Blood constituent measuring device and method

Claims

What is claimed is:

1. A blood constituent measuring device for measuring a constituent of blood in a person's body, said device comprising:

means for sensing electromagnetic energy passing through a portion of the body at a plurality of wavelengths and for producing, for each wavelength, a respective electrical signal comprising a pulsatile component and a non-pulsatile component, wherein said sensing and producing means produces each of the signals in response to the electromagnetic energy received at the respective wavelength;

means for subtracting and storing at least a portion of the non pulsatile component from the signal for each wavelength;

means for processing the pulsatile component of the signal for each wavelength and for computing the amount of the blood constituent as a function of the processed pulsatile component of each signal and the stored portion of the non pulsatile component of each signal; and

means for amplifying, with a controllable gain, the pulsatile component of each signal after at least a portion of the non pulsatile component is subtracted from the respective signal, wherein said amplifying means amplifies the signals, after subtraction by said subtracting and storing means, to a sufficient extent that the amplified subtracted output signals are within a predetermined sensitivity range of said processing and computing means.

2. The device of claim 1, wherein said subtracting and storing means comprises a digital-to analog converter.

3. The device of claim 1, further comprising second amplifying means for amplifying the signals before said subtracting and storing means subtracts the portion of the non pulsatile components from the signals.

4. The device of claim 1, wherein said amplifying means amplifies with a controllable gain which is selected from among a plurality of predetermined values.

5. The device of claim 3, wherein each of said amplifying means amplifies with a respective controllable gain, each of said gains being selected, independently of each other, from among a respective plurality of predetermined values.

6. The device of claim 3, further comprising:

electromagnetic energy emitting means for emitting electromagnetic energy at each of the wavelengths through the body portion seriatim, thereby to produce multiplexed electromagnetic energy and wherein said sensing and producing means comprises a photodetector for converting the electromagnetic energy transmitted through the body portion into analog electrical current signals and a current-to-voltage converter for converting the analog electrical current signals into analog voltage signals whose voltage varies with time;

a demultiplexer for demultiplexing the signals produced by said current to voltage converter so as to produce two separate signals in first and second channels, respectively, representing electromagnetic energy of different wavelengths from said electromagnetic energy emitting means passing through the body portion;

two low pass filters, each connected to said demultiplexer through a respective one of said channels, wherein each said filter receives a respective one of the two separate signals; and

a remultiplexer connected to both channels for remultiplexing the two separat signals after filtering by said filters, wherein the output from said multiplexer is received by said means for amplifying the signals before subtraction of a portion of the non-pulsatile component; and

wherein said digital to analog converter also converts the pulsatile component of the demultiplexed, subtracted and amplified signals into digital output signals.

7. The device of claim 6, wherein said filters are frequency-matched to each other.

8. The device of claim 6, further comprising a sequencer for controlling said current-to voltage converter, and wherein said processing and computing means comprises a microprocessor subsystem, distinct and separate from said sequencer, which microprocessor subsystem controls said sequencer.

9. The device of claim 6, further comprising means for preventing an electrosurgical unit being used on the body from interfering with the operation of said photodetector, wherein said interference preventing means comprises a partially transparent window, to be positioned between said photodetector and the body portion.

10. The device of claim 6, wherein said electromagnetic energy emitting means comprises first and second emitters for emitting electromagnetic energy at a first of said wavelengths and a third emitter for emitting electromagnetic radiation at a second of said wavelengths, said third emitter being disposed generally between said first and second emitters, and said three emitters being disposed sufficiently close together to ensure that the body portion through which the radiation that is sensed by said sensing means passes, receives on the average substantially equal luminance due to the energy at said first and second wavelengths.

11. The device of claim 1, further comprising means for digitizing the subtracted portion of the non pulsatile component of the signals and for computing the instantaneous value of the non pulsatile component of the signals.

12. The device of claim 11, further comprising means for digitizing the pulsatile component after the portion of the non-pulsatile component is subtracted from the signals, wherein said amplifying means amplifies the pulsatile component before digitization by the digitization means, and wherein said amplifying means amplifies the pulsatile component sufficiently that the digitized pulsatile component has a height of at least fifty times the resolution of said digitizing means.

