CA1327899C - Spectrophotometric method for quantitatively determining the concentration of a dilute component in a light or other radiation-scattering environment - Google Patents

Abstract

ABSTRACT OF THE INVENTION
A spectrophotometric method is described of quantitatively determining the concentration of a dilute component in either a clear or a strongly light-scattering environment containing same in unknown concentration together with a reference component of known concentration, by a series of contemporaneous radiation-directing and measurement steps of radiation of selected varying wavelengths. Specific applications are disclosed involving the in situ, in vivo, non-invasive spectrophotometric determination of blood-borne as well as tissue species, e.g., hemoglobin, and oxyhemoglobin, and intra-cellular enzyme cytochrome c oxidase, in human body parts such as fingers, hands, toes, feet, earlobes, etc., as well as organs such as the brain, skeletal muscle, liver, etc.

Description

!,; .
~ 1327~.9 ~
"SPECTROPHOTr)METRlC ~lETHOD FOR QUANTITATI~IE' Y DETERMlf~IllG
THE CONCE~TRATION Of A Dll UTE COMPONE~T 1~ A B IGHT- OR OT~ER
RADIATION-SCATTERING ENVIRONME~T~

~ BACKGROUND OF THE INVENTION

Field Of The Invention This invention generdlly relates to a spectrophotometric method of quantitatively determin~ng the corlcentration of a dilute component in a light- or other radiation-scattering environment containing the dilute component in combination with a reference component of known concentration.

Descri tion Of The Related Ar~
P
In many fields of technology there is a need for quantitative determination of dilute component concentrations in environments where the dilute component is in combination with a reference component of known concentration. Examples of illustrative environment 5 of such type include enzymes, proteins, and metabolit'es in corporeal fluids; acidic fumes or gaseous components (e.g., hydrogen sulfide and sulfuric acid, nitric acid, carbon monoxide, etc.) in the atmosphere; salt concentrations in sea water undergoing desalination; ozone ~n oz'one-enriched air utilized in waste water o20nat10n systems, etc.

In part~cular, there has been a specific need in the medlcal and health care fields for a non-invasive, continuous, atraumatic, ~n vivo, in situ determlnation of amounts of -' crit1cal metabolic 1ndicators In body flu1ds or tissues of human patients. Examplss of such body fluids include the blood and fluids associated with the lymphatic and neurological systems of the body~ Further specific examples involving the human circulatory system include the monitoring of glucose and of o~ygenated/de-oxygenated, arterial/venous j~;
colored hemoglobin in the blood stream. In additiong monitoring in localized tissue, such as brain and muscle, of certain enzyme species such as the cytochrome c oxidase enzyme tunofficially better known as cytochrome ,a31 or met'abolic substrates ~such as glucose) or products (such as carbon ~$'' :
' ,. ~ ': ' . .
: , -2- 1327~9 dioxide) i5 becoming an increasingly urgent practical applic~tlon of spectrophotometric technology.

Spectrophotometric methods have been proposed in the art to monitor metabolites in corporeal fluids. Such methods involve the impingement of radiation, typically in the visible or near-infrared region, onto the exterior body portion of the subject for transdermal and interior tissue penetration of the radiation, which is moni~ored as to its reflectance or transmission, at a wavelength condition at which the metabollte or other monitored component is selectively absorpt~ve for the radiation. This technique is mainly limited to yielding a qualltative determination from the measured output radiation (reflected or transmitted) of the qual1tatlve eharacter of the metabollsm. At best a semi^quantitat1ve result can be obtained in a so-called trend monitoring mode where concentration changes can be monitored in ter0s of an original baseline condition of unkown concentration.

Solute concentrations in dilute fluid media can be theoretically quantified by the Beer-~ambert law, i.e., log(1O/I~ = d x E x c 20 wherein: ;
Io ~ lntens~ty of source radiation ~mpinged on the sample;
1 5 lntensity of radiation transmitted through the sample, E = absorption (extinction) coefficient oF the solute species at the wavelength of the source radiation implnged on the sample;
d u optical d~stance ~travel pathlength of the radiat~on transmltted through the sample); and ;;
c ~ concentrat~on of the solute (dilute component) in the ;` ..
solut~on sample.

Although the ~oregoing Beer-!ambert ~aw equation per~its~ -~
a ready determinatlon of solute concentrat~on to be made ~n ln vitro or other non-corporeal discrete sample systems utll~zed ~or convent~onal spectrophotometric assays, such direct, quantitative measurement is not possible ln the lntact bo~yD

, .

.

.

~?

-3- 1 3 2 7 8 9 ~

even though the lnfl~ent radiation is penetrative of the body elements of the corporeal system, e.g., bones, musculature, org~ns, and the like~ slnce the scattering of radlation during its passage through the corporeal system ~s e~tensi~e and highly variable ~n character. Such scattering not only adds an unknown loss of radiation to the required information regarding specific absorption but by multiple scattertng ~t also lengthens to an unknown degree the path length of those photons eventua1ly emerging form the body element. As a result, it has not been possible to determine in an in viYo situation what the effective path length, d, of the impinged radiation actually is, prior to measurement of the transmit~ed or reflected radiation derived therefrom. In cons~quence9 the absolute quantitation of solute concentrations in corporeal systems has been severely adversely limited.

Faced with the alternatives of invasive and traumatic sampling of the corporeal fluids of interest, or spectrophoto-metric methods whlch realize only qualitat~ve or at best semi-quantitative measurement of changes in t~ssue or body fluid solute concentrations, there is a substantial perceived need in the art for a non-invasive, in vivo method of 20 quantitatively determtning the concentration in corporea~
solution of a solute in a body f1u1d solvent.

~ similar need exists in numerous other fields in ~thich absolute concentrations of dilute component species in fluid media wo~ld materially assist the characterization of the fluid system. An example is atmospheric monitoring of "acid rain~" i.e., airborne acidic contaminants which have in recent years proliferated and been determined to cause widespread biospheric damage, including the defoliation~ of forest stocks and spol~ation of natural bodies of water and other aqueous environments. It is anticipated that in coming years with the inoreasing severity of the acid rain problem, correspondingly greater sc1entifi~ and legislati~ve efforts will be focused on the ~on~torlng of acid rain with a ~iew to controlling and , .~ , ~ .' ~, ` '' ~
. ,:

-4- i~27~

minimizing its adverse impacts. Determinations in the naturally ex~sting medium such dS turbid water and hazy or cloudy atmosphere will be a great boon for direct, effect~ve and rapid monitoring of these environments.

In numerous other indus$rial and natural systems there ls a need to quanitatively monitor solute species in an indirect manner not involving the time, effort, and cost of discrete sample collection, purif~cation and analysis.

U. S. Patent 4,281,645 to F. F. Jobsis describes a spectrophotometric system for monitoring cellular oxidat~ve metabolism by non-~nvasively measuring in vivo changes in the steady state oxidat~on-reduction of cellular cytochromes together w1th changes ~n blood volume, the oxidatlon state of hemogloblll and the rate of blood flow in the brain, heart, lS kldneys, other organs, limbs, or other parts of a human or animal body.

The methodology described in the Jobsis patent involves transmitting near-infrared radiation ~n at least two d1fferent and periodically recurring wavelengths through the corpo~eal environment, and detecting and measuring the radiatlon ~nten-slty which emerges at another, distant point or on the opposite side of the body, for the monitoring of biochemical reactiuns, utilizing an approximation of the ~eer-Lambert law.
One of such wavelengths selected for the measurement is in a range for which oxidized cytochrome a, a3 is selectively highly absorptive~ One or more reference signals are provided at corresponding wavelengths outside the peak of the cytochrome absorption band but preferably in close proximity to the measuring wavelength. The d~fference or ratio between the measuring and reference signals is determined and non-speclf~c changes ~n the intensity of transmitted radiatlon not ~ttr~butable to absorption by the cytochrome species are eliminated. Thus, the system of this patent produces~an ,~" . ,, .. "

` ~ ~
~327899 output signal representing the difference in or ratio of absorption of the measuring and reference wavelengths by the organ or other corporeal portion of the body as a function of the state o~ the metabolic activity in vivo, which may be S converted to a signal providing a substantially cont~nuous measure of such activity. A related spectrophotometric reflectance technique ~s disclosed in U.S. Patent 4,223,680 to F. F. Jsbsis.

U. S. Patent 4,655,225 to C. Dahne et al discloses a spectrophotometric system for non-invasive determination oF
glucose concentration in body tissue. The system involves irradiation of the exterior body portion with an optical light whose transmittance or reflectance is collected at selected band wave)ength vatues for the glucose absorption spectrum and at a selected band wavelength value for the absorption spectru~ of background tissue containing no or insignificant amounts of glucose. The measuring and reference radiation collected is then converted into electrical signals and uti-lized to deter~ine glucose concentrations.

