Multivariable artifact assessment

• US 6,287,328 B1
• Filed: 04/08/1999
• Issued: 09/11/2001
• Est. Priority Date: 04/08/1999
• Status: Expired due to Term

• Current Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Original Assignee: Agilent Technologies, Inc.
• Inventors: Lyster, Thomas D, Snyder, David E
• Primary Examiner(s): Layno, Carl H.
• Application Number:
  US09/288,584
• Time in Patent Office: 887 Days
• Field of Search: 607/509, 607/513
• US Class Current: 600/509
• CPC Class Codes: A61B 5/30 Input circuits therefor
• CPC Class Codes: A61B 5/7203 for noise prevention, reduction or removal
• CPC Class Codes: A61B 5/721 using a separate sensor to detect motion or using motion information derived from signals other than the physiological signal to be measured
• CPC Class Codes: A61B 5/7217 of noise originating from a therapeutic or surgical apparatus, e.g. from a pacemaker
• CPC Class Codes: A61B 5/7257 using Fourier transforms
• CPC Class Codes: A61B 5/726 using Wavelet transforms
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: A61N 1/3904 External heart defibrillators [EHD]

Abstract

An apparatus detects corruption of an event signal by two or more non-event artifacts by using multivariable artifact assessment. The apparatus comprises: 1) a sensor for providing an input signal comprising an event signal coexisting with two or more non-event signals, 2) a measurement system including a receiver for receiving the input signal from the sensor and separating the received input signal into its constituent parts, 3) detectors for detecting both the event signal and non-event signals from the received input signal, and 4) an inference processor for analyzing the correlation signals to produce an indication of signal corruption. In a second embodiment, the apparatus comprises: 1) a plurality of sensors for providing additional non-event input signals, 2) a corresponding measurement system having a plurality of receivers for receiving the additional input signals, 3) detectors for detecting the non-event signals from the received additional input signals, 4) correlators for comparing the detected event signal with each of the detected non-event signals to produce respective correlation signals, and 5) an inference processor for analyzing the correlation signals to produce an indication of signal corruption. The apparatus may be a patient monitoring system or a defibrillator. Methods of operating the apparatus are also described.

Claims

1. An apparatus for detecting corruption of an event signal with artifact in an environment where the potentially corrupted event signal coexists with two or more non-event signals which may be related to the artifact, the apparatus comprising:

a sensor for providing an input signal, wherein the input signal comprises an event signal coexisting with two or more non-event signals;

a measurement system for receiving the input signal from the sensor and separating the received input signal into its constituent parts, wherein the measurement system comprises a receiver for receiving the input signal, a detector for detecting the event signal from the received input signal, and detectors for detecting the non-event signals from the received input signal;

correlators for comparing the detected event signal with each of the detected non-event signals to produce respective correlation signals; and

an inference processor for analyzing the correlation signals to produce an indication of signal corruption.

2. The apparatus of claim 1 wherein the inference processor analyzes the correlation signals to determine if the event signal is corrupted.

3. The apparatus of claim 1 wherein the inference processor analyzes the correlation signals to determine a degree of corruption of the event signal.

4. The apparatus of claim 1 wherein the inference processor is selected from the group consisting of:

peak detectors, polynomial regressions, fuzzy set operations, neural networks, probability estimators, and statistical classifiers.

5. The apparatus of claim 1 wherein the coexistent signals are selected from the group consisting of:

differential voltage, impedance, and common-mode current.

6. The apparatus of claim 1 wherein the apparatus has a plurality of signal processors for processing the event and non-event signals prior to correlation.

7. The apparatus of claim 6 wherein the signal processors are selected from the group consisting of linear and non-linear transforms.

8. The apparatus of claim 7 wherein the signal processors are selected from the group consisting of:

filters, Fourier transforms, wavelet transforms, and joint time-frequency spectrograms.

9. An apparatus for detecting corruption of an event signal with artifact in an environment where the potentially corrupted event signal and non-event signals which may be related to the artifact exist independently, the apparatus comprising:

a sensor for providing a first input signal, wherein the first input signal comprises the event signal;

a plurality of additional sensors for providing a plurality of additional input signals, wherein the additional input signals comprise non-event signals;

a measurement system for receiving the input signals from the sensors and detecting the event and non-event signals, wherein the measurement system comprises a receiver for receiving the first input signal, a detector for detecting an event signal from the received first input signal, receivers for receiving the additional input signals, and detectors for detecting the non-event signals from the received additional input signals;

correlators for comparing the detected event signal with each of the detected non-event signals to produce respective correlation signals; and

an inference processor for analyzing the correlation signals to produce an indication of signal corruption.

10. The apparatus of claim 9 wherein the inference processor analyzes the correlation signals to determine if the event signal is corrupted.

11. The apparatus of claim 9 wherein the inference processor analyzes the correlation signals to determine a degree of corruption of the event signal.

12. The apparatus of claim 9 wherein the inference processor is selected from the group consisting of:

peak detectors, polynomial regressions, fuzzy set operations, neural networks, probability estimators, and statistical classifiers.

13. The apparatus of claim 9 wherein the apparatus has a plurality of signal processors for processing the event and non-event signals prior to correlation.

14. The apparatus of claim 13 wherein the signal processors are selected from the group consisting of linear and non-linear transforms.

15. The apparatus of claim 13 wherein the signal processors are selected from the group consisting of:

filters, Fourier transforms, wavelet transforms, and joint time-frequency spectrograms.

16. The apparatus of claim 9 in an environment where the input signals may have additional coexistent non-event signals, wherein the measurement system additionally separates the input signals from the sensors into their constituent parts providing additional non-event signals.

17. In an apparatus for detecting corruption of an event signal with artifact in an environment where the potentially corrupted event signal coexists with two or more non-event signals which may be related to the artifact, a method for determining the integrity of the event signal comprising:

(a) sensing an input signal;

(b) detecting the input signal by a measurement system;

(c) separating the input signal into constituent parts, comprising an event signal and two or more non-event signals;

(d) correlating the resulting event signal with each of the resulting non-event signals to produce respective correlation signals;

(e) providing the correlation signals to an inference processor; and

(f) determining whether the event signal can be used with confidence.

18. The method of claim 17 wherein the inference processor examines the correlation signals to determine if the event signal is corrupted.

19. The method of claim 17 wherein the inference processor analyzes the correlation signals to determine a degree of corruption of the event signal.

20. The method of claim 17 wherein step (b) is repeated a plurality of times to provide more than one input to the device.

21. In an apparatus for detecting corruption of an event signal with artifact in an environment where the potentially corrupted event signal and non-event signals which may be related to the artifact exist independently, a method for determining the integrity of the event signal comprising:

(a) sensing a plurality of input signals with a plurality of sensors;

(b) detecting an event signal from one input signal;

(c) detecting a non-event signal from an additional input signal;

(d) correlating the resulting event signal with each non-event signal to produce respective correlation signals;

(e) providing the correlation signals to an inference processor;

(f) determining whether the event signal can be used with confidence; and

(g) repeating step (c) a plurality of times to provide more than one non-event signal to the device.

22. The method of claim 21 wherein a detected input signal is further separated into its constituent parts to provide additional non-event signals to the device.

23. The method of claim 21 wherein the inference processor examines the correlation signals to determine if the event signal is corrupted.

24. The method of claim 21 wherein the inference processor analyzes the correlation signals to determine a degree of corruption of the event signal.