13. The device of claim 1, wherein the portion of the non-pulsatile component varies with time in such a manner as to change the value of the pulsatile component, and wherein said processing and computing means comprises means for compensating for the change in the value of the pulsatile component due to the varying of the non pulsatile component.

14. The device of claim 13, wherein the pulsatile component comprises first and second pulses, wherein the non pulsatile component causes the pulsatile component to vary at a drift rate, and wherein said compensation means comprises means for calculating the drift rate by the following formula:

[(1/2)(Max+Min).sub.1 -(1/2)(Max+Min).sub.2 ]/.DELTA.T

wherein (Max+Min).sub.1 represents the sum of the maximum value and minimum value of the voltage of the first pulse, (Max+Min).sub.2 represents the sum of the maximum value and the minimum value of the voltage of the second pulse, and wherein .DELTA.T represents the time elapsed between the minimum values of the voltages of said first and second pulses.

15. The device of claim 13, wherein the pulsatile component comprises a plurality of pairs of pulses, wherein said compensation means comprises means for calculating the drift rate for each pair of pulses by the following formula:

[(1/2)(Max+Min).sub.n -(1/2)(Max+Min).sub.n+1 ]/ .DELTA.T

wherein (Max+Min).sub.n represents the sum of the maximum value and minimum value of the voltage of the nth pulse of the pulsatile component, wherein (Max+Min).sub.n+1 represents the sum of the maximum value and the minimum value of the voltage of the (n+1)st pulse of the pulsatile component, the (n+1)st pulse occurring later in time than the nth pulse, wherein .DELTA.T represents the time elapsed between the minimum values of the voltages of the nth and (n+1)st pulses, and wherein n assumes the value of each of a predetermined set of positive integers.

16. The device of claim 15, wherein said compensation means further comprises means for subtracting the product of the drift rate and the duration of the nth pulse from the average value of the voltage of the (n+1)st pulse.

17. The device of claim 16, wherein said compensation means computes the average value of the voltage of the (n+1)st pulse only during the systolic portion of the (n+1)st pulse.

18. The device of claim 16, wherein said compensation means computes the average value of the voltage of the (n+1)st pulse by adding pairs of values for the voltage of the (n+1)st pulse, which pairs are selected such that the difference between the values of the pair is greater than three-fourths of the difference between the maximum and minimum values of the voltage of the (n+1)st pulse, and dividing the resulting sum by the number of such pairs.

19. The device of claim 1, further comprising electromagnetic energy emitting means for alternately emitting red and infrared wavelengths of light, wherein said processing and computing means comprises means for computing the percentage of oxygen saturation of hemoglobin in the blood of the body.

20. The device of claim 19, wherein said processing and computing means comprises means for computing said percentage of oxygen saturation by dividing a first quotient: ##EQU12## by a second quotient: ##EQU13##

21. The device of claim 20, wherein the pulsatile component comprises a plurality of pulses, each corresponding to a pulse of the blood of the body, wherein each pulse of the pulsatile component has a voltage varying over time, and wherein said computing means computes the value of the voltage of each pulse of the pulsatile component of the red wavelength by computing, for each of a plurality of pairs of values for the voltage of one pulse wherein the difference between the values of the pair is greater than three-fourths of the difference between the maximum and minimum value of the voltage of that pulse, the quotient of their difference divided by the number of such pairs, and repeating said computing step for each pulse.

22. The device of claim 20, wherein said means for computing said percentage of oxygen saturation includes at least one look-up table for looking up said percentage of oxygen saturation as a function of the quantity obtained by dividing said first quotient by said second quotient.

23. The device of claim 22, wherein said means for computing said percentage of oxygen saturation includes at least two such look-up tables, each of said look-up tables being for use in connection with the electromagnetic energy being passed through a different respective body portion.