It ls an object of the present invention to provid~ an improved method and apparatus for indirectly quantitatively, spectrophotometrically determln~ng the amounts oF a dilute component by using a reference component in the environment of 1 nterest .

It is another ob~ect of the ~nvention to provide such a method for non-lnvas~ve, in vivo quantitative determ~nation oF
the concentrat~on of a dilute solute component in a corpore~1 -solvent environment.
.
Other ob~ects and advantages of the present invention will be more fully apparent fro~ the ensuing disclosure and appended cla~s~
.

,. , . ;.
;............ , , :
'.,, " , ' ' . .
: ~ ' . ' ' ~

. .

~ ~qo9;
-- -6- ~ 3 2 7g 99 SUMMARY OF THE lNVE~TION

The present lnvention relates to d method of determining the true concentration (e.g., in terms of grams or moles o~ a dilute component per volume of a reference component) in environ~ental media in whlch the optical pathlength ~s 5 ill-defined due to the extensive occurence~ of scatterin~ of incident radiation, such as in very long distance atmospheric monitor~ng as well as in more intensely light scattering media duriny transi11umination as well as d;ffuse reflectance modes of spectrophotometry.

As used herein~ the term "environment'l refers to a selected spatial region in which the directed and measured radiation is transmitted andtor reflected along substant~ally the same path.

is rhe crux o~ the invention is to measure the transmi~ted and/or reflected radiation for both the dilute component of unknown concentration and the reference component of known concentration with wh~ch it is associated. Multiple scattering spoils the optical pathlength parameter in the Beer-~ambert equation and does so to different degrees depending on the wavelength. It ;s, therefore, necessary to measure the d~lute and reference components in closely the same spectral region. Measuring the ~ntensity of the light absorption and/or re~lectance by the two types of molecu1es, dilute component and reference component, in the environment and apptying the extinction coe~ficients of each provides the opportunity to relate them to one another ln terms of relative amounts, which is recognized as being the essential character o~ concentration. Thus, by absorption (and/or reflectance) measurements and with knowledge of the extinction coefficents, the amount of dilute component in the light path and the amount of reference component lt is associated with, ~ . .
,' , ~ , , ~ . `
; . ... .
. .
. , , , - . ~
-.: ~
. . .

~ ~.
- . . ; . .... . . .
.

- ~ab f~
~7~ 1327~99 determined s~m~larly at other near-by wavelength(s),may be employed to calculate the coneentrat~on of the dilute component relative to the reference component.

In a system in wh~ch the reference component is present in known concentration, the apparent pathlength may be determined by absorption measurements taken in the environment of unknown pathlength in the same electromagnetic spectral region. The difference between the resulting absorption Yalues is calculated as the differential absorbance in the environment whose pathlength is to be determined~ The tabulated or previously determined extinction coefficient values of the pure reference component at such wavelengths are then employed to calculate the differential extinction coefficient, I as the difference between the respective extinction coefficient values. When the d~fferential absorbance i5 then divided by the dlfferential extinction coefficient, the result is the apparent effective pathlength of the environ~ent.
When the electromagnetic radiation emitter and detector spac~ng distance is measured, the pathlengthening f~ctor for the system is determined as the ratio of the apparent effective pathlength to the actual e~itter-to-detector spacing distance.

In another selected, specific aspect, the invention relates to a spectrophotometric method of quantitat1vely determining the concentration of a dilute component in an environment containing the dilute component of known identity but of unknown concentration in combination with a reference çomponent of known concentration, by a series of suçcessive, substantially contemporaneous measurements of transmitted and/or reflected rad~ation at selected wavelengths, comprising:

(a) determin~ng the apparent effective pathlength in sa~d ~nv~ronment; ' ~- -,:

... "
'' '' .

~ `
-8- ~3278~ .

(b) directing at the environment incident electromag-netlc radiation of a first wavelength in a selected spectral region at whlch the dllute and/or reference componentts) exhibit absorption for the e1ectromagnetic radiat~on;

S (c) measurlng the first wavelength radiation transmitted and/or reflected by the envlronment;

(d) directing at the environment incident electromag-netic radiation of at least one other wavelength in the selected spectral region at which the dilute and/or reference component(s) exhibit absorption of different relative intensities than for the first wavelength incident radiation, whereby absorption ls exhiblted in sald selected spectral region comprlslng sald flrst and a~ least one other wavelength by both the dllute and the reference component(s);

~e) measuring the other wavelength radiation transmitted or reflected by the environment; and (f) determinlng extinction coefficient values for the ~: dilute component at said first and other wavelengths in said envlronment; and 9 (9) based on the apparent effective pathlength determined for the environment, extlnction coefficient values for the dllute component at said first and other wavelengths, and the measured absorbed and/or reflected radiation at said wavelengths, determining the relative amount of the dilute component to the amount of the reference component, as the concentrat~on of the dilute component ln the environment.
.
In another aspect of the method broadly described above3 the determination of step (f) is effected by establishing simultaneous mod;fled Beer-~ambert equations for each of the radiatlon absorption and/or re~lectance measurement steps, and solving the equations for the concentratlons of the dllute component and the reference component in the environment.

.. . . .

~.. ,, .., ,,. ., "
~, . . .

;i: :'.. , : .

-9- i3~78~ .

A further aspect of the invention relates to a spectrophotometric method of quantitatively determinin~ the concentration of a dilute component in an environment containing the dilute component of known identity but of unknown concentration in combindtion with d reference component of known concentration, comprising:

(a) directing at the environment incident electromagnetic radiat~on at a number of wavelengths in a selected spectral region at which the dilute and/or re~erence components exhibit absorption for the electromagnetic radiation, the number of such wavelengths being determined by the number of dilute and reference components in the environment, and the scattering characteristi.cs of the environment;

(b) determining the absorbance by the environment . of the eleetromagnetic radiation at the various wavelengths and the relative intensfties of the absorption contributions of the d~lute and reference components and scattering losses from the environment at each of such wavelengths;
~ .
:: 20 (c) at each of such waYelengths, establishing absorption equations of the form:

Absw a ~ X jA; ~ zR + S, i =l wherein: Absw i~ the absorbance by the environment, containing the d~lute and reference components, of the incident : electromagnetic radiation of wavelength w; xj is the relative ~ntens~ty of the ~bsorpt10n contribution of the assoc~ated dilute component A~, and wherein terms of the form xjA~ are set forth for each of the dilute components; n ~s the,number of d~lute components; 2 iS the relative intenslty of the -lo- 13278~9 absorption contribution o~ the reference component; R is the concentrat~on of the reference component; and S is the normalized scattering o~ the environment at wavelength w, thereby establishing for each of such wavelengths an S absorbance equation, to yield a set of simultaneous equations whose number equals the number of dilute and reference components and the number of wavelengths required to characteri2e the scattering of the environment;

(d) deriving algorithms by matrix solution of the aforementioned simultaneous equations, said algorlthms being of the form:
m e rc] = ~ a jAbsWj, ~=1 wherein: ~c] is the concentration of the specific dilute or reference component, aj is a determined numerical constant; m is the number of said wavelength determ1nations; and Abswi is the absorbance at wavelength w, thereby establishing the concentration of each of the dilute and reference components in the en~ironment.

In the above desoribed method aspects of the invention for quantitatively determining the concentrat;on o~ the dilute component in the enviornment containing same in combination with a re~erence component of known concentration, the various radiation-direct~ng and measuring steps at the var~ous selected wavelengths will be carried in a contemporaneous fashion to determine the dilute component concentration. I~
will be recogni2ed, however, that in many systems of interest, ~0 especlally including corporeal environments, it may be desirable to establish the dilute component concentration by ;. :~;
such contemporaneous radiation-directing and measurement .
_ steps, and that subsequent to such determ1nation, it may be ~ ~
advantageous to monitor the system for a period of time9 ~ -35 e~ther at discreet intervals or on a continuous basis. ~.

~`

,: ,. . . -: , . ~ ,, ; ~ , , . . .
.

-11- 11 327~99 Still another aspect of the invention relates to apparatus for spectrophotometricallY ~uantitatively determining the concentration of a dilute component in an environment containing the dilute component of known identity S but of unknown concentration in combination with a reference component of known concentration, comprising:

(a) means for producing electromagnetic radiation of known wavelengths and directing said radiation into the environment to be characteri~ed for the dilute component;

(b) means for detecting electromagnetic radiation emanating from and/or reflected from the environment and producing therefrom an electrical signal corresponding thereto;

(c) means for receiving said electrical signal and producing therefrom electrical signals at corresponding to said different wavelengths;

(d) means receiv~ng and operatively responsive to said electrical signals corresponding to said different wavelengths7 to establish absorbance equations responslve to sa~d electricat signals corresponding to said different ~
wavelengths, where~n absorbance at each of the wavelengths is expressed as a function of the relatlve intensities of the absorption contributions of the dilute and reference components and the concentrations of the dilute and reference components, and for calcutating the amounts of the absorbing species by solution of said absorbance equations; and (e) means for displaylng the calculated conoentrations of said d~lute and reference components.

