`
`US005259381A
`
`(11) Patent Number:
`
`[45] Date of Patent:
`
`5,259,381
`* Nov. 9, 1993
`
`FOREIGN PATENT DOCUMENTS
`
`0104772 4/1984 European Pat. Off.
`
`............ 128/633
`
`Primary Examiner—Lee S. Cohen
`Assistant Examiner—Robert L. Nasser, Jr.
`Attorney, Agent, or Firm—Christensen, O’Connor,
`Johnson & Kindness
`
`United States Patent 19
`Cheungetal.
`:
`
`[54] APPARATUS FOR THE AUTOMATIC
`CALIBRATION OF SIGNALS EMPLOYEDIN
`OXIMETRY
`
`[75]
`
`Inventors: Peter W. Cheung, Mercer Island;
`Karl F. Gauglitz, Kirkland; Scott W.
`Hunsaker, Seattle; Stephen J.
`Prosser, Lynnwood; Darrell O.
`Wagner, Monroe; Robert E. Smith,
`Edmonds, all of Wash.
`
`{73] Assignee:
`
`Physio-Control Corporation,
`Redmond, Wash.
`
`[*] Notice:
`
`The portion of the term of this patent
`subsequent to Apr. 3, 2007 has been
`disclaimed.
`
`ABSTRACT
`[57]
`Under the present invention, a method and apparatus
`are provided for compensating for the effect tempera-
`ture variations have on the wavelength oflight emitted
`by the oximeter sensor light source (40, 42). In pulse
`oximetry, LEDs (40, 42) are typically employed to
`expose tissue to light at two different wavelengths. The
`light illuminating the tissue is received by a detector
`(38) where signals proportional to the intensity of light
`are produced. These signals are then processed by the
`oximeter circuitry to produce an indication of oxygen
`saturation. Because current oximetry techniques are
`dependent upon the wavelengthsoflight emitted by the
`LEDs(40, 42), the wavelengths must be known. Even
`when predetermined combinations of LEDs (40, 42)
`Related U.S. Application Data
`having relatively precise wavelengths are employed,
`[63]|Continuation of Ser. No. 897,663, Aug. 18, 1986, Pat.
`variations in the wavelength oflight emitted mayresult.
`No. 4,913,150.
`Because the sensor (12) may be exposedto a significant
`range of temperatures while in use, the effect of temper-
`ature on the wavelengths may besignificant. To com-
`pensate for this effect, a temperature sensor (50)is in-
`cluded in the sensor (12) to produce a signal indicative
`of sensor temperature. This signal is interpreted by the
`oximeter circuitry including, for example, a microcom-
`puter (16), where the effect of temperature on wave-
`length is compensated for. In a preferred arrangement,
`this compensation takes the form of a computation of an
`alternative calibration curve from which the oxygen
`saturation is indicated, given a particular processing of
`signals from the detector (38).
`
`[21] Appl. No.: 377,722
`
`[22] Filed:
`
`Jul. 10, 1989
`
`Int. ChS oe cceneesceseseeeeeseneneeeees A61B 5/00
`[$2]
`[52] U.S. CV. eee cecceeeeeeeeteneeeeees 128/633; 128/666;
`356/41
`[58] Field of Search................ 128/633, 634, 664-667;
`330/59, 308; 250/214 A
`
`[56]
`
`References Cited
`U.S. PATENT DOCUMENTS
`
`1/1973 Clemens ........-ecseecescseeneees 128/633
`3,709.612
`5/1986 Hamaguri.............
`.-- 128/633
`4,586,513
`
`ws 128/633
`8/1987 Goldbergeret al.
`4,685,464
`. 250/354.1
`4,710,631 12/1987 Aotsuka et al.
`......
`
`4.723.554 2/1988 Oman etal. oe 128/664
`
`3 Claims, 9 Drawing Sheets
`
`
`
`APPLE 1007
`
`APPLE 1007
`
`1
`
`
`
`U.S. Patent
`
`Nov.9, 1993
`
`Sheet 1 of 9
`
`5,259,381
`
`
`
`2025ZDL0UNE
`
`LIAVHIVNIS
`
`O/~_
`
`O/
`
`SIMO
`
`DIATIOS
`
`OF
`
`2
`
`
`
`
`U.S. Patent
`
`Nov. 9, 1993
`
`Sheet 2 of 9
`
`5,259,381
`
`|
`
`Ly
`
`d,
`
`L,-4alQ
`
`a:
`
`A———FA£3
`A:-ylo
`£3-=£1>
`
`ITEG,S.