24. The device of claim 23, wherein one of said look-up tables is suitable for use in a case in which the electromagnetic radiation is passed through an ear lobe, and a second of said look-up tables is suitable for use in a case in which the electromagnetic reduction is passed through a finger.

25. A blood constituent measuring device for measuring at least one constituent of blood in a body, comprising:

means for sensing electromagnetic energy passing through a portion of the body at a plurality of wavelengths and for producing electrical signals comprising a pulsatile component and a non pulsatile component for each wavelength in response to the electromagnetic energy received by said sensing and producing means at a plurality of wavelengths, wherein the electrical signals comprise voltage signals whose voltage varies over time, and wherein the voltage of the non-pulsatile component varies over time in such a manner as to change the value of the voltage of the pulsatile component over time; and

means for processing the pulsatile component of the signals for each wavelength and for computing the amount of the blood constituent as a function of the processed pulsatile component and the non pulsatile component of the output signals, wherein said processing and computing means further comprises means for compensating for the change in the value of the voltage of the pulsatile component over time due to the varying of the voltage of the non pulsatile component over time.

26. The device of claim 25, wherein the non-pulsatile component varies in such a manner as to change the value of the pulsatile component linearly; and said device further comprising means for compensating for the linear change in the value of the pulsatile component due to the varying of the non-pulsatile component.

27. The device of claim 26, wherein the pulsatile component comprises first and second pulses, wherein the non-pulsatile component varies the value of the voltage of the pulsatile component over time at a predetermined drift rate, wherein said compensation means comprises means for calculating the drift rate by the following formula:

[(1/2)(Max+Min).sub.1 (1/2)(Max+Min).sub.2 ]/.DELTA.T

wherein (Max+Min).sub.1 represents the sum of the maximum value and minimum value of the voltage of the first pulse of the pulsatile component, (Max+Min).sub.2 represents the sum of the maximum value and the minimum value of the voltage of said second pulse of the pulsatile component, and .DELTA.T represents the time elapsed between the minimum values of the voltage of the first and second pulses.

28. The device of claim 26, wherein the pulsatile component comprises a plurality of pairs of pulses, wherein said compensation means comprises means for calculating the drift rate for each pair of pulses by the following formula:

[(1/2)(Max+Min).sub.n -(1/2)(Max+Min).sub.n+1 ]/.DELTA.T

wherein (Max+Min).sub.n represents the sum of the maximum value and minimum value of the voltage of the nth pulse of the pulsatile component, (Max+Min).sub.n+1 represents the sum of the maximum value and the minimum value of the voltage of the (n+1)st pulse of the pulsatile component, wherein the (n+1)st pulse occurs later in time than the nth pulse, and wherein .DELTA.T represents the time elapsed between the minimum values of the voltage of the nth and (n+1)st pulses, and wherein n is assumes the value of each of a plurality of positive integers.

29. The device of claim 28, wherein said compensation means further comprises means for correcting the (n+1)st pulse of the pulsatile component by subtracting the product of the drift rate and the duration of nth pulse from the average value of the voltage of the (n+1)st pulse.

30. The device of claim 29, wherein said compensation means computes the value of the voltage of the (n+1)st pulse of the pulsatile component only during the systolic portion of the pulse.

31. The device of claim 29, wherein said computing means computes the average value of the voltage of the (n+1)st pulse of the pulsatile component by computing, for each of a plurality of pairs of values for the voltage of the (n+1)st pulse such that the difference between the values of the members of a pair is greater than three quarters of the difference between the maximum and minimum values of the voltage of the (n+1)st pulse, the quotient of that difference divided by the number of such pairs.

32. The device of claim 25, wherein said processing and computing means includes a plurality of look-up tables which are respectively for looking up the blood constituent amount as a function of the signals, each of said look-up tables being for use in connection with the electromagnetic energy being passed through a different respective body portion.