In general, the number of wavelengths employed for concentration determination of the dilute component(s~ ln the , system will be equal to the number of absorbing species (~.eO, the number of dilute component~s) and the reference -12- 13273~

component). If the environment exhibits a flat non-spectfic baseline for background scattering due to wave1ength independent scattering~ an additional one wavelength muse be added, while if the baseline is linearly sloped an additional two wavelengths must be introduced, and if the scattering is non-linearly wavelength dependent, a number of extra wavelengths will be required to correct for the curvature of the baseline~ with increasing wavelength determinations providing increased accuracy to the determined concentrations.

In another aspect of the invention, in instances where an absorption band of the reference component is not present in the spectral region in which the selected dilute component ~i.e., the dilute component whose concentration i5 desired) has its absorption spectra, but such spectral region contains a band of a second dilute component, and a relatively near near spectral region exists in which the reference component and the second dilute component exhibit absorption, the concentration of the second dilute component may be determined and used as a bridging reference for deter~ination of the selected dilute component concentration by calibration of the selected dilute component against the second dilute component.

Appltcations of the present invention extend to the ent~re fteld of analytical chemistry utilizing a spectrophotometrtc means to determine concentrations in radiation scattering circumstances. Special attention shall be given in the following description to applications in the bio-medical field of non-invasive monitoring of metabolism.

One of these bio-medical applications is the determina-tion of the oxldation-reduction state of enzymes and the degree of oxygenation of the blood flowing through act~Yely metabolizing solid organs such as the brain, skeletal muscle~ the ltver, etc. An optical reflectance technique is used when such organs are too large for effective transillumt-nation. Such measurements provide much needed diagnoYtic lnformat~on on the adequacy of oxygen delivery and oxygen utiltzation to vttal organs at any moment in time and on an on-going basls.

,, 1327~

A second preferred application using a transillumination mode is the measurement of hemoglobin content in blood by the quantitative determination of the hemoglobin and water content of the pulsatile increases in blood content of a finger or earlobe as it pulsates with each heartbeat. Measurements can similarly be made of other blood-borne constituents (e.g.
glucose, lipids, cholesterol, carbon dioxide, etc.) that possess absorptive characteristics in the invisible, infrared or other parts of the electromagnet~c spectrum.

1~ In other general applications of the present invention, quantitative determinations can be made in light- or other radiation-scattering media, mixtures, or solutions, of any and all constituents possessing absorption characteristics in a spectral range in which a reference compound of known concentration also possesses significant absorption characteristics. Such applications are not limited to the visible or near infrared parts of the spectrum but can be performed in any part of the electromagnetic spectrum in which absorption of the radiation in either a transillumination or in a reflective mode is not so intense as to preclude the acquisition of a radiation signal strong enough for routine instrumental analysis.

~RIEF DESCRIPTION OF T~E DRAWINGS
.

Figure 1 is a ptot of the absorption spectrum of pure component water.

Figure 2 is a plot of the absorption spectrum of water in an environment exhibitin~ a flat baseline9 ~5 associated with rad~ation scattering.

Figure 3 is a ~lot of the absorption spectra of water.
-- 30 hemoglobin, and oxyhemoglobin, with a first baseline attributable to radiation scattering by the environment.

: , ~ ': ;

. . .
: .. .

,.

-14 1327~

Figure 4 is a plot of the absorption spectra of water~
hemoglobin, and oxyhemoglobin, with a linearly sloped baseline for wavelength-dependent radiation scattering in the environment.

Figure 5 is a plot of the absorption spectra of water, hemoglobin, and oxyhemoglobin, with a curved baseline evidencing wavelength-dependent radiation scattering in the environment.

Figure 6 is a plot of the absorption spectra of water, hemoglobin, oxyhemoglobin, and cytochrome a,a3 over the near-infrared range of about 700 to about 1400 nanometers, illustrating six wavelength values ~indicated by arrows) at which absorption measurements are taken in an illustrative syste~.

Figure 7 is a schematic illustration of an apparatus system for _ vivo determination of absorbent species in a human finger using transmission.

Figure 8 is a schematic illustration of the reflect3rce mode used with apparatus of the same general type shown in Figure 7, in a relatively denser body organ such dS an adult's head.

DETAIIEO DESCRIPTION OF THE INVENTION, METHOD0l06Y, ~: A~D PREFERRED EMBODIMENTS THEREOF

The present invention not only remedies the shortcomings of prior art infrared monitoring techniques but more generally provides a method to ascertain the concentration of a spectrophotometrically absorbing material under conditions ; where the pathlength of the llght beam 1s unknown~ This uncertainty exists ~n any light scatter1ng medium~ be it a ,:, , . ~; ~ .
- . ~ ... . ~ , ~.
. .

..

r~
-15- 132789~

liquid, a solid, or a gaseous mi~ture, in which multiple scattering lengthens the yeometricallY measurable optical pathlength. The present ~nvention overcomes this problem of undefined pathlength in media which have light-absorbing properties of their own in the electromagnetic radiation spectral region of the wavelength range in which the ~aterial whose concentration is to be determined is radiation absorptive.

The general broad applicability of the invention will be clear from the ensuing description of the invention, even though this description utilizes the near infrared spectral range in the illustrative examples hereinafter set forth. It should also be noted that in a spectral range (e.g., the visible range) in which the solution or any other known component does not absorb the impinged radiation, an indicator may in some instances be desirably added (specifically in 1n vitro determinations1 to the solution to establish the effective pathlength traversed by the photons.

The Beer Lambert law defines the basis of the spectrophoto~etric determinations of concentrations af radiation absorbing materials. It emphasizes that the absorption o~ light (or other radiation) depends on just two conditions: the efficiency of the molecules or atoms to absorb light and the number of such molecules or atoms in the light path. Two aspects should immediately be noted. The effiGiency of radiation absorption varies at d;fferent wavelengths ~aking it necessary to use a narrow "monochromatic" band of light and to state the efficiency with which the material absorbs that light. This parameter is called the extinction coefficient. The other aspect is the consequence that when the length of the light path through the selection or mixture is known, the only variable determining the number of molecu7es or atoms in the solution or mixture is the concentration. The entire field of quantitative analys~s using bench-top spectrophotometers applies this fact by us~ng ; .

~ -16- 1327~99 spectrophotometric vessels ("cuvettes") of known pathlength dimensions. Of course, clear solutions or ~as mixtures are required in such analysis to avoid lengthening of the optical pathlength by multiple light scattering. Scattering S media are also avoided whenever possible because the light 10st by scattering complicates the determination of light lost by absorption. Differentidl spectrophotometry has been developed to decrease errors due to mild scattering where light losses may occur but where pathlengthening is still a negligible factor. This technique employs either two closely spaced monochromatic wavelengths with different absorption coefficients, or two samples with equal light-scattering properties but one 7acking the absorbing material to be determined. It should be emphasized that these two well-known approaches to differential spectrophotometry can not, and do not, correct for optical pathlengthening by scattering.

The Beer-lambert law for solutions states that the logarithm of the fraction of the light absorbed by the dissolved material (the solute) equals the mo1ar extinction coefficient (E) of the solute times the concentration (c) times the pathlength td); or, as rewritten to determine an unknown concentration:
o C = - d log The quantity log Io/I is called the Absorbance tformerly the Optical Density). In these equat~ons Io is the original intensity of the light without an absorbing species and I the intensity after the beam has traversed the solution. In practice the intensitg Io is defined as the light falling on the detector of the spectrophotometer after it has traversed a cuvette containing pure solvent. The signal thus obtained is designated ~o and the signal obtained with an identical cuvette containing the solution of unknown concentration is designated I.
., .

.
;

r ~ ~_ -17- 132789~

The extinction coefficient is standardly given in the form of the amount of absorption produced over a 1 c~
pathlength by a 1 molar solution (one molecular weight of solute contained in one liter of solution). Values for molar extinction coeff~cients are commonly available in published tables and are usually given for wavelengths of maximal absorption.

