivariate physiologic signals undesirable signals attributable to distal, global and external signal sources through comparison of the undesirable signals acquired on two or more of the signal channels so as to isolate a local cardiac signal.

10. The method of claim 1, further comprising the step of:

externally generating one or more signals containing information relative to external undesirable signal sources.

11. The method of claim 1, wherein:

each multivariate physiologic signal is coupled to one of the signal sensors by a plurality of electrically conducting leads of similar trajectory not all of which are in electrical contact with the signal sensor, such that a common spurious signal is acquired in each lead attributable to one of the multiple signal sources that is external to the body; and

the deriving step further comprises the step of extracting the common signal from the acquired multivariate physiologic signals.

12. The method of claim 1, further comprising the step of:

generating from the output signal a synthetic ECG signal including indications of cardiac electrical activity.

13. A system for determining a condition associated with a live body organ by separating desired physiologic signals produced by the live body organ from superimposed contaminant signals acquired during physiologic monitoring, comprising:

a plurality of sensors adapted to be located at different positions relative to a live body organ for acquiring multivariate physiologic signals, wherein the physiologic signals acquired on each of the sensors reflects the respective sensitivity of the particular signal sensor to multiple signal sources, including a desired signal and one or more superimposed signals;

a separability operator that includes a matrix of separation coefficients;

a data processor in data communication with the plurality of sensors and executing the steps of;

converting the acquired physiologic signals into signal data upon which mathematical operations are performed by representing the acquired signal data as an observation vector O(t) whose values are representative of the physiologic signals acquired by the corresponding sensors;

applying the separability operator S by dot product to the observation vector O(t) to produce an output;

extracting one or more output signals from the output to determine a condition associated with the live body organ; and

enabling the physiological monitoring of the condition associated with the live body organ;

wherein the matrix of separation coefficients collectively specify a model of the conditions of physiologic and external signal sources encountered during data acquisition.

14. The system of claim 13, wherein the data processor further executes steps for training the separability operator S for a particular physiologic monitoring arrangement comprising:

selecting a template signal T(t) associated with a training sensor as a measure of success;

acquiring a new observation O(t) from the training sensor;

initializing the separability operator S to be an i x j matrix initialized to the identity matrix, wherein the components of Sixj represent the sensitivity of the i sensors to j signal sources, where i represents a template signal data coefficient associated with the i^th lead, and j represents the particular sensor;

redefining the coefficients of S by obtaining a result signal R(t) by dot product of S.O(t), and then adjusting the coefficients of S based on a minimization of an error function calculated as the difference between a channel of the result signal R(t) and the template signal T(t),

$E_1=\underset{t}{\sum <br><br>}<br><br>{\left(R_1<br><br>\left(t\right)-T_1<br><br>\left(t\right)\right)}^2.$

15. The system of claim 14, wherein the template signal T(t) is a resting multi-lead EKG signal.

16. The system of claim 14, wherein:

the separability operator S comprises a 2×

2 matrix;

$O<br><br>\left(t\right)=\left[\begin{array}{c}a<br><br>\left(t\right)\\ p<br><br>\left(t\right)\end{array}\right],$

 

wherein a(t) is the observed signal at an anteriorly positioned sensor, and p(t) is the observed signal at a posteriorly positioned sensor;

the template signal T(t) comprises EKG v1 for the anteriorly positioned sensor, and a signal representing great vessel flow for the posteriorly positioned sensor; and

the error function comprises

$E_1=\underset{t}{\sum <br><br>}<br><br><br><br>{\left(R_1<br><br>\left(t\right)-\mathrm{EKG}<br><br><br><br>v<br><br><br><br>1<br><br>\left(t\right)\right)}^2.$

17. The system of claim 14, wherein:

the data processor executes a step of selecting multiple template signals T(t); and

the error function comprises a sum of independent error terms,

$E=\underset{i}{\sum <br><br>}<br><br><br><br>E_i,$

 

where E_i is error function for component i.

18. The system pf claim 14, wherein the data processor executes steps correcting for DC offsets in each sensor comprising:

expanding the operator S to support an additional degree of freedom; and

appending the value 1 to the observation vector O(t).