`
`keofje-sa/—>|
`(Fig. G.
`
`3
`
`
`
`U.S. Patent
`
`Nov. 9, 1993
`
`Sheet 3 of 9
`
`5,259,381
`
`2Org
`
`SyOoIW.
`
`FW
`
`
`
`(IW)ALNITFZAUM
`
`890
`
`60|9O
`IOZ0|9¢O
`
`
`
`CO)GSINAIHL
`
`FO7G
`
`IINIPTAS
`
`NOLMLASOSEY
`
`ALNYVAALFO)
`
`pr?O7=‘7
`
`K
`
`8
`
`DVS
`
`(%)
`
`NOISSIWSNCLLO72071
`
`SLTON
`
`DOSSIWINVALA724
`
`SLTON
`
`4
`
`
`
`
`Nov. 9, 1993
`
`Sheet 4 of 9
`
`5,259,381
`
`DETECTOR
`
`
`&
`
`
`x
`
`|j { | | | | L
`
`FED
`FILTER
`
`
`
`5
`
`
`
`U.S. Patent
`
`Nov. 9, 1993
`
`5,259,381
`
`Sheet 5 of 9
`
`
`
`6
`
`
`
`5,259,381
`
`‘2sO74;
`
`-U.S. Patent
`
`Nov. 9, 1993
`
`Sheet 6 of 9
`
`b/
`
`FT1TAWVISE|L-—~—j7---—-—--+-~---4 LIMPO/T
`
`FTIGVWWVSIOEd#1
`
`LINTdWeMOP
`
`AISTLOIEG
`
`HMA&/
`
`7
`
`
`
`
`U.S. Patent
`
`Nov. 9, 1993
`
`Sheet 7 of 9
`
`5,259,381
`
`$0
`48
`doh
`
`Sig. £4.
`
`DETECTOR
`
`LV
`
`8
`
`
`
`
`
`U.S. Patent
`
`Nov. 9, 1993
`
`Sheet 8 of 9
`
`5,259,381
`
`‘27‘a2
`
`DMNUGY7AO
`(SLIOAO=)
`
`CIYISSA
`
`IWISQi
`
`oO
`
`WeMOT
`
`W088NAM
`
`9
`
`
`
`
`U.S. Patent
`
`Nov. 9, 1993
`
`Sheet 9 of 9
`
`5,259,381
`
`
`
`ATVASPISAIMSOL
`
`‘DEBae
`
`2900S%LO7VNWVS
`
`
`
`LYOSTW1LISQL
`
`WaV7l’
`
`ot
`
`W
`
`
`
`PHIASWIDPOSIW
`
`SAVIASSIOOL
`
`ALLINDIID
`207
`
`OL
`
`10
`
`10
`
`
`
`
`
`1
`
`5,259,381
`
`APPARATUS FOR THE AUTOMATIC
`CALIBRATION OF SIGNALS EMPLOYED IN
`OXIMETRY
`
`This is a continuation of the prior application Ser.
`No.897,663, filed Aug. 18, 1986, the benefit of the filing
`date of whichis hereby claimed under 35 USC 120, now
`US. Pat. No. 4,913,150.
`
`2
`such as the volume of blood in the tissue. The intensity
`of light transmitted through the tissue, when expressed
`as a function of time, is often said to include a baseline
`component, which varies slowly with time and repre-
`sents the effect of the fixed components on thelight, as
`well as a periodic pulsatile component, which varies
`more rapidly with time and represents the effect that
`changing tissue blood volumehas on the light. Because
`the attenuation produced by the fixed tissue compo-
`nents does not contain information about pulse rate and
`arterial oxygen saturation, the pulsatile signal is of pri-
`This invention relates to oximetry and, more particu-
`mary interest. In that regard, many of the prior art
`larly, to automatic calibration techniques employed in
`transmittance oximetry techniques eliminate the so-
`oximetry.
`called “DC” baseline component from the signal ana-
`The arterial oxygen saturation and pulse rate of an
`lyzed.
`individual may be of interest for a variety of reasons.