Transillumination of a material with intense light-scattering properties results in a significant ~raction of photons falling on the detector having had a tortuous pathway that increased the distance traversed beyond the direct geometric length of the sample. In the most extreme mode.
Vi2., reflectance spectrophotmetry, absorbance spectra are taken utilizing the photons scattered out of the sample, either obllquely ~typically 90 angle observatlon) or back-scattered (same surface observation). Not only is the pathlength then unknown it 1s undefined especiallY ln the case of same surface observations. The mean depth of penetration before back scattering occurs is difficult or impossible to determine. In either case the effective pathlength is unknown and concentrdtions can not be determined by the Beer-lambert law.

The present invention substitutes a totally new approach to determine concentrations in various media - solutions, gas mixtures, and solids - an approach not predicated on pathlength but on simultaneous ~easurement of the amount of medium traversed. In solutions this method provides therefore a statement concerning the amount of solvent encountered. From this para~eter and the strength of absorption by the solute, the concentration of the solution can be derived. A
prerequisite is that the absorption bands occur at relatively closely-spaced wavelengths, scattering being wavelength dependent, especially in the visible and ultraviolet wavelengths regions of the speetrum.

i ~
,. , ~327g9~

Figure 1 shows the absorption spectrum of pure component water over a selected spectral region in the wavelength range of approximately 900-1100 nanometers. In a non-scattering environment containing only the reference component, the concentration of that component can be determined if the pathlength is known as is done in routine benchtop spectrophotometry, Conversely, if the concentration is known but the pathlength unknown the latter can be calculated instead, as should be clear from inspection of the Beer-'ambert law.

The absorbing component of Figure 1 can be either an indicator or the solvent itself. For ease of understandin~ it is instructive to consider the absorbing component to be an indicator. Such indicator must be present in known concentration. If an appropriate indicator component is lacking in the spectral range of the experiment, a suitable indicator which is absorptive in the spectral range of interest can be added to the solution at a known concentration. What also must be kno~n is the ~olar extinction coefficient of the indicator per centimeter pathlength at the measuring wave7ength. Intensity of the measured absorption peak then ~ndicates the length of the opticdl path. This indicator technique, although useful in bench top spectrophoto~etry, does clearly not lend itself well ~o in vivo monitoring. In that situation, however, the ubiquitous presence of water in biological tissues makes water the indicator of choice, just as in atmospheric applications nitrogen gas can fill this need.
.
It is relevant to point out that it is possible to express water in terms of its own concentration, which then provides a case formally identical to the indicator case -- above. Concentrations are often expressed either in grams pcr liter but more stringently in terms of molarity, i.e., moles .: : , . . .: , .
. :. ~ ~ .

::

-l9- 132~39~

per liter of solution which means ~rams per liter.
moleeular weight Since the weight of a liter of water is 1000 grams and its molecular weight is 18, pure water exists in the form of of 1000/18 = 55.6 molar "solution." Thus water is an indicator present at known concentration. However, in terms of the present discussion it is clarifying to note, for the special case of water or any other spectrophotometrically measurable solvent, that if it is known that a 1 cm cuvette filled w~th pure water shows an Absorbance of say 0.50 at a given wavelength, then an Absorbance value of 6.00 found at that same wavelength over an unknown pathlength shows that that pathlength must have been 12 cm. ~;

In this discussion, it may also be noted that the solute molecules do displace water molecules to a certain extent, lowering thereby its concentration in the solution as compared to its own concentration in pure water, This effect is, however, very small for the dilute solutions encountered in most situations, For example, the most concentrated salt component of blood (NaCl) produces a less than 2.6 ml increase in volume when dissolved in a liter of water~ Thereforet the water content of the resulting "physiological salt solution" is decreased by less than 0.26X. The errors thus created are far smaller than many uncertainties inherent in this or any other spectrophotometric methodology.

The parallel argument for the macro-molecular and so-called "formed" components of tissues, howe~er, is best reduced to terms of water content. Typicallyg for soft tlssues the water content is 85X percent. A correct~on of about 15 to 20~ is significant and could be applied in such cases. ~n that case the determined concentrations would be in terms of total tissue mass. This may, however, not be preferable to an expression in terms of total tissue water ... .
. .

-20- 13278~

which would result if the 15 to 20X correction is not applied.
It might be noted here that an identical approaeh is applicable to the analysis of gaseous mixtures that exhibit light scattering, i.e. such as haze9 steam, or clouds. For many atmospheric applications, nitrogen gas can serve as the reference component at known concentration, i.e., approximately 79%. Altitude or changes in barometric pressure do not decrease its usefulness since the N2 percentage will remain the same and a contaminant can still be determined in terms of percent, per mil or parts per million.
:
The above examples, referring to Figure 1, are based on totally clear, 1.e7, non-scattering, solutions or mixtures allowing a single wavelength approach. When scattering occurs, we must correct for light losses not related to absorption and for pathlengthening due to multiple scattering.
When several components produce overlapping absorption spectra this adds to their complexity. In the following four cases~
light scattering and absorption situations of increasing complexity will be anaylzed.

In Figure 2 the case for wavelength independent scattering is illustrated. Iight loss, equal throughout the relevant range of the spectrum, creates a ~baseline shift"
(identified by line B i'n Figure 2) which adds to the spectrum of the material to be determined (water in this example~. A
two wavelength differential approach is now indicated as the first, minimal step to cope with the problem tt amounts to subtracting signals at two wavelengths (~ I and ~ 2~ thereby eliminating the light lost by scattering.

For the spectrum of water in the wavelength range of 900-lI00 nanometers~ the extinction coefficients of water at two points along the wavelength range, e.g., at 980 and 1100 nanometers may be readily determined or obtained from tabulated values for the pure water component. The difference , in the observed Absorbance values at these respective wavelengths is a measure of either the amount of water encountered by the photon stream along their optical path, or in the case of a dilute solution (i.e~, a solution in which the water concentration remains approxi~ately 55.6 molar) a measure of the pathlen~th. Therefore, if the actual differential Absorbance, i.e., the difference in absorption values at the wavelength values of 980 and 1100, nanometers is determined, the effective pathlength of the test system can be derived by dividing the measured differential absorbance by the di~ferential absorbance value between 980 and 1100 nanometers for 1 cm oF water.

For example, Figure 2 shows the absorption spectrum for water (curve A), over the spectral range of from about 900 to about 1100 nanometers, with a flat baseline attributable to environmental scattering (baseline ~). The plot shows the peak of the absorption spectrum for water (curve A) at a first w~velength,~ 1, and a trough in the spectrum at a second wavelength ~2. The difference in absorption for water at the respective ~1 and ~ 2 values is indicated by the quantity Al - A2, as the differential absorption in an environment exhibiting background scatter producing a flat baselineO The practice of subtracting absorption signals at adjacent wavelength v~lues in this manner, where the known component differs significantly in its absorbance at the respective wavelengths, amounts to subtracting an existing wavelength-independent baseline of loss by scattering. The intensity of the remaining Al-A2 value provides a measure of the effective pathlength. What has not yet been discussed is the determination of the concentration of other absorbing components dissolved in the water and their effect on the water measurement.

, ., `: , ' ': 7 ', ~ ' , ' :, .

' ' ~ ,' ' ` ` .
;~

-~ -22- 13278~

Figure 3 shows absorption spectra ~or water (curve A1, oxyhemo~lobin ~curve C), and de-nxyhemoglobin (curve ~), against a flat baseline (curve ~) associated with scattering losses in the system. The derivation of the effective pathlength through the sample in the Figure 3 system is more complex since other absorption curves occur in the spectral range needed to determine the amount of the "known" or "reference~l material (i~e., water). For three absorbing species (water, hemoglobin, and oxyhemoglobin), a minimum of three wavelengths is required so that three absorbance equations or "algorithms~' can be established and solved for the unknowns, i.e., the contributions of the three absorber species. If a flat non-specific baseline exists (due to wavelength~independent scattering, as shown), a fourth wavelength must be added, to yield four equations to solve for the four unknown contributions. Solution of these four equations provides information on the amount of each material encountered. The known concentration of water (55.6 Molar) provides the opportunity to calculate the pathlength that must have been traversed and thereby enables the calculation of the concentration of the other components. In simplified terms, this can also be done by calculating the apparent effective pathlength from the water signal and deriving the ~ i concentration of the solute(s) in this manner.

A simi)ar type of complexity is produced by scattering that varies in intensity as a function of wavelength over the spectral region considered. Figure 4 is a plot of absorption spectra for water (curve A), oxyhemoglobin (curve C), and de-oxyhemoglobin (curve D), against a linear, sloped baseline (curve B~. In absorption systems of this type, a fifth wavelength must be introduced in the absorbance relationships to determine the degree of steepness of the slope.