33. A blood constituent measuring device for measuring a constituent of blood in a person's body, said device comprising:

means for sensing electromagnetic energy passing through a portion of the body at a plurality of wavelengths and for producing, for each wavelength, a respective electrical signal comprising a pulsatile component and a non-pulsatile component, wherein said sensing and producing means produces each of the signals in response to the electromagnetic energy received at the respective wavelength;

means for filtering the signals;

means for subtracting and storing at least a portion of the non-pulsatile component from the filtered signal for each wavelength;

means for processing the pulsatile component of the signal for each wavelength and for computing the amount of the blood constituent as a function of the processed pulsatile component of each signal and the stored portion of the non-pulsatile component of each signal; and

at least two means for amplifying the signals to a sufficient extent that the amplified subtracted output signals are within a predetermined sensitivity range of said processing and computing means, each of said amplifying means having a gain controllable independently of that of the other.

34. The device of claim 33, wherein each said amplifying means amplifies the signal for each wavelength, and each of said amplifying means has its gain controlled independently for each wavelength.

35. The device of claim 33, wherein one of said amplifying means amplifies the signals after said subtracting and storing means has subtracted the portion of the non-pulsatile component of each signal.

36. A blood constituent measuring device for measuring a constituent of blood in a person's body, said device comprising:

means for emitting electromagnetic energy at a plurality of wavelengths, said emitting means comprising a first emitter for emitting energy at a first of said wavelengths, and second and third emitters for emitting energy at a second of said wavelengths, said second and third emitters being disposed one to each side of said first emitter such that, on the average, substantially equal luminance due to the energy at the first and second wavelengths is incident on a body portion at a predetermined position relative to said emitting means;

means for sensing electromagnetic energy from said emitting means and passing through a portion of the body and for producing, for each wavelength, a respective electrical signal comprising a pulsatile component and a non-pulsatile component, wherein said sensing and producing means produces each of the signals in response to the electromagnetic energy received at the respective wavelength; and

means for processing the pulsatile component of the signal for each wavelength and for computing the amount of the blood constituent as a function of the pulsatile component of each signal.

37. A blood constituent measuring device for measuring a constituent of blood in a person's body, said device comprising:

electromagnetic energy emitting means operable in a monitoring mode and in a test mode, wherein, in said monitoring mode, said electromagnetic energy emitting means emits electromagnetic energy at each of N wavelengths (N an integer greater than 1) seriatim through a portion of the body during a predetermined period of time, thereby to produce multiplexed electromagnetic energy passing through the body portion; and wherein, in said test mode, said electromagnetic energy emitting means emits electromagnetic energy at only one of said wavelengths;

means for sensing the electromagnetic energy passing through the body portion and for producing during said predetemined period of time during operation in said monitoring mode, a respective electrical signal corresponding to each of said N wavelengths, wherein said sensing and producing means produces each of the signals in response to the electromagnetic energy received at the respective wavelength; and said sensing and producing means producing, during said predetermined period of time during operation in said test mode, N electrical signals in response to the electromagnetic energy received at said one wavelength; wherein each of the signals comprises a pulsatile component and a non-pulsatile components in both modes;

means for processing the pulsatile component of each signal and for computing the amount of the blood constituent as a function of the processed pulsatile component of each signal; and

means for controlling said electromagnetic energy emitting means to operate selectively in said monitoring mode or in said test mode,

wherein said processing means is so structured and arranged that, when said electromagnetic energy emitting means operates in said test mode, the amount computed by said processing means is compared to a predetermined value.

38. A method of determining the amount of at least one constituent of the blood in a person s body, comprising the steps of:

sensing a plurality of wavelengths of electromagnetic radiation passing through a portion of the body, with a sensing means;

producing voltage signals corresponding to the electromagnetic radiation sensed in said sensing step, the voltage of each signal varying with time, and wherein the voltage signals each comprise a pulsatile component and a non-pulsatile component;

subtracting and storing at least a portion of the non-pulsatile component from each voltage signal;

processing the pulsatile component of each voltage signal and computing the amount of the blood constituent as a function of the processed pulsatile components and the stored portion of the non-pulsatile components; and

amplifying the pulsatile components after at least a portion of the non pulsatile components is subtracted from the voltage signals.