`For example, in U.S. Pat. No. 2,706,927 (Wood) mea-
`For example, in the operating room up-to-date informa-
`surements of light absorption at two wavelengths are
`taken under a “bloodless” condition and a “normal”
`tion regarding oxygen saturation can be used to signal
`changing physiological factors, the malfunction of an-
`condition. In the bloodless condition, as much blood as
`aesthesia equipment, or physician error. Similarly, in
`possible is squeezed from the tissue being analyzed.
`the intensive care unit, oxygen saturation information
`Then, light at both wavelengths is transmitted through
`can be used to confirm the provision of proper patient
`the tissue and absorption measurements made. These
`measurements indicate the effect that all nonbloodtis-
`ventilation and allow the patient to be withdrawn from
`a ventilator at an optimalrate.
`sue components have on the light. When normal blood
`In many applications, particularly including the oper-
`flow has been restored to the tissue, a second set of
`measurements is made that indicates the influence of
`ating room and intensive case unit, continual informa-
`tion regarding pulse rate and oxygen saturation is im-
`both the blood and nonblood components. The differ-
`portantif the presence of harmful physiological condi-
`ence in light absorption between the two conditionsis
`tions is to be detected before a substantial risk to the
`then used to determine the average oxygen saturation of
`patient is presented. A noninvasive technique is also
`the tissue,
`including the effects of both arterial and
`desirable in many applications, for example, when a
`venous blood. As will be readily apparent, this process
`home health care nurse is performing a routine check-
`basically eliminates the DC, nonblood component from
`up, because it increases both operator convenience and
`the signal that the oxygen saturation is extracted from.
`patient comfort. Pulse transmittance oximetry is ad-
`For a numberof reasons, however, the Wood method
`dressed to these problems and provides noninvasive,
`fails to provide the necessary accuracy. For example, a
`continual information about pulse rate and oxygen satu-
`true bloodless condition is not practical to obtain. In
`ration. The information produced, however,
`is only
`addition, efforts to obtain a bloodless condition, such as
`useful when the operator can depend onits accuracy.
`by squeezing the tissue, may result in a different light
`The method and apparatus of the present invention are,
`transmission path for the two conditions. In addition to
`therefore, directed to the improved accuracy of such
`problems with accuracy, the Wood approach is both
`information without unduecost.
`inconvenient and time consuming.
`As will be discussed in greater detail below, pulse
`A more refined approach to pulse transmittance ox-
`transmittance oximetry basically involves measurement
`imetry is disclosed in U.S. Pat. No. 4,167,331 (Nielsen).
`of the effect arterial blood in tissue has on the intensity
`The disclosed oximeter is based upon the principle that
`of light passing therethrough. More particularly,
`the
`the absorption of light by a material is directly propor-
`volumeof blood in the tissue is a function ofthearterial
`tional to the logarithm ofthe light intensity after having
`pulse, with a greater volume present at systole and a
`been attenuated by the absorber, as derived from the
`lesser volumepresentat diastole. Because blood absorbs
`Beer-Lambert law. The oximeter employs light-emit-
`some of the light passing through thetissue, the inten-
`ting diodes (LEDs) to producelight at red and infrared
`sity of the light emerging from the tissue is inversely
`wavelengths for transmission through tissue. A photo-
`proportional to the volumeofblood in the tissue. Thus,
`sensitive device responds to the light produced by the
`the emergentlight intensity will vary with the arterial
`LEDs,after it has been attenuated by the tissue, and
`pulse and can be used to indicate a patient’s pulserate.
`produces an output current. That output current is am-
`In addition, the absorption coefficient of oxyhemoglo-
`plified by a logarithmic amplifier to produce a signal
`bin (hemoglobin combined with oxygen, HbOz)is dif-
`having AC and DC components and containing infor-
`ferent from that of unoxygenated hemoglobin (Hb)for
`mation about the intensity of light transmitted at both
`most wavelengthsof light. For that reason, differences
`wavelengths. Sample-and-hold circuits demodulate the
`in the amount of light absorbed by the blood at two
`red and infrared wavelength signals. The DC compo-
`different wavelengths can be used to indicate the hemo-
`nents of each signal are then blocked byaseries of
`60
`globin oxygen saturation, % SaOQ2 (OS), which equals
`bandpass amplifier and capacitors, eliminating the effect
`([HbO2]/((Hb] + [HbO2])) x 100%. Thus, measurement
`of the fixed absorptive components from the signal. The
`of the amountof light transmitted through, for example,
`resultant AC signal componentsare unaffected byfixed
`a finger can be used to determine both the patient's
`absorption components, such ashair, bone,tissue, skin.