In systems of the t~pe shown in Figure 5, wherein the ~ater, oxyhemQglo~in~ de-oxyhemoglobinp and baseline curves -23- ~3278~

are denoted by the letters A, C, D, and B respectively, and wherein the baseline indicates scattering of a curvilinear wavelength-dependent character, a number of extra wavelengths will be required to correct for the curvature of the baseline.
The higher the degree of accuracy required for the calculated concentrations, the greater the number of wavelength determinations that must be employed.

~ETHQDOLOGY

The method of the present invention has particular applicability to the determination of concentrations of blood components, such as the aforementioned hemoglobin and oxyhemoglobin, in body extremities where transillumination is employed, i.e., a source of radiation is impinged on the body part and collected at another exterior region of such body part. This methodology is applicable to body parts such as fingers, toes, earlobes, and other organs up to and including infants' heads~ Alternatively, reflectance spectrophotometry may be employed in portions of the body where transillumination is impractical due to the mass and optical density of the body part involved, e,g., the adult head, lungs, kidneys, etc.

The spectra of water (curve 1), hemoglobin (curve 2), and oxyhemoglobin ~curve 3) are shown in Figure 6 in the near infrared spectrum, over the range from about 700 to 1400 nanometers. In addition, the absorption curve of cytochrome a,a3 is illustrated for later discussion. These spectra were obtained by bench-top spectrophotometry using transillumination. The water spectrum is a so-called absolute spectrum, i.e., obtained from a cuvette full of water using an empty cuvette as a "blank" to determine lO at each wave length. The other spectra are of the hemoglobin and oxyhemoglobin compounds each dissolved in watery solution against a water blank.

.
:

-24- ~278~9 In order to make spectrophotometric determinations of the amount of a particular molecular species in a material o~ body organ of unknown volume and/or unknown optical pathlength, the minimum number of wavelengths required equals the number of absorbing molecular species. Additional wavelengths may be required if the geometry of the input and collection o~ the photons in the system being measured is complex, or varies from case to case, or if the wavelength dependence of scattering adds significantly differently at the two extremes of the spectral range used. In the following examples 10 increasing levels of complexity will be used as illustrations.

For the sake of simplicity, the first example is restricted to two dilute components of unknown concentration:
hemoglobin (Hb; also known as de-oxyhemoglobin) and oxyhemoglobin (HbO2). The environment to be analyzed for these dilute components will be a small body organ (from the 15 slze of a finger tip to a baby's head) and the constant media absorber, or "reference component". of known concentration in this example is water. The mode of observation is transillu-mination, with radiation input and collection (detection) points diametrically across the finger or head. When the 20 absorption curves do not overlap, a four wavelength meth~od can be practical with the tacit implieation that the "scattering baseline" is flat,i.e. that for the accuracy required the ;
scattering can be considered wavelength independent.

In Figure 6, the relative absorption contributions by 25 water and the two hemoglobin species are shown which approximate the normal relative contributions of these three absorbers in the human head. Similarly9 the cytochrome a,a3 contribution to the brain absorption is sho~n in approximate scale (with the oxidized cytochrome a,a3 spectrum shown as 30 curve 4, and the corresponding reduced enzyme spectrum shown as curve 4a), but it should be noted that in the tissues of o the finger only a negligible concentration of this enzyme ts present.

- , ~ ~.' 1327~99 Considering now the 900 to 1400 nm region of the near infrared spectral range, it is noted that the contributions of hemoglobin become negligible beyond 1150 nm approximately.
Thus, the effective optical paLhlength through a very srnall body part such as a finger can be deter~ined by measuring at S the trough, at 1270 nm, and at either adjacent peak, i.e. at approximately 1200 or 1400 nm. ~y subtracting the Absorbance values at the two wavelengths from each other~ the differential absorption value is found. When such differential absorption value is divided by the differential r 10 absorption ~extinction) coefficient for 1 cm of water, the apparent effective pathlength is determined. Assuming an equally flat scattering effect in the ad~acent 700 to 900 nm region and with the knowledge that the finger tissues do not contain a measurable amount of cytochrome a,a3 or other 15 species absorbing in th~s range, it is possib1e to calculate the exact amounts of the two hemoglobins by transilluminating with any two wavelengths in the 700 to 900 nm range and using the pathlengthening factor established above.

This ~ost simple case is often complicated by a number oF
20 factors. In the case of transillumination of a baby's h~ad, for example, the thickness of the baby's head, makes it impractical to use a wavelength such as 1400 nm at which the intensity of the water absorption results in so much light loss that the re~aining signal becomes difficult to detect.
In this case four wavelengths are chosen in the 900 to 1100 nm range and the absorption 1ntensities are measured. From previous experiments the contributlons to the extinction by each absorbing species plus that by the light scattering have been established using approximate models. Best suited ~0 for the latter are the corresponding body parts frnm corpses or appropriate animal models. These may be perfused -- alternately with hemoglobin-free solution, w~th oxy~en-free blood (for the Hb contribution), and with fully oxygenated , .

~ 13278~

blood (for the HbO2 contribution). In the last case a small amount of a poison such as cyanide, that inhibits 2 utilization by the tissue, is added to ensure that the blood remains oxygenated during the observation. Thus, when the environment is a corporeal corpse body moiety, a suitable indicator is an aqueous solution of indocyanine green dye.
The individual contributions of the three absorbing species and the scattering to the light losses (absorption) in the irradiated environment are then summed at each of the four wavelen~ths to derive the total absorption equations (referred to herein as "modified Beer-Lambert equations") at each wavelength. These equations take a form as shown below for the 980 nm wavelength:
Abs980 = x; Hb ~ xz HbO2 + z H20 ~ Scatter, wherein Abs980 is the Absorbance by the irradiated system of the incident radiation of wavelength 980 nm; Hb, HbO2, and H20 are the concentrations o~ hemoglobin, oxyhemoglobin, and water in the irradiated system: and the factors x, y, and z express the relative intensities of the absorption contributions of the associated components. The scattering term ("Scatter'~) has the factor 1, i.e., its contribution has been normalized.
The four wavelengths are chosen so that the three factors x, y, and z will show a considerable range of values. For example, a choice of 940, 980, 1030, and 1070 nm produces a good variation in relative values of the H20, Hb, and HbO2 contributions.
From this infQrmation algorithms are derived by matrix solution of the four unknowns (three absorbing component concentrations and the scattering losses) in the four equations. These resulting algorithms take the form Hb a Abs940 + b Abs980 + c Abs1060 + d Abs1100, wherein the values of the constants a, b, c, and d are numerically determined, yielding an expression for the amount of hemoglobin in the system. Similar expressions for HbOz and H20 are also obtained. The scattering contribution is not calculated Por the algorithms since it is irrelevant to the analytical assessment. The constants a, b, c, etc. can be positive or negative or larger or smaller than unity.

,~, .~ .;
.r ' ~
.
.

.

~ ,~,~

The above descrlbed methodology appllcable to hemoglobln and oxyhemoglobin concentrations, may be generali~ed and broadly stated as a spectrophotometric method for quantitatively determ1ning the concentration of a dilute S component in an environment containing the dilute component of known identity b~t of unknown concentration in combination with a reference component of known concentration, in which the following steps are carried out:

(a) directtng at the environment incident electromagnetic radiation at a number of wavelengths in a selected spectral region at which the dilute and/or reference components exhibit absorption for the electromagnetic radiation, the number of such wavelengths beiny determined by the number of dilute and reference components in the 5! enviranment, and the scattering characteristlcs of the env1ronment;

(b) determin~ng the absorbance by the environment of the electromagnetic radiat~on at the var10us wavelengths and the rclat1ve lntensit1es of the absorption contr~but~ons of the dilute and reference components and scattering losses from the env~ronment ~t each of such wavelengths;

(c~ at each of the aforementioned wavelengths, establishing absorption equations of the form:

Absw ~ ~ xjA1 + xR + S
i =l wherein: Absw is the absorbance ~y the environment, containing the dilute and reference components, of the incident electromagnetic radiation of wavelength w; xj is the relative 3~ intensity of the absorption contribution of the associated dilute component A~, and wherein terms of the form x~A; are set forth for each of the d1lute components; n ls the number of d11ute components; z 1s the rel~t1ve intensity of ~he absorpt10n contr~but10n of the reference component; R is the .
`~
. ;, .. .

-28- 1327~9~ .

concentrat~on of the reference component; and S is the normallzed sc~tterlng of the environment dt WdVelell9th W, thereby establishing for each of the aforement~oned wavelengths an absorbance equation, to yield a set of simultaneous equations whose number equals the number of dilute and reference components and the number of wavelengths required to characterlze the scattering of the environment;

(d) deriving algorithms by matrix solution of the aforement~oned simultaneous equations, such algorithms being of the form:
m [c~ = ~ aiAbsw;, i =l where1n: ~c~ 1s the concentration of the specif1c dt~ute or reference component; a; ls a determined numerical cor\stant; m is the number of wavelength determinations; and Absw~ ~s the absorbance at wavelength w; thereby establishing the ~ concentration of each of the dilute and reference co~ponents j 1n the environment.