39. The method of claim 38, wherein said producing step further comprises producing analog electrical signals, and wherein said subtraction step further comprises digitally subtracting and storing at least a portion of the non-pulsatile component of each of the voltage signals.

40. The method of claim 38, further comprising the step of adding a predetermined negative voltage to the voltage signals before said subtraction and storage step.

41. The method of claim 38, further comprising the step of amplifying the signals before said subtraction step and after said producing step.

42. The method of claim 39, further comprising the steps of:

emitting and transmitting electromagnetic energy at a plurality of predetermined wavelengths through the body portion seriatim to produce multiplexed electromagnetic energy transmitted through the body;

converting the multiplexed electromagnetic energy transmitted through the body portion into analog electrical current signals with a photodetector and converting the current signals into the voltage signals with a current-to-voltage converter;

demultiplexing the voltage signals so as to produce two separate signals in first and second channels representing electromagnetic energy passing through the body portion at different wavelengths;

low-pass filtering the separate signals in the first and second channels, using low pass filters connected to the demultiplexer through respective channels;

remultiplexing the two separate signals after said filtering step; and

amplifying the signals after said remultiplexing step and before said subtraction step.

43. The method of claim 42, further comprising the step of digitizing the pulsatile component of the analog voltage signals after said subtraction step.

44. The method of claim 43, further comprising the step of amplifying the pulsatile component before said digitization step and after said subtraction step, sufficiently that the digitized pulsatile component has a resolution of at least eight bits.

45. The method of claim 38, wherein the non-pulsatile component varies in such a manner as to change the value of the voltage of the pulsatile component, and further comprising the step of compensating for the change in the value of the voltage of the pulsatile component due to the varying of the non pulsatile component.

46. The method of claim 45, wherein the pulsatile component comprises first and second pulses, wherein the non-pulsatile component causes the average voltage of the pulsatile component to vary at a predetermined drift rate, and wherein said compensation step comprises the step of calculating the drift rate by the following formula:

[(1/2)(Max+Min)-(1/2)(Max+Min).sub.2 ]/.DELTA.T

wherein (Max+Min).sub.1 represents the sum of the maximum value and minimum value of the voltage of the first pulse of the pulsatile component, (Max+Min).sub.2 represents the sum of the maximum value and the minimum value of the voltage of the second pulse of the pulsatile component, and .DELTA.T represents the time elapsed between the minimum values of the voltages of the first and second pulses.

47. The method of claim 45, wherein the pulsatile componet comprises a plurality of pairs of pulses, wherein said compensation step further comprises the step of calculating the drift rate for each aair of pulses by the following formula:

[(1/2)(Max+Min).sub.n -(1/2)(Max+Min).sub.n+1 ]/.DELTA.T

wherein (Max+Min).sub.n represents the sum of the maximum value and minimum value of the voltage of the nth pulse of the pulsatile component, (Max+Min).sub.n+1 represents the sum of the maximum value and the minimum value of the voltage of the (n+1)st pulse of the pulsatile component, wherein the (n+1)st pulse occurs later in time than the nth pulse, and .DELTA.T represents the time elapsed between said minimum values of the voltages of of said nth and (n+1)st pulses, and wherein n assumes the values of each of a plurality of positive integers.

48. The method of claim 47, wherein said compensation step further comprises the step of correcting the (n+1)st pulse of the pulsatile component by subtracting the product of the drift rate and the duration of the nth pulse from the average value of the voltage of the (n+1)st pulse.

49. The method of claim 48, wherein said compensation step further comprises the step of computing the value of the voltage of the (n+1)st pulse only during the systolic portion of the pulse.

50. The method of claim 49, wherein said compensation step further comprises the step of computing the average value of the voltage of the (n+1)st pulse by computing, for each of a plurality of pairs of values for the voltage of the (n+1)st pulse which pairs are such that the difference between the members of the pair is greater than three quarters of the difference between the maximum and minimum values of the voltage of the (n+1)st pulse, the quotient of their difference divided by the number of such pairs.