`pulse rate and hemoglobin oxygen saturation.
`An average value of each ACsignal is then produced.
`Aswill be appreciated, the intensity of light transmit-
`The ratio of the two averagesis then used to determine
`ted through a finger is a function of the absorption
`the oxygen saturation from empirically determined val-
`coefficient of both “fixed” components, such as bone,
`ues associated with the ratio. The AC components are
`tissue, skin, and hair, as well as “variable” components,
`also used to determine the pulserate.
`
`BACKGROUNDOF THE INVENTION
`
`10
`
`20
`
`25
`
`35
`
`45
`
`50
`
`55
`
`65
`
`11
`
`11
`
`
`
`3,259,381
`
`3
`Another reference addressed to pulse transmittance
`oximetry is U.S. Pat. No. 4,407,290 (Wilker). In that
`reference, light pulses produced by LEDsat twodiffer-
`ent wavelengthsare applied to, for example, an earlobe.
`A sensor respondsto the light transmitted through the
`earlobe, producing a signal for each wavelength having
`a DC and AC componentresulting from the presence of
`constant and pulsatile absorptive components in the
`earlobe. A normalization circuit employs feedback to
`scale both signals so that the DC nonpulsatile compo-
`nents of each are equal and the offset voltages removed.
`Decoders separate the two signals, so controlled, into
`channels A and B where the DC componentis removed
`from each. The remaining AC componentsofthe sig-
`nals are amplified and combined at a multiplexer prior
`to analog-to-digital (A/D) conversion. Oxygen satura-
`tion is determined by a digital processor in accordance
`with the following relationship:
`
`OS = X1R(Ay) + X2R(A2)
`X3R(A1) + X4R(A2)
`
`wherein empirically derived data for the constants X1,
`X2, X3 and X4 is stored in the processor.
`European Patent Application No. 83304939.8 (New,
`Jr. et al.) discloses an additional pulse transmittance
`oximeter. Two LEDsexpose a body member,for exam-
`ple, a finger, to light having red and infrared wave-
`lengths, with each LED having a one-in-four duty cy-
`cle. A detector producesa signal in response that is then
`split
`into two channels. The one-in-four duty cycle
`allows negatively amplified noise signals to be inte-
`grated with positively amplified signals including the
`detector response and noise, thereby eliminating the
`effect of noise on the signal produced. The resultant
`signals include a substantially constant DC component
`and an AC component. To improve the accuracy of a
`subsequent analog-to-digital (A/D) conversion,a fixed
`DC value is subtracted from the signal prior to the
`conversion. This level is then added backin by a micro-
`processor after the conversion. Logarithmic analysis is
`avoided by the microprocessor in the following man-
`ner. For each wavelength oflight transmitted through
`the finger, a quotient of the AC component over the
`constant componentis determined. Theratio of the two
`quotients is then determined and fitted to a curve of
`independently derived oxygen saturations. To compen-
`sate for the different
`transmission characteristics of
`different patient’s fingers, an adjustable drive source for
`the LEDsis provided.
`In European Patent Application 83304940.6 (New et
`al.), a calibrated oximeter probe is disclosed. That probe
`includes a coding resistor or coding connector used to
`identify the particular combination of wavelengths of
`light emitted by the two LEDs contained thereon. Ox-
`imeter circuitry then senses the code ofthe resistor or
`connector to determine the wavelengthsof light emit-
`ted by the LEDs. In this manner, the effect that differ-
`ent wavelengths have on the oxygen saturation compu-
`tations can be compensated for. The basis upon which
`oxygen saturation is measured involves the determina-
`tion of the quotient of the pulsatile component over the
`constant componentoflight transmitted at each wave-
`length. The ratio of the quotients for the two wave-
`lengths is then fitted to a curve of independently de-
`_Tived oxygen saturations. Outputs include pulse rate and
`oxygen saturation.
`
`4
`Even with the calibration techniques of New, Jr. et
`al. employed, however, the wavelengths oflight emit-
`ted by the LEDs may changein a mannerthat the oxim-
`eter circuitry is usable to detect. As will be appreciated,
`such variations can significantly affect the accuracy of
`the oxygen saturation measurements. The disclosed
`invention is directed to the provision of more complete
`information about the actual wavelengths of the light
`emitted and, hence, the production of more accurate
`oxygen saturation measurements.