In the previously described spec~fic example of determin1ng hemoglobin and oxyhemoglobin concentrations~ the H20 stgnal can be used to provide the pathlengthening fact~r for concentration calculations. Tn this manner the amount of Hb and Hb2 can be converted into concentrations in terms of 2S grams per l~ter. tt should be emphas~zed that thts term although a true concentration term refers to an inhomo~eneows disperse system. Not only is the hemoglobin sontained separately in the red blood cells, the observed water comprises not only the water in the blood pl~sma but also the 30 water ~n the cells and lymph spaces. Thus, the concentration units are not directly comparable to the usual ones (grams per lO0 ml) used clinically for bloodO The latter unlts can be obtalned, however, if we consider only the extra amount of blood that swells the finger with each heartbeat. This pulsatile signa1 ts used for example in the well-know~
technique of pulse oximetry.

" ~ .
':~
.; ~ .
,' r . ~
' ~ ~

^29- ~3~7~9~

In pulse oxlmetry, the color of the extra blood that swells the flnger with each pulse ~s determined, ~.e., the relative amounts of Hb and Hb2, thus providing a measure of the degree of oxygen saturation of the blood. This technique S does not provide information on the total a~ount of hemoglob~n in the blood. Adding a measure of the increase of water with each pulse can be accomplished by using an H20 absorption signal to ~easure each pulsatile increase in blood volume in the finger. In this way the actual hemoglobin concentration in the blood can be calculated~ The value of this number 1s general, and not limited to the specific organ (such as the ~inger) from which it was derived. The hemoglobin content thus determ~ned provides important diagnostic information for such conditions as anemia or polycythemia. In addition~ the hemoglobin content is required to determine the actual oxygen cantent o~ the blood since most of it is carried combined with hemoglobin in the form of oxyhemoglobin. With the hemoglobln content known and the percent of n2-saturation of the hemoglobin obtained by standard oximetry the much more signif~cant 02-oontent parameter can be calculated quite simply.

In other organs of larger diameter~ e.g. the head, limb musculature~ etc., transilluminatlon can be performed as~long as the thtckness of the tlssue does not preclude the acquls1t~on of an ~nstrumentally useful signal after the 25 transmitted rad~ation has passed through the tissue. In this respect it ts to be noted that one cm of water absorbs approxi~ately 80X of the 1400 nm near infrared radiat~on beamed throught lt. Transillumination of an infant's head of 5 cm diameter would show an extinction of approximately 99.97X
of the inc~dent near ~nfrared photons by absorption alone. In turb~d s~mples~ however, this loss can be increased and oYershadowed by losses due to scattering away from the detector and by additional absorption attributable to pathlengthening produced by the multiple scattering encountered by the photons eventually arriving at the detector. Although these light losses are very sever~D useful .
, .", .... .

:, ~L327~
signals have stlll been der~ved in such situations. The above example is, however, the limit of the present translllum1na_ ~ tion technique applied to human organs and body parts.

For ~ppl~cation to larger sol~d organs, a reflectance S mode must be employed in lieu of trans~llumination. As an example of the reflectance mode ~n appl~cation to the adult human head, light may suitably be entered at a first location on the forehead and collected at a second location on the forehead several centimeters distant from the first location, the collected photons havlng traversed scalp and skull and interacted with the cort~cal layers of the braln before being scattered out again. In this reflectance mode the crucial need of the present invention ls especia11y clearly illustrated, Cerebra1 content of Hb and HbO2 determined by reflectance measurements of radlation intensity can then be referenced to the total water observed, providing a measure of the a~ount of blood ln the brain related to the amount of water encountered by the photon stream from the entry to the collecting points.
The blood in the brain is highly compartmental~zed ln the erythrocites (red blood cells) wh1ch are, of course, lo~ated in the vascular space. The measured quantities are, howeYer, referred to the total water content, wh~ch ls present notably in the blood9 the brain cells and the meningial spaces, and also to a smaller extent in the bones and sk~n. The determined concentration therefore is expressed as amounts o~
: Hb (or HbO2) per amount of total "head water" or "tissue water." Most but not nearly all of this tissue water wlll have been that of the brain and is quite comparable to the water content of other soft tissues. Although this unusual expression appears at first somewhat awkward, in the context of the method of analysis of the present ~nvention it takes on a s~gnificance of its own. It should be noted that total water eontent of tissue and of the brain especially ~s a qu~te constant fraction of the total weight and thus such ~ater fulfills the requ1rement for a spectroscopic "reference ,.~

.~ . .

-3l-~32789~
component". It should be noted parenthetically that ~n cases - of edema a shi~t of water from blood and lymphspaces into the cells takes place. In the bratn, due to the nonelastic nature of the cran~um wh~ch for~s a prat~cally closed system, cerebral edema leads to increased intracranial pressure and conse~uently a forcing out of meningeal fluid and blood.
However total intracranial water content remains the same.
Incidentally, the loss of hemoglobin compared to the total water signal constitutes an excellent noninvasive indication of ~ntracranial pressure build-up, a potentially fatal affl~ction.

In the preced~ng example, absnrpt~on by the main oxygen ut~l~z1ng exzyme cytochrome c oxidase (~Icytochrome a,a3" or cyt a,~3") was lgnored. In view of th1s en2yme~s relat~vely smal1 contrlbutlon to the overall spectrum the resulting error produced ~n the hemog10btn data ls negllgible. In the event, however, that cyt a,a3 information is desired, two more appropriately spaced wavelengths are required, one ;n the 825 nm region and the other 1n the 865 nm region. Although a water band does not exist in that exact region the 980 nm peak is relatively near, and hemoglobin wh~ch absorbs in both reg~ons can be used as a bridging reference to deter~ine the enzyme concentrat~on.

PREFERRED EMBODIMENTS

Before describing the apparatus that can be utilized to make the measurements referred to above~ three caveats should be added to the general principles used in the above example.

The first ~s the fact that a narrow banded, 1.e., relat~vely monochromatic, light source is an 1mportant adv~ntage 1n construct~ng incisive algorithms. It is qu~te 30 clear from Fig. 6 that a photon source provid~ng a narrow band of l~ht~, say 5 nm ~idth, wi 11 prod~lce much less overlap between absorpt~on characteristlcs than a broad one, say w1th 50 nm spread of wavelengths among its photons.

. ~. . .
,~ ' ' .
. : '' ` ', - ~ ~

~ 32-~32789~
The second caveat follows directly from the flrst. When new light sources are used, differing even slightly in center wavelength or in bandw~dth, new algorithms must be constructed.

The third caveat is that the practice of the present 5 invention depends strongly on the development of either a means of translating the results ~n terms of accepted standards, such as spectrophoto~etric data in clear solutions, or on the de novo development of an extensive data base where accepted standards are not relevant, i.e., in heterogeneous systems such as the brain.

The apparatus system required to make the determination to practice the present invention ~ay suitably comprise the following component systems:

~a) means ~or producing light of varying wavelengths to enter ~nto the tissue or body part to be characterized for dilute components;

tb) means for detecting light emanating from or reflected from the body part;

c) m@ans for separating ampl1fylng and otherw~se treat~ng the signals obta~ned from the ltght source(s) at different wavelengths;

d1 ~eans for calculating the amo~nts of the absorbing species, us~ng the algorith~s der1ved for these llght sources; -and . . .
e) ~eans for d~playing the -results ~n dimensions of :.
concentrat~ons t~.e., amounts per amount of water or other ~ :
reference component) or in fract~onal quant~ties (such as :
a~ount of d~l~te component in relat~on to the amount of a reference component other than water ~n the body part to be meas~red for determ~nation of the dilute component of,unknown concentrat1On).

:

`` -33-Figure 7 1s a schematic dlagram of a spectrophotmetric system for quantitatively determining the concentration of blood dilute components in a human finyer with reference to water cont~1ned in the finger.