51. The method of claim 38, further comprising the step of alternately emitting red and infrared wavelengths of light and transmitting the red and infrared wavelengths of light through the body portion, and wherein said processing and computing steps further comprise the step of computing the percentage of oxygen saturation in the blood of the body.

52. The method of claim 51, wherein said processing and computing steps further comprise the step of computing the percentage of oxygen saturation by dividing a first quotient: ##EQU14## by a second quotient: ##EQU15##

53. The method of claim 52, wherein the pulsatile component comprises a plurality of pulses, each corresponding to a pulse of the blood in the body, wherein each pulse of the pulsatile component has a voltage varying over time; and wherein said computing step further comprises the step of computing the value of the voltage of each pulse of the pulsatile component of the red wavelength signal by computing, for each of a plurality of pairs of values for the voltage of one pulse such that the difference between the members of one pair is greater than three quarters of the difference between the maximum and minimum values of the voltage of that one pulse, the quotient of their difference by the number of pairs of values, and repeating said computing step for each pulse.

54. The method of claim 52, wherein said step of computing the percentage of oxygen saturation comprises looking up values of oxygen saturation in a look-up table based on the quotient of said first and second quotients.

55. A method for measuring at least one constituent of the blood in a body, comprising the steps of:

sensing electromagnetic energy passing through a portion of the body at a plurality of wavelengths;

producing electrical signals comprising a pulsatile component and a non-pulsatile component for each wavelength in response to said sensing of the electromagnetic energy, wherein the voltage of the signals varies over time, and wherein the voltage of the non-pulsatile component varies over time in such a manner as to cause the average value of the voltage of the pulsatile component to change over time; and

processing the pulsatile component of the output signals for each wavelength and computing the amount of the blood constituent as a function of the processed pulsatile component and the non-pulsatile component, wherein said processing and computing step further comprises the step of compensating for the change in the value of the voltage of the pulsatile component over time due to the varying of the non-pulsatile component over time.

56. The method of claim 55, wherein the pulsatile component comprises first and second pulses, wherein the non-pulsatile component varies the value of the average voltage of the pulsatile component over time at a predetermined drift rate, and wherein said compensation step comprises the step of calculating the drift rate by the following formula:

[(1/2)(Max+Min).sub.1 -(1/2)(Max+Min).sub.2 ]/.DELTA.T

wherein (Max+Min).sub.1 represents the sum of the maximum value and minimum value of the voltage of the first pulse of the pulsatile component, (Max+Min).sub.2 represents the sum of the maximum value and the minimum value of the voltage of the second pulse of said pulsatile component, and .DELTA.T represents the time elapsed between said minimum values of the voltage of the first and second pulses.

57. The method of claim 55, wherein the pulsatile component comprises a plurality of pairs of pulses, wherein said compensation step further comprises the step of calculating the drift rate for each pair of pulses by the following formula:

[(1/2)(Max+Min).sub.n -(1/2)(Max+Min).sub.n+1 ]/.DELTA.T

where (Max+Min).sub.n represents the sum of the maximum value and minimum value of the voltage of the nth pulse of the pulsatile component, (Max+Min).sub.n+1 represents the sum of the maximum value and the minimum value of the voltage of the (n+1)st pulse of the pulsatile component, where said (n+1)st pulse oocurs later in time than said nth pulse, and .DELTA.T represents the time elapsed between said minimum values of the voltage of the nth and (n+1)st pulses, wherein n takes on the value of each of a set of positive integers.

58. The method of claim 57, wherein said compensation step further comprises the step of correcting the (n+1)st pulse of the pulsatile component by subtracting the product of the drift rate and the duration of the nth pulse from the average value of the voltage of the (n+1)st pulse.

59. The method of claim 58, wherein said compensation step further comprises the step of computing the average value of the voltage of the (n+1)st pulse only during the systolic portion of the (n+1)st pulse.

