Method and apparatus for determining the state of an oximetry sensor

• US 6,035,223 A
• Filed: 11/19/1997
• Issued: 03/07/2000
• Est. Priority Date: 11/19/1997
• Status: Expired due to Term

• Current Assignee: Covidien LP (Medtronic Plc)
• Original Assignee: Nellcor Puritan Bennett Incorporated (Medtronic Plc)
• Inventors: Baker, Clark R. Jr.
• Primary Examiner(s): Wincker, Eric F.
• Application Number:
  US08/974,274
• Time in Patent Office: 839 Days
• Field of Search: 600/310, 600/322, 600/323, 600/331, 600/336, 600/300, 600/301, 600/309, 356/41, 128/920, 128/925
• US Class Current: 600/323
• CPC Class Codes: A61B 5/0059 using light, e.g. diagnosis by transillumination, diascopy, fluorescence A61B5/0093, A61B5/1455, A61B5/02416 and A61B5/0261 take precedence
• CPC Class Codes: A61B 5/14551 for measuring blood gases
• CPC Class Codes: A61B 5/6843 Monitoring or controlling sensor contact pressure
• 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

Abstract

A method for determining whether a sensor is coupled to a tissue sample is described. The sensor is operable to generate a detector signal indicative of absorption of electromagnetic radiation by the tissue sample. A plurality of metric values corresponding to the detector signal are generated. A neural net is employed with the metric values to determine whether the sensor is coupled to the tissue sample.

Claims

1. A method for determining whether a sensor is coupled to a tissue sample, the sensor being operable to generate a detector signal indicative of absorption of electromagnetic radiation by the tissue sample, the method comprising:

generating a plurality of metric values corresponding to the detector signal, each of the metric values guantifying a behavior of the detector signal which is correlated with whether the sensor is coupled to the tissue sample;

inputting the plurality of metric values to a neural net,

performing data reduction with the neural net based on dependencies among the metric values, and generating a probability value indicative of whether the sensor is coupled to the tissue sample; and

determining whether the sensor is coupled to the tissue sample by referring to the probability value.

2. The method of claim 1 wherein generating the plurality of metric values comprises:

performing analog signal processing on the detector signal;

after the analog signal processing, converting the detector signal to a digital signal;

generating AC and DC components of the digital signal corresponding to AC and DC components of the detector signal; and

generating the plurality of metric values using the AC and DC components of the digital signal.

3. The method of claim 2 further comprising taking the natural logarithm of the digital signal.

4. The method of claim 2 further comprising bandpass filtering the digital signal.

5. The method of claim 2 further comprising normalizing the digital signal.

6. The method of claim 2 wherein the detector signal comprises a first portion corresponding to electromagnetic radiation of a first frequency, and a second portion corresponding to electromagnetic radiation of a second frequency, selected ones of the metric values being generated using AC and DC components of the digital signal corresponding to the first portion of the detector signal.

7. The method of claim 1 wherein the detector signal comprises an infrared portion corresponding to infrared radiation and a red portion corresponding to red radiation, and wherein generating the plurality of metric values comprises generating infrared AC and DC component signals corresponding to the infrared portion of the detector signal, and red AC and DC component signals corresponding to the red portion of the detector signal.

8. The method of claim 7 wherein the plurality of metric values comprises the average amplitude of the infrared AC component signal.

9. The method of claim 8 wherein the average amplitude is determined using an infinite impulse response filter.

10. The method of claim 7 wherein the plurality of metric values comprises the variability of the amplitude of the infrared AC component signal.

11. The method of claim 7 wherein the plurality of metric values comprises the degree to which the infrared and red portions of the detector signal are correlated.

12. The method of claim 11 wherein the degree to which the infrared and red portions of the detector signal are correlated is determined with reference to a digitized and normalized portion of the detector signal.

13. The method of claim 7 wherein the plurality of metric values comprises the variability of the amplitude of the infrared DC component signal.

14. The method of claim 7 wherein the plurality of metric values comprises the slope of the infrared DC component signal.

15. The method of claim 7 wherein the plurality of metric values comprises a pulse shape metric which is reflective of a pulse shape associated with the infrared portion of the detector signal.

16. The method of claim 7 wherein the plurality of metric values comprises a maximum autocorrelation value corresponding to a digitized and normalized portion of the detector signal.

17. The method of claim 1 further comprising training the neural net to determine whether the sensor is coupled to the tissue sample using a database of previously recorded sensor data.

18. The method of claim 17 wherein the neural net comprises a plurality of processing nodes each having at least one coefficient associated therewith, and wherein training the neural net comprises determining values for each of the coefficients.

19. The method of claim 18 wherein training the neural net comprises mapping an input space defined by the plurality of metric values.

20. The method of claim 19 wherein the plurality of metric values are explicitly constrained during the mapping of the input space.

21. At least one computer readable medium containing program instructions for determining whether a sensor is coupled to a tissue sample, the sensor being operable to generate a detector signal indicative of absorption of electromagnetic radiation by the tissue sample, said at least one computer readable medium comprising:

computer readable code for generating a plurality of metric values corresponding to the detector signal, each of the metric values quantifying a behavior of the detector signal which is correlated with whether the sensor is coupled to the tissue sample;

computer readable code for inputting the plurality of metric values to a neural net;

computer readable code for performing data reduction with the neural net based on dependencies among the metric values, and generating a probability value indicative of whether the sensor is coupled to the tissue sample; and

computer readable code for determining whether the sensor is coupled to the tissue sample by referring to the probability value.

22. A method for determining whether a sensor is coupled to a tissue sample, the sensor being operable to generate a detector signal indicative of a parameter associated with the tissue sample, the method comprising:

generating a plurality of metric values corresponding to the detector signal, each of the metric values quantifying a behavior of the detector signal which is correlated with whether the sensor is coupled to the tissue sample;

inputting the plurality of metric values to a filter;

performing data reduction with the filter based on dependencies among the metric values, and generating a probability value indicative of whether the sensor is coupled to the tissue sample; and

determining whether the sensor is coupled to the tissue sample by referring to the probability value.