METHOD AND APPARATUS FOR PROCESSING A CYCLIC PHYSIOLOGICAL SIGNAL

• US 20120302900A1
• Filed: 02/07/2011
• Published: 11/29/2012
• Est. Priority Date: 02/12/2010
• Status: Abandoned Application

• Current Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Original Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Inventors: Yin, Bin, Duric, Haris, Morren, Geert Guy Georges, Muhlsteff, Jens, Fokkenrood, Steven Antonie Willem
• Application Number:
  US13/576,687
• Publication Number:
  US 20120302900A1
• Time in Patent Office: Days
• US Class Current: 600/484
• CPC Class Codes: A61B 2562/0219 Inertial sensors, e.g. accelerometers, gyroscopes, tilt switches
• 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 5/024 Detecting, measuring or recording pulse rate or heart rate A61B5/0205, A61B5/021 take precedence
• CPC Class Codes: A61B 5/0255 Recording instruments specially adapted therefor
• CPC Class Codes: A61B 5/0816 Measuring devices for examining respiratory frequency measuring frequency of electric signals G01R23/00
• CPC Class Codes: A61B 5/113 occurring during breathing
• 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 5/7221 Determining signal validity, reliability or quality A61B5/7207 and A61B5/7217 take precedence
• 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: 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: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

The invention relates to a method and apparatus for processing a cyclic physiological signal (30, 40, 52, 53, 54). The method comprises the steps of repeatedly collecting (2) the physiological signal (30, 40, 52, 53, 54) over a time period (31, 32, 33) covering two or more cycles of the cyclic physiological signal (30, 40, 52, 53, 54), wherein a next time period (31, 32, 33) is adjacent to or overlaps with a previous time period (31, 32, 33), extracting values (3, 13) of a set of predefined parameters from the physiological signal (30, 40, 52, 53, 54) within each time period (31, 32, 33) which parameter values characterize the physiological signal (30, 40, 52, 53, 54) within the time period (31, 32, 33), and classifying (4, 14) the physiological signal (30, 40, 52, 53, 54) within each time period (31, 32, 33) based upon the extracted set of predefined parameter values. This provides for an efficient analysis of a cyclic physiological signal which is especially suitable for continuous monitoring of patients where a trend of a reliable physiological signal is more important than an instantaneous measurement of a reliable physiological signal, such as in a general ward environment of a hospital and/or in a home environment.

Claims

1. A method of processing a cyclic physiological signal, the method comprising the steps of:

repeatedly collecting the physiological signal over a time period covering two or more cycles of the cyclic physiological signal, wherein a next time period is adjacent to or overlaps with a previous time period;

extracting values of a set of predefined parameters from the physiological signal within each time period which parameter values characterize the physiological signal within the time period; and

classifying the physiological signal within each time period based upon the extracted set of predefined parameter values.

2. The method as defined in claim 1, further including a step of extracting the frequency of the classified physiological signal for each time period, after the step of classifying the physiological signal.

3. The method as defined in claim 2, further including a step of calculating a confidence value for the extracted frequency of the physiological signal within each time period after the step of extracting the frequency and based upon the values of the set of predefined parameters extracted from the physiological signal in the corresponding time period and the extracted frequency value.

4. The method as defined in claim 1, wherein the step of classifying labeling each time period as accept or as reject.

5. The method as defined in claim 2, wherein the frequency of the physiological signal is not extracted for each time period that is labeled as reject.

6. The method as defined in claim 1, wherein the physiological signal represents a respiration and/or pulse of a patient.

7. The method as defined in claim 1, wherein the predefined parameters characterize the physiological signal as a function of time, frequency and/or position in space.

8. The method as defined in claim 1, wherein in the case of an overlap between the next and previous time period, the step of classifying is performed only for the part of the next time period that does not overlap with the previous time period.

9. The method as defined in claim 1, wherein the step of classifying is also based upon a statistical database of the set of predefined parameters.

10. The method as defined in claim 1, further including a step of conditioning the physiological signal before the step of repeatedly collecting the physiological signal, wherein the step of repeatedly collecting the physiological signal comprises repeatedly collecting the conditioned signal.

11. The method as defined in claim 1, wherein multiple physiological signals are collected simultaneously and wherein the steps of extracting and classifying are performed for each of the multiple physiological signals separately.

12. An apparatus for monitoring a cyclic physiological signal, the apparatus comprising:

a sensor adapted for measuring the cyclic physiological signal;

a time period definition unit adapted for repeatedly defining a time period covering one or more cycles of the cyclic physiological signal in which time period the physiological signal is analyzed, wherein a next time period is adjacent to or overlaps with a previous time period;

an extraction unit adapted for extracting values of a set of predefined parameters from the physiological signal within the time period which parameters characterize the physiological signal within the time period; and

a classifying unit adapted for classifying the physiological signal within the time period based upon the extracted set of predefined parameter values.

13. The apparatus as defined in claim 12, wherein the sensor comprises a multi-axial accelerometer and wherein the measured physiological signal comprises multiple sub-signals corresponding to the multiple axes and which sub-signals are analyzed synchronously in time.

14. The apparatus as defined in claim 12, further comprising a frequency determination unit for determining the value of the frequency of the classified physiological signal.

15. The apparatus as defined in claim 12, wherein the apparatus comprises multiple sensors and is suitable for measuring multiple physiological signals each of which is analyzed either separately or in combination.