As prevlously alluded to, the amount of water encountered by the photons in a bo~ part must be estab1ished flrst, Th1s can be done either by measurements ln the spectr~l range ln ~hich water ~s pratlcally the only absoring species or by multiwavelength differentia1 spectrophotometry if other absorbing spec~es are present. In the human finger water may best be determined by suitable spectrophotometrlc de~erminat~ons on fingers of corpses. If such measurements must be made in a reg~on of absorption band overlap with hemoglob1n the blood must be re~laced by a sultable non-absorbing scatter~ng fluid to mimick the scattering by the red blood cells. Examples of suitable scattering fluids include fluorocarbon blood substitute solutions, calcium carbonate suspensions in saline solution, etc. In such ~bloodless~ systems, the spectrophotometric characteristics of the corpse f~ngers may be determined against pure water as a reference standard, to determine the apparent e~ective optical pathlength for radiation ln a given spectral reg~on, as pasçed through the finger to effect transillumination thereof. 8y numerous determinat10ns of such type, a database of opttcal lengths for various types of human fingers (erg., baby, adolescent, adult; Black, Caucasian, Oriental; etc.) may be developed. In this respect, it is to be noted that metanin is a pigmentation species which is present in varying degrees depending on the race and origins of the human sub~ect. It in some ~nstances may be desirable to treat melanin or other pigmentation-re)ated agents as additional absorbing species in the syste~ and to add further radiation directing and measurement steps of additional wavelength~s) in determining the concentration of the des1red dilute component ~n the corporeal system under study. Alternatively, routines for data acqu1sition and alogr~thms calculation can be incorporated as software in a microprocessor-based system to ~ .

: ~ . , ', -, ~

~ -3~-132789~
provlde a set of algorithms appropriate for a given patient at the start of a monitoring period.

Subsequently, the finger of a human test subject may be translllum1nated using radiation at various wavelength values, the number of which correspond to the degree of accuracy requ~red ~n the concentration to be determined for the dilute component of lnterese. The number of wavelengths at which measurements are made depends, as previously discussed, on the nu~ber of absorber species in the system (reference component and dllute co~ponent(s)) and the wavelength dependent character of the baseline lndicatlve of scat~ering losses ~n the en~lron~ent be~ng transillumtnated or irradiated for reflectance measurements~

From thc var10us absorbance and/or reflectance measurements at the selected wavelengths, a serles of slmu1taneous equattons of the type previously discussed are established. After solution by matr~x algebra~ algorithms are constructed ~n which the concentration is expressed as the summat~on of lndiv~dual absorbance values determlned at a g~ven wavelength multiplied by a dimensionless coeff1cient, 20 Once hav~ng determined the amount of the d~lute unknown ~
present 1n the sample ~l,e,, body organ) this can be ratioed aga1nst the uater slgnal to obtain a value 1n terms o~
conoentratlon ~n the total water of the organ.

In another embodiment~ total b~ckground subtract~on can 25 be used to obtaln only the increases ~n blood ln a pulsat~ng organ or body part suc~ as the f~nger, to ach1eve determlnatlon o~ the concentrat~on ~n the blood of such blood-borne spectes, e.g., hemoglob~n; waste products such as ammon~a, urea, ereat1nlne, and carbon d~oxide; substrates and metabolltes such as glucose~ lipids, and cholesterol; polsons such as carbon monoxlde, cyanide, and arsenic; etc.

. ~ .
.

. . .
'"~

-35- 1 3 2 78 ~g As applied to the apparatus shown in Figure 7, the finger 10 has mounted thereon two "optrode" assemblies 12, comprising a source optrode 14 and a co11ection optrode 16. The direct~on of trdns~llumination is lmmaterial: a path through the finger nail may be preferable in certain instances.

The source optrode 14 is connected v~a optical fiber cable 16 tc a l~ght source 18, wh~ch in this illustrative embud~ment comprises multiple sol~d state lasers energized by power supply 20 via power supply feedline 22. The laser source 18 emits electromagnetic radiations in the near infrared region, each of a monochromatic character, which are transmitted by the fiber optic cable 16 to the optrode 14 ~or impingement on the associated surface of finger 10, dnd transillumination thereof.

The resulting transmitted electromagnetic radiation is collected by the detector optrode 16 and passed via fiber optic cable 24 tD an appropriate transducer 26 which in this illustrative embodiment comprises a photomultiplier tube energized by high voltage power supply 28 via power supply feedline 30. The sensed transilluminated signal passing from optic21 flber cable 24 to the photomultiplier tube is q amplified therein and passed by signal transfer means 32 to the stgnal processing ~odule 34. fn the event that a low voltage powered, "solld state" detector of small size ~s employed, the detector can be ~ncorporated in the detector optrode. Fiber opt~c cable 24 is then replaced by an :
electr~cal cable d~rectly to ~he subject.

As shown, the f1ber optic cable 16 contains a small separate bundle branch li~e 36 which transmits a fraction of the ~onochromatic light from laser source 18 wh~ch is directly ~:
scattered back by the sktn~ It is coupled by cable branch --- line 36 to a photodiode 38, which transm~ts an electrical signal ~n signal wire 40 to the calcu1ation module 34, whlch may ~cr ex~mple compr~se a digital e1ectronic co~puter or may comprise a dedicated m~croprocessor unlt or un~ts.
~`

-36- ~3278~ .

In the computat1On module 34~ the electrical s~gnals transmitted by the photodiode 38 and photo~ultipl~er tube 26 are stored, prov~ding a measure of the incident and detected rad~aSion 1ntensities, together with stored or calculated systems p~ra~eters, From these var~ables and pre-programmed algorlthms, or by the aforementioned simultaneous absorbance equations established and solved by matrix solution to y1eld these algor1thms ~on llne"~ the concentration(s) of the selected dilute component(s) are calculated and expressed as amounts of such component related to amounts of water or other reference co~ponent in the finger syste~

In the preferred appl~cation of the ~nvent~on to physlologlcal systems, the successtve rad~atlon-d~rectlng and measurement steps must be carr~ed out ~n periods substant~ally much briefer than the metabol~c reaction kinet~cs of the corporeal environment.

The computation module may continuously prov1dc such concentrat~on data as output through deY1ces 34a compr~s~ng su~table meters and/or stripchart recorders, and the l~ke, and store the output concentration data for later access by digltal disc recording or si0ilar storage means assoclated with the computat~on module.

f~gure 8 shows a schematic dep~ction of a system whose components correspond to those shown in Figure 79 ~ith the exception of the optrodes 114 and 116~ which are arranged for reflectance mode spectrophotometry at spaced-apart reg~ons of the forehead of a human subject 110. All other system elements shown in F~gure 8 system elements are numbered ident~oally to their Figure 7 counterparts~ but with addltior of 100 to the reference numerals used in Figure 7~ :

; . ~, ....................... .

,' . , ' :' . . - . ~ , : ~ ~

In the Flgure 8 system, incident electromagnet1c rad1at10n 1s emitted from optrode 114 and provides photons capable of penetrating both the skin and bone layer as welt as the qray matter and wh1te matter of the sub~ect's head. Those photons which are reflected to the optrode 116 are sensed and the resulting detect10n s~gnal 1s transmitted by fiber opt~c cable 124 to the photodetection and calculation module co~ponents, as previously described in connection with Figure 7.
Although the 1nvent10n has been described with primary reference to the detect10n and determ~nat10n of concentrations of dilute components such as tissue components and blood-borne species 1n body parts (whole tissue/whole organ environments) such as fingers, hands, toes, feet, earlobes~
heads, and the 11ke, using near infrared radiation, it will be , apparent that the appl1cablity of the invention is not so lS lim1ted. The method of the invention may be applied to the determ1natlon of an~ d11ute component 1n an environment contain1ng a reference component of known concentration and in any range of the electromagnetic spectrum in whtch spectrophotometric absorbance techniques can be practiced.

Illustrat1ve examples of such alternative applicat~pns ~nclude but ~re not l~m1ted to, the measurement of ac1d ra1n const~tuents, carbon monox1de~ or other a~r pollut10n species 1n atmospher1c and ocean1c/riparian environments; and the detect10n of tox1c gas spec1es 1n sem1conductor manufactur1ng operations and industr~al gas pur1f1catlon processes.

Further9 wh11e the 1nvention has been shown and descrlbed w~th reference to 111ustrat1ve embodiments, 1t will be apparent th~t other var1at~ons, modificat10ns and embodiments are poss1ble, and all such apparent var1at~ans, mod1f1c~tlons and embod~ments ~re to be regarded as be1ng with the sp~rlt and scope o~ the present invent10n.
:~, ::' ., .. - :. , .. ..

-38- ~32789~

Best Mode for Carrylng Out the Invention The best mode for prac~ic~ng the present inventlon uti1~zes at least as many wavelengths of electromagnet~c radiat~on as the number of absorbing molecular components to be deter~ined in the environment to be analyzed. The ~avelengths are ~n selected regions wherein the components exhiblt absorption. Analysis of the components of a body organ envtronment comprises transillum~nation of the organ w~th the selected wavelengths, with the radiation input and detection potnts being placed diametrically across the organ, followed by determin~tion of the absorbence of each wavelength. Algor~thms are then derived from the absorption equations at each wavelength by matrix solution of the equattons for the unknowns. For larger, solid organs 15' reflected rad~ation may be measured and used for determination of the co~ponent concentrat~ons.