60. The method of claim 58, wherein said compensation step further comprises the step of computing the average value of the voltage of the (n+1)st pulse by computing, for each of a plurality of pairs of values for the voltage of the (n+1)st pulse such that the difference between the members of a pair is greater than three quarters of the difference between the maximum and minimum values of the voltage of the (n+1)st pulse, the quotient of their difference divided by the number of such pairs.

BACKGROUND OF THE INVENTION

1. Field of the Invention

The invention relates to a blood constituent measuring device and method, and more particularly relates to a non-invasive device and method for determining the concentration of oxygen in the blood.

2. Description of Pertinent Background Information

The well-known explosion in electronics technology over the past few decades has found many diverse areas of application. On such area is the monitoring of physiological functions. The present invention relates to such monitoring, and specifically, to the measurement of tissue oxygenation.

Monitoring oxygenation levels is desirable in the more critical areas of the hospital, espcially when a patient is being ventilated by machine. There is potential for mishap, both physiological and mechanical. Foremost examples are patients under anesthesia in the operating room, and patients in intensive/critical care units.

Two forms of electronic monitoring have gained widespread acceptance for the monitoring of oxygenation--transcutaneous monitoring of the partial pressure of oxygen, and optical monitoring of the percent hemoglobin saturation (oximetry).

Transcutaneous monitoring seeks to measure directly the partial pressure of oxygen in the tissues by measuring the oxygen which diffuses through a locally heated area of the skin. An implicit assumption of transcutaneous monitoring is good correlation between the partial pressure of diffused oxygen and the partial pressure of oxygen in the tissues. Thick and fatty skin is the Achilles' heel of this approach.

Oximetry seeks to determine the percentage of available hemoglobin in the red blood cells carrying oxygen to the tissues from the lungs. This percentage is related to the partial pressure of oxygen in the blood by the well established oxygen-disassociation curve. The higher the partial pressure, the greater is the diffusion of oxygen from the capillaries to the tissues. Thus, although oxygen saturation is not a direct measurement of the degree of tissue oxygenation, unless the cardiac output (rate at which the heart pumps blood to the body) is impaired, the two measurements will be strongly correlated.

The oximetry measurement is optical--it essentially measures how red the blood is. As most are aware from common experience, oxyhemoglobin (hemoglobin bound with oxygen) is "redder" than hemoglobin.

The method employed in such measurements is spectrophotometry. Spectrophotometry can determine the relative concentrations of N substances in a mixture by measuring the absorption by the mixture of N wavelengths of light, if the absorptions by the individual substances are sufficiently different. Mathematically, the approach amounts to solving N equations in N variables.

In the blood, hemoglobin and oxyhemoglobin are the primary substances which absorb light in the red and near-infrared region of the spectrum. Thus, two wavelengths of light (typically one red and one near-infrared are employed for maximum discrimination) are required to measure the percentage saturation (oxyhemoglobin as a percentage of total hemoglobin and oxyhemoglobin).

In vitro devices (whose use requires drawing a blood sample for measurement external to the body) have existed for a number of years. More recently, in vivo devices (which perform the measurement in blood in the body) have appeared, but these were invasive, requiring a fiber optic tube to be inserted into the bloodstream. Making a practical non-invasive device which could continuously monitor percent saturation did not await only the electronics revolution, however. There were other practical difficulties, for it is the percent saturation of the arterial blood which correlates to tissue oxygenation, and one aspect of the problem, therefore, is how to measure, non-invasively, the absorption of the arterial blood and exclude the contributions by venous blood, bone, skin, etc. One approach by Wood in the 1940's was to squeeze the earlobe to get a reading of the absorption of everything but blood, and then heat the ear to arterialize the blood which entered when the pressure was taken off. In the 1970's, Hewlett-Packard marketed a device which used eight wavelengths of light in an attempt to account for contributions from the non-blood portions of the earlobe. Use of that device also involved heating the ear to arterialize the blood. Neither of these devices were suitable for use in the operation room or intensive/critical care units: they were too large, expensive and complicated to use.