`SUMMARYOF THE INVENTION
`
`The present invention discloses a method of deter-
`mining the oxygen saturation of arterial blood flowing
`in tissue. The method includes an initial step in which
`the tissue is exposed to light from two sourcesat sepa-
`rate temperature-dependent wavelengths. An indication
`of the temperature of the sources is produced, as are
`signals produced in response to the exposure of the
`tissue to the light at the separate temperature-dependent
`wavelengths. A preliminary indication of the oxygen
`saturation is then produced from the signals. A compar-
`ison of independently derived oxygen saturations with a
`continuum of such preliminary indications of oxygen
`saturation is then selected in accordance with the indi-
`cation of the temperature of the sources earlier pro-
`duced. From this comparison, the actual oxygen satura-
`tion corresponding to the preliminary indication previ-
`ously obtained is produced.
`In accordance with a particular aspect of the inven-
`tion, an indication of the separate temperature-depend-
`ent wavelengths of light emitted by the sources at a
`reference temperature is produced. This indication is
`used to further aid in the selection of the appropriate
`comparison of independently derived oxygen satura-
`tions to the preliminary indications of oxygen satura-
`tion.
`In accordance with a further aspect of the invention,
`an oximeter is disclosed that employs the foregoing
`method to determine the oxygen saturation of arterial
`blood flowing in tissue. The oximeter includes first and
`second light sourcesthat illuminate the tissue with light
`at separate temperature-dependent wavelengths. The
`oximeter also includes a temperature detector that pro-
`duces an indication of the temperature of the light
`sources. Signal that are proportionalto the intensity of
`light received from the tissue at each of the tempera-
`ture-dependent wavelengths are produced bya light
`detector and a processor analyzesthesignals to produce
`a preliminary indication of the oxygen saturation of the
`blood. A selection circuit selects a particular compari-
`son of oxygensaturations with the continuum ofprelim-
`inary indications of the oxygen saturation in accordance
`with the indication of temperature received. Finally, a
`converter converts the preliminary indication of oxy-
`gen saturation into an oxygen saturation determination
`by reference to the comparison selected.
`In accordance with additional aspects of the inven-
`tion, a red opticalfilter filters the light received by the
`light detector. The signals produced by the light detec-
`tor can,similarly, be amplified by a differential current-
`to-voltage amplifier before being analyzed by the pro-
`cessor. A sensor housing, having first and second ele-
`ments, is employed to receive the tissue being analyzed
`and to define a light path between the light sources and
`the detector. A mirror, attached to the housing,is posi-
`tioned between the light sources and detector and
`breaks the light path up into first and second segments
`
`20
`
`2wn
`
`45
`
`50
`
`35
`
`60
`
`65
`
`12
`
`12
`
`
`
`5,259,381
`
`6
`
`DETAILED DESCRIPTION
`
`5
`at a predetermined angle with respect to each other.
`The two elements of the housing may pivot and be
`closably biased. In another arrangement, an apparatusis
`constructed in accordance with this invention indepen-
`dently of the light sources and light detector.
`In accordance with anotheraspect of the invention, a
`sensoris disclosed for use with an oximeter to determine
`the oxygen saturation ofarterial blood flowingin tissue.
`The sensor includes first and second light sources for
`illuminating the tissue with light at separate tempera-
`ture-dependent wavelengths. A temperature indicatoris
`also included to produce an indication of the tempera-
`ture of the light sources. Signals are produced in re-
`sponse to the illumination of the tissue at each of the
`temperature-dependent wavelengths by a light detec-
`tor.
`
`5
`
`10
`
`’
`
`15
`
`20
`
`25
`
`30
`
`35
`
`Referring to the overall system block diagram shown
`in FIG. 1, a pulse transmittance oximeter 10 employing
`this invention includes a sensor 12, input/output (1/O)
`circuit 14, microcomputer 16, power source 18, display
`20, keyboard 22 and alarm 24. Before discussing these
`elements in detail, however, an outline of the theoretical
`basis of pulse transmittance oximetryas practiced by the
`oximeter of FIG. 1 is provided.
`An understanding of the relevant theory begins with
`a discussion of the Beer-Lambert law. This law governs
`the absorption of optical radiation by homogeneous
`absorbing media and can best be understood with refer-
`ence to FIGS. 2 and 3 in the following manner.