~ndustrlal_~ppllcab~liey The present ~nvent~on may be used for quantitative determtnation o~ dilute component concentrations where the dilute component 1s in an environment with a reference component of known concentration. For example, the 1nventlon may be used to determine the concentration of enzymes, prote~ns and metabolites ~n corporeal fluids. The concentratton of ac~d~c fumes or ga~eous components such as hydrogen sulftde, sulfuric acid, nitric acid and carbon monox~de in the atmosphere may also be quant~tat~vely deter~ned with the invention, as may salt concentrat~ons tn sea ~ater ~n the desalinatlon process~ and ozone ln waste water ozonatlon systems.

. ;

~, , , ., .

Claims (23)

1. A spectrophotometric method of quantitatively determining the concentration of a dilute component in an environment containing the dilute component of known identity but of unknown concentration in combination with a reference component of known concentration, by a series of successive, substantially contemporaneous measurements of at least one of transmitted and reflected radiation at selected wavelengths, comprising:

(a) determining the apparent effective pathlength in said environment;

(b) directing at said environment incident electromagnetic radiation of a first wavelength in a selected spectral region at which the dilute and/or reference component(s) exhibit absorption for the electromagnetic radiation;

(c) measuring the first wavelength radiation transmitted and/or reflected by the environment;

(d) directing at the environment incident electromagnetic radiation of at least one other wavelength in the selected spectral region at which the dilute and/or reference component(s) exhibit absorption of different relative intensities than for the first wavelength incident radiation;

(e) measuring the other wavelength radiation transmitted and/or reflected by the environment;

(f) determining extinction coefficient values for the dilute component at said first and other wavelengths in said environment; and (g) based on the apparent effective pathlength determined for the environment, the extinction coefficient values for the dilute component at said first, and other wavelengths, and the measured absorbed and/or reflected radiation at said wavelengths, determining the relative amount of the dilute component to the amount of the reference component, as the concentration of the dilute component in the environment.

2. A method according to claim 1, wherein said apparent effective pathlength of said environment is determined by the steps comprising:
(i) measuring the absorbance of said reference component at different selected wavelengths in said selected spectral region;

(ii) calculating the differential absorbance from the measured absorbance values;

(iii) determining the extinction coefficient values for said reference component at the different selected wavelengths of step (i);

(iv) calculating the differential extinction coefficient from the determined extinction coefficient values for said reference component; and (v) dividing the differential absorbance by the differential extinction coefficient to yield the apparent effective pathlength of said environment.

3. A method according to claim 1, wherein said determination of step (f) is effected by establishing simultaneous modified Beer-Lambert equations for each of said absorption/scattering measurement steps, and solving said equations for the concentrations of said dilute component and said reference component in said environment.

4. A method according to claim 1, wherein a second dilute component of unknown concentration is contained in said environment, and wherein electromagnetic radiation of a third wavelength in said selected spectral region is directed at said environment at which said second dilute component exhi-bits an absorption for the electromagnetic radiation, and the third wavelength radiation transmitted and/or reflected by the environment is measured, and employed to determine the relative amount of the second dilute component to the amount of the reference component.

5. A method according to claim 4, wherein three simultaneous modified Beer-Lambert equations are established for the concentration of the dilute components and reference component in said environment.

6. A method according to claim 1, wherein said environment effects scattering of said electromagnetic radiation which is independent of wavelength of said radiation, and wherein the transmitted and/or reflected radiation is measured at a single additional wavelength.

7. A method according to claim 1, wherein said environment effects scattering of said electromagnetic radiation to an extent which is of sloped linear relationship to wavelength, and wherein the transmitted and/or reflected radiation is measured at an additional two wavelengths.

8. A method according to claim 1, wherein said environment effects wavelength scattering of said electromagnetic radiation which is a non-linear function of wavelength, and wherein the transmitted and/or reflected radiation is measured at an additional at least three wavelengths.

9. A method according to claim 1, wherein said environment is a corporeal environment.

10. A method according to claim 9, wherein said environment is selected from the group consisting of corporeal tissue and corporeal organs.

11. A method according to claim 9, wherein the dilute component is selected from the group consisting of tissue components and blood-borne species.

12. A method according to claim 9, wherein the dilute component is selected from the group consisting of enzymes, metabolites, substrates, waste products and poisons.

13. A method according to claim 9, wherein the dilute component is selected from the group consisting of glucose, hemoglobin, oxyhemoglobin, and cytochrome a,a3.

14. A method according to claim 1, wherein said environment comprises an externally added indicator solution as a reference component.

15. A method according to claim 14, wherein said environment is a corporeal corpse body moiety, and said indicator is an aqueous solution of indocyanine green dye.

16. A method according to claim 1, wherein a first dilute component and reference component exhibit absorption for the electromagnetic radiation is said selected spectral region and a second dilute component does not exhibit absorption for the electromagnetic radiation in said spectral region, and wherein the first and second dilute components exhibit absorption for electromagnetic radiation in a second spectral region closely proximate to said selected spectral region and in which the reference component does not exhibit absorption for the electromagnetic radiation, comprising determining the relative amount of the first dilute component to the amount of the reference component in said selected spectral region as the concentration of the first dilute component in the environment, and utilizing the first dilute component whose concentration is thus determined, as a reference component for the second dilute component in said second spectral region.

17. A method according to claim 9, wherein said corporeal environment comprises a body portion selected from the group consisting of heads, fingers, hands, toes, feet, and earlobes.

18. A method according to claim 1, wherein said electromagnetic radiation is infrared radiation having a wavelength in the range of from about 700 to about 1400 nanometers.

19. A method according to claim 9, wherein said reference component is water.

20. A spectrophotometric method of quantitatively determining the concentration of a dilute component in an environment containing the dilute component of known identity but of unknown concentration in combination with a reference component of known concentration, comprising:

(a) directing at the environment incident electromagnetic radiation at a number of wavelengths in a selected spectral region at which at least one of the dilute and reference components exhibit absorption for the electromagnetic radiation, the number of said wavelengths being determined by the number of dilute and reference components in the environment, and the scattering characteristics of the environment;

(b) determining the absorbance by the environment of the electromagnetic radiation at the various wavelengths and the relative intensities of the absorption contributions of the dilute and reference components and scattering losses from the environment at each of said wavelengths;

(c) at each of said wavelengths, establishing absorption equations of the form:

wherein: Absw is the absorbance by the environment, containing the dilute and reference components, of the incident electromagnetic radiation of wavelength w; xi is the relative intensity of the absorption contribution of the associated dilute component Ai, and wherein terms of the form xiAi are set forth for each of the dilute components; n is the number of dilute component; z is the relative intensity of the absorption contribution of the reference component; R is the concentration of the reference component; and S is the normalized scattering of the environment at wavelength w;
thereby establishing for each of said wavelengths an absorbance equation, to yield a set of simultaneous equations whose number equals the number of dilute and reference components and the number of wavelengths required to characterize the scattering of the environment;
(d) deriving algorithms by matrix solution of said simultaneous equations, said algorithms being of the form:

, wherein: [c] is the concentration of the specific dilute or reference component; aj is a determined numerical constant; m is the number of said wavelength determinations; Abswj is the absorbance at wavelength w; thereby establishing the concentration of each of the dilute and reference components in the environment.

21. Apparatus for spectrophotometrically quantitatively determining the concentration of a dilute component in an environment containing the dilute component of known identity but of unknown concentration in combination with a reference component of known concentration, comprising:

(a) means for producing electromagnetic radiation of known wavelengths and directing said radiation into the environment to be characterized for the dilute component;

(b) means for detecting electromagnetic radiation received from the environment by at least one of emanating from the environment and being reflected from the environment, and producing from said radiation an electrical signal corresponding thereto;

(c) means for receiving said electrical signal and producing therefrom electrical signals corresponding to said different wavelengths;

(d) means receiving and operatively responsive to said electrical signals corresponding to said different wavelengths, to establish absorbance equations responsive to said electrical signals corresponding to said different wavelengths, wherein absorbance at each of the wavelengths is expressed as a function of the relative intensities of the absorption contributions of the dilute and reference components and the concentrations of the dilute and reference components, and for calculating the amounts of the dilute and reference components by solution of said absorbance equations;
and (e) means for displaying the calculated concentrations of said dilute and reference components.

22. A method according to claim 1, wherein subsequent to determination of the concentration of the dilute component in the environment, the environment is monitored for changes in said concentration.

23. A method according to claim 20, wherein subsequent to determination of the concentration of each of the dilute and reference components in the environment, the environment is monitored to determine changes in said concentrations.