Newer devices, which are gaining widespread acceptance, are of a type called "pulse oximeters". The principle upon which they are based is simple. The light transmitted through the monitoring site (typically the finger, ear or toe), has a pulsatile component related to the extra blood pumped into the arterial vessels of the monitoring site with each heartbeat. This extra blood is arterial. Therefore, analysis of the pulsatile signal yields the percentage oxygen saturation of the arterial blood.

There is another complication related to the in vivo measurement. Strictly speaking, spectrophotometric analysis is based upon a model wiich includes pure collimated light, the intensity of which is reduced only by aborption by the mixture to be analyzed. The intensity is reduced by an exponential process known as "Beer's Law". Calculations used in in vivo measurement assume this exponential process. In non-invasive pulsatile oximetry, the light is diffused by the tissues being analyzed and the pulsatile signal received is due to scattering by the red blood cells as well as absorption by the hemoglobin and oxyhemoglobin molecules in the arterial vessels.

Fortuitously, it is found that if a "Beer's Law" type relationship is assumed, the coefficients which determine the exponential characteristic can be determined experimentally by measurement over a population of patients. Since a scattering process is involved as well as an absorption process, the coefficients are larger, and yet they are consistent enough over a population to be the basis of a useful device.

Such devices are described in U.S. Pat. Nos. 3,998,550, 4,266,554, 4,407,290 and 4,621,643. All are pulsatile oximeters and differ only by the means in whichthe signals are processed. The device of U.S. Pat. No. 3,998,550 solves the exponential Beer's Law equations by using a logarihmic circuit, while that of U.S. Pat. No. 4,266,554 takes the derivative. U.S. Pat. No. 4,407,290 recognizes that the pulse is sufficiently small to allow linerization of the equations, thus obviating the need to solve exponential equations.

While the above patents illustrate the basic principles upon which pulse oximetry is founded, and are directed to devices which are based upon these principles, all of them fail to focus upon some of the specific difficulties associated with the use of such devices in practice. It is important to recognize that these devices are typically utilized to monitor patients who are not healthy. Thus, these devices must operate under conditions of unstable physiological states and on patients who may have very weak pulses. In addition, these devices must operate from monitoring sites which exhibit a wide variation in light transmission properties.

SUMMARY OF THE INVENTION

It is an object of the present invention to provde a non-invasive oximeter capable of accurately measuring the percent oxygen saturation of arterial blood in a wide variety of patients, including patients who have very weak pulses and/or unstable physiological states.

It is another object of this invention to provide a non-invasive oximeter which will operate successfully in the presence of great amounts of electrical noise such as is generated by an electrosurgical unit (ESU) as is typically used in the operating room.

It is a further object of this invention to provide an oximeter which minimizes the number of electronic circuits required, thus making the instrument less expensive and more reliable.

It is yet a further object of this invention to provide a means for the user of any non-invasive oximeter (such as those to which U.S. Pat. No. 3,998,550, 4,266,554, 4,407,290 and 4,621,643 are directed) to perform a complete functional test of the entire system while the sensor probe is attached to the patient, thus allowing the user to have full confidence in the operation of the monitor at any time.

There is a need for a pulsatile oximeter which meets these objects.

The present invention, as do the devices of the referenced patents, comprises means for sensing electromagnetic energy of at least two wavelengths as it passes through a portion of a patient's body, processes the signals so produced so as to separate out a pulsatile portion of each signal which is related to the physiological pulse, and then determines the percent saturation as a function of the relative sizes of the pulsatile and non-pulsatile components.

According to one aspect of the present invention, in processing the signals to separate out the pulsatile component, a number of discrete gains are used to compensate for variations in the total amount of electromagnetic energy received due to variation in the strength of the emitting source, the thickness of the portion of the body through which the electromagnetic energy is being sent, and placement of the detector of energy with respect to the emitters. A digital-to-analog converter is provided to allow variable amounts of voltage to be subtracted off these signals, and another series of discrete gains are applied to the residual signal, which is primarily composed of the pulsatile signal, to allow variable pulse strengths (i.e., weak or strong) to be digitized for analysis by a microprocessor subsystem. This structure enables the unit to respond to changes in signal sizes essentially