`Asshownin FIG.2,incident light having an intensity
`Ip impinges upon an absorptive medium 26. Medium 26
`BRIEF DESCRIPTION OF THE DRAWINGS
`has a characteristic absorbance factor A that indicates
`The invention can best be understood by reference to
`the attenuating affect medium 26 has on the incident
`the following portion of the specification, taken in con-
`light. Similarly, a transmission factor T for the medium
`junction with the accompanying drawings in which:
`is defined as the reciprocal of the absorbance factor,
`FIG. 1 is a block diagram of an oximeter including a
`I/A. The intensity of the light 1; emerging from me-
`sensor,
`input/output
`(I/O) circuit, microcomputer,
`dium 26is less than Ip and can be expressed functionally
`alarm, displays, power supply, and keyboard;
`as the product TIp. With medium 26 divided into a
`FIG.2 is a block diagram illustrating the transmission
`number of identical components, each of unit thickness
`of light through an absorptive medium;
`(in the direction of light transmission) and the same
`FIG.3 is a block diagram illustrating the transmission
`transmission factor T, the effect of medium 26 on the
`of light through the absorptive medium of FIG. 2,
`incident light Ip is as shown in FIG. 3.
`wherein the medium is broken up into elemental com-
`There, medium 26isillustrated as consisting of three
`ponents;
`components 28, 30, and 32. As will be appreciated, the
`FIG. 4 is a graphical comparison ofthe incident light
`intensity I; of the light emerging from component28is
`intensity to the emergent light intensity as modeled in
`equal to the incident light intensity Ip multiplied by the
`FIG. 2;
`transmission factor T. Component30 hasa similar effect
`FIG. 5 is a graphical comparison of the specific ab-
`on light passing therethrough. Thus, because the light
`sorption coefficients for oxygenated hemoglobin and
`incident upon component30 is equal to the product Tlo,
`deoxygenated hemoglobin as a function of the wave-
`the emergent light intensity Iz is equal to the product
`length of light transmitted therethrough;
`TI, or T2Io. Component 32 has the same effect on light
`FIG. 6 is a block diagram illustrating the transmission
`and, as shown in FIG. 3, the intensity of the emergent
`of light through a block model of the components of a
`light I3 for the entire medium 26 so modeled is equal to
`finger;
`the product TIor To. If the thickness d of medium 26
`FIG. 7 is a graphical comparison of independently
`is n unit lengths,it can be modeled as includingnidenti-
`derived oxygen saturation measurements with a vari-
`cal components of unit thickness. It will then be appre-
`able that is measured bythe oximeter;
`ciated that the intensity of light emerging from medium
`FIG.8 is a schematicillustration of the transmission
`26 can be designated I, and the productis equal to T"Ipo.
`of light at two wavelengths through a finger in accor-
`Expressed as a function of the absorbance constant A,
`dance with the invention;
`I, can also be written as the product (1/A*)lo.
`FIG. 9 is a graphical plot as a function of time of the
`From the preceding discussion,
`it will be readily
`transmittanceoflight at the red wavelength through the
`appreciated that the absorptive effect of medium 26 on
`finger;
`the intensity of the incident light Ip is one of exponential
`FIG. 10 is a graphical plot as a function oftime of the
`decay. Because A may be an inconvenient base to work
`transmission of infrared light through the finger;
`with, I, can be rewritten as a function of a more conve-
`FIG. 11 is an exploded view showing the sensor of
`nient base, b, by recognizing that A” is equal to b2”,
`FIG. 1 in greater detail;
`wherea is the absorbance of medium 26perunit length.
`FIG.12 is a more detailed schematic of the I/O cir-
`The term a is frequently referred to as the extinction
`cuit illustrated in the system of FIG.1;
`coefficient and is equal to log,A.
`FIG. 13 is a schematic diagram of a conventional
`Given the preceding discussion,it will be appreciated
`current-to-voltage amplifier circuit;
`that the intensity of the light I, emerging from medium
`FIG. 14 is a schematic diagram ofa differential cur-
`26 can be expressed in base 10 as I910—@1", or in base e
`rent-to-voltage preamplifier circuit included in the I/O
`as Ipe~%2", where a and a2 are the appropriaterelative
`circuit of FIG.1;
`extinction coefficients for base 10 and base €
`respec-
`FIG. 15 is a graphical representation of the possible
`tively. The effect that the thickness of medium 26 has on
`ranges of I/O circuit output, showing the desired re-
`the emergent lightintensity I,, is graphically depicted in
`sponse to the I/O circuit and microcomputerat each of
`FIG. 4. If the light incident upon medium 26 is estab-
`the various possible ranges;
`lished as having unit intensity, FIG. 4 also represents
`the transmission factor T of the entire medium as a
`FIG. 16 is a more complete schematic diagram of the
`function of thickness.
`microcomputerillustrated in FIG. 1; and
`FIG. 17 is a family of curves similar to the oneillus-
`The discussion above can be applied generally to the
`trated in FIG. 7.
`medium 26 shown in FIG. 2 to produce:
`
`45
`
`50
`
`35
`
`60
`
`65
`
`13
`
`13
`
`
`
`5,259,381
`
`8
`nent 34 can be written as a function ofthe incident light
`intensity Ip as follows:
`
`N= Ige— 4
`
`q)
`
`whereI) is the emergentlight intensity, Io is the incident
`light intensity, a is the absorbance coefficient of the
`medium per unit length, d is the thickness of the me-
`dium in unit lengths, and the exponential nature of the
`relationship ha arbitrarily been expressed in terms of
`base e. Equation (1) is commonly referred to as the
`Beer-Lambert law of exponential light decay through a
`homogeneous absorbing medium.
`While the basic understanding of the Beer-Lambert
`law, a discussion of its application to the problems of
`pulse rate and hemoglobin oxygen saturation measure-
`mentis now presented. As shownin FIG. 5, the absorp-
`tion coefficients for oxygenated and deoxygenated he-
`mogliobin are different at every wavelength, except
`isobestic wavelengths. Thus,it will be appreciated that
`if a person’s finger is exposed to incident light and the
`emergent
`light
`intensity measured,
`the difference in
`intensity between the two, which is the amountoflight
`absorbed, contains information relating to the oxygen-
`ated hemoglobin content of the blood in the finger. The
`manner in whichthis information is extracted from the
`Beer-Lambert law is discussed below. In addition, it
`will be appreciated that the volume of blood contained
`within an individual’s finger varies with the individual's
`arterial pulse. Thus,
`the thickness of the finger also
`variesslightly with each pulse, creating a changing path
`length for light transmitted through the finger. Because
`a longer lightpath allows additional
`light
`to be ab-
`sorbed, time-dependent information relating to the dif-
`ference between the incident and emergentlight intensi-
`ties can be used to determinethe individual's pulse. The
`manner in which this information is extracted from the
`Beer-Lambert law is also discussed below.
`information
`As noted in the preceding paragraph,
`about
`the incident and emergent
`intensities of light
`transmitted through a finger can be used to determine
`oxygen saturation and pulse rate. The theoretical] basis
`for extracting the required information, however,
`is
`complicated by several problems. For example,
`the
`precise intensity of the incident
`light applied to the
`finger is not easily determined. Thus, it may be neces-
`sary to extract the required information independently
`of the intensity of the incident light. Further, because
`the charging volumeof blood in the finger and, hence,
`thickness of the lightpath therethrough, are not exces-
`sively dependent upon the individual’s pulse,it is desir-
`able to eliminate the changing path length as a variable
`from the computations.
`The mannerin which the Beer-Lambert lawis refined
`to eliminate the incident intensity and path lengths as
`variables is as follows. With reference to FIG. 6, a
`humanfinger is modeled by two components 34 and 36,
`in a manner similar to that shown in FIG. 3. Baseline
`component 34 models the unchanging absorptive ele-
`mentsof the finger. This componentincludes, for exam-
`ple, bone, tissue, skin, hair, and baseline venous and
`arterial blood and has a thickness designated d and an
`absorbance a.
`Pulsatile component 36 represents the changing ab-
`sorptive portion ofthe finger, the arterial blood volume.
`As shown, the thickness of this componentis designated
`Ad, representing the variable nature of the thickness,
`and the absorbanceof this componentis designated a,
`representing the arterial blood absorbance.
`Aswill be appreciated from the earlier analysis with
`respect to FIG. 3, the light 1; emerging from compo-
`
`I= Ie 04
`
`(2)
`
`Likewise, the intensity of light Iz emerging from com-
`ponent 36 is a function ofits incident light intensity 1),
`and:
`
`Ih=Ne—*Ahd
`
`(3)
`
`Substitution of the expression for I; developed in equa-
`tion (2) for that used in equation (3), when simplified,
`results in the following expression for the intensity I2 of
`light emerging from thefinger as a function of the inten-
`sity of light Ip incident upon the finger;
`
`In=Ipe—lad+ add)
`
`(4)
`
`Because our interest lies in the effect of the light pro-
`duced by the arterial blood volume, the relationship
`between Ip and 1, is of particular interest. Defining the
`change in the transmission produced by the arterial
`component 36 as Ta4, we have:
`
`G)
`
`Substituting the expressions for I; and I; obtained in
`equations (2) and (3), respectively, equation (5) be-
`comes:
`
`35
`
`Tas =
`
`Ige7 l0d- aAAd)
`—ad
`ioe
`
`(6)
`
`45
`
`50
`
`It will be appreciated that the Ip term can be cancelled
`from both the numerator and denominator of equation
`(6), thereby eliminating the input light intensity as a
`variable in the equation. With equation (6) fully simpli-
`fied,
`the change in arterial
`transmission can be ex-
`pressed as:
`
`Taa=e7 tAad
`
`(7)
`
`A device employing this principle of operation is effec-
`tively self-calibrating, being independentofthe incident
`light intensity Ip.
`At this point, a consideration of equation (7) reveals
`that the changing thickness of the finger, Ad, produced
`by the changing arterial blood volumestill remains as a
`variable. The Ad variable is eliminated in the following
`manner. For convenience of expression, the logarithms
`of termsin equation (7) are produced with respect to the
`same base originally employed in equation (1). Thus,
`equation (7) becomes:
`
`In Ta4=1n (e— 9444) = —aydd
`
`(8)
`
`A preferred technique for eliminating the Ad variable
`utilizes information drawn from the changein arterial
`transmission experienced at two wavelengths.
`The particular wavelengths selected are determined
`in part by consideration of a more complete expression
`of the arterial absorbance ay:
`
`65
`
`a4=(aga)(OS)—(apai —OS)
`
`(9)
`
`14
`
`14
`
`
`
`9
`where agg is the oxygenated arterial absorbance, ap, is
`the deoxygenated arterial absorbance, and OS is the
`hemoglobin oxygen saturation of the arterial blood
`volume. As will be appreciated from FIG. 5, ag, and
`apa are substantially unequalat all light wavelengths in
`the visible-red and near-infrared wavelength regions
`except for an isobestic wavelength occurring at approx-
`imately 805 nanometers. With an arterial oxygen satura-
`tion OS of approximately 90 percent, it will be appreci-
`ated from equation (9) that the arterial absorbance a, is
`90 percent attributable to the oxygenated arterial absor-
`bance ao, and 10 percentattributable to the deoxygen-
`ated arterial absorbance apy. At the isobestic wave-
`length, the relative contribution of these two coeffici-
`ents to the arterial absorbance ay is of minimal signifi-
`cancein that both ag, and ap, are equal. Thus, a wave-
`length roughly approximately the isobestic wavelength
`of the curvesillustrated in FIG. 5 is a convenient one
`for use in eliminating the changein finger thickness Ad
`attributable to arterial blood flow.
`A second wavelength is selected at a distance from
`the approximately isobestic wavelength that is sufficient
`to allow the twosignals to be easily distinguished. In
`addition, the relative difference of the oxygenated and
`deoxygenated arterial absorbancesat this wavelength is
`more pronounced. In light of the foregoing consider-
`ations,
`it
`is generally preferred that
`the two wave-
`lengths selected fall within the red and infrared regions
`of the electromagnetic spectrum.
`The foregoing information, when combined with
`equation (8) is used to produce the followingratio:
`
`5,259,381
`
`10
`For simplicity, a measured ratio Ros is defined from
`equation (11) as:
`
`:
`Ratio = Ros =
`
`a4 @arR
`a4 @ IR
`
`(12)
`
`It is this measured value for Ros that is plotted on the
`x-axis of independently derived oxygen saturation
`curves, as shown in FIG. 7 and discussed in greater
`detail below, with the hemoglobin oxygen saturation
`being referenced on the y-axis.
`Measurement of the arterial absorbances at both
`wavelengths is performed in the following manner. As
`shownin FIG.8, a detect