 |
Claims  |
|
|
What is claimed is:
1. A method of measuring the concentration in a sample of a selected
substance which absorbs radiation in a particular region of the spectrum,
said sample further containing a plurality of sites which scatter
radiation in said particular region, said method comprising the steps of:
illuminating said sample with radiation from a radiation source which
provides radiation in said particular region of the spectrum, said
radiation source being selected and located to provide illumination of
said sample;
collecting radiation which is transmitted from said sample with a detector,
said detector being selected and located such that said detector collects
radiation only from a limited solid angle, said radiation being collected
from an area having a diameter or width at least comparable in size to the
thickness of said sample; and
determining concentration of said selected substance from said detected
radiation;
whereby the ratio of directly transmitted radiation to diffusely scattered
radiation collected by said detector from said sample is higher than if
said method is not used.
2. The method of claim 1 wherein said sample is illuminated with a beam of
restricted solid angle.
3. The method of claim 2 wherein said restricted solid angle of said
illuminating radiation is achieved by collimating the radiation from said
radiation source prior to illuminating said sample.
4. The method of claim 3 wherein said illuminating radiation is collimated
with collimating optics.
5. The method of claim 1 wherein said sample is illuminated with a beam
having a diameter or width at least comparable in size to the thickness of
said sample.
6. The method of claim 1 wherein said radiation source comprises a laser.
7. The method of claim 1 wherein said limited solid angle of said detected
radiation is achieved by excluding uncollimated radiation transmitted by
said sample prior to being collected by said detector.
8. The method of claim 7 wherein said exclusion of uncollimated transmitted
radiation is achieved using collimating optics.
9. The method of claim 1 wherein a first converging lens is selected and
located on said illumination side of said sample, with the focal point of
said first converging lens for said illuminating radiation located at a
location on the detector side of said sample.
10. The method of claim 9 wherein limiting the solid angle viewed by said
detector is achieved, at least in part, by locating an aperture between
said sample and said detector to enhance the ratio of directly transmitted
radiation to scattered radiation reaching said detector.
11. The method of claim 9 wherein limiting the solid angle viewed by said
detector is achieved, at least in part, by locating a second converging
lens between said sample and said detector, said second converging lens
having a focal point more distant from said sample than the focal point of
the first converging lens.
12. The method of claim 1 wherein said illuminating radiation comprises
broad spectrum illumination.
13. The method of claim 1 wherein said radiation source for illumination of
said sample comprises a plurality of radiation sources arranged to provide
congruent illumination.
14. The method of claim 1 wherein said detector comprises a plurality of
detectors arranged to provide congruent sampling.
15. The method of claim 1 wherein scattered radiation outside of said
limited solid angle is tested by an additional detector located off axis
to provide information obtainable from said scattered radiation.
16. The method of claim 1 wherein said particular region of the spectrum is
between about 700-1500 nm.
17. The method of claim 1 wherein said sample is a portion of a mammalian
body.
18. The method of claim 17 wherein said sample is a portion of a human
body.
19. The method of claim 18 wherein said portion of the human body is a
region of tissue selected so that person-to-person variations in tissue
thickness is minimized.
20. The method of claim 18 wherein said selected substance is selected from
the group consisting of glucose, glucose identifying substances, and
mixtures thereof.
21. The method of claim 18 wherein said selected substance is selected from
the group consisting of hemoglobin, deoxyhemoglobin and mixtures thereof.
22. The method of claim 21 wherein said concentration measurement is used
to derive a hematocrit.
23. The method of claim 21 wherein said concentration measurement is used
to derive oxygen saturation of the blood.
24. The method of claim 12 wherein said detector comprises a plurality of
detector units, at least two of said detector units having overlapping
spectral ranges.
25. A device for measuring the concentration in a sample of a selected
substance which absorbs radiation in a particular region of the spectrum,
said sample also containing a plurality of sites which scatter radiation
in said particular region, said device comprising:
a radiation source for illuminating said sample with radiation from said
particular region of the spectrum; and
a detector for collecting radiation which is transmitted by said sample,
said detector being selected and located such that said detector collects
radiation only from a limited solid angle with an area having a diameter
or width at least comparable in size to the thickness of said sample;
whereby the ratio of radiation collected by said detector which is directly
transmitted by said sample to that scattered by said sample is higher than
if said device is not used.
26. The device of claim 25 wherein said radiation source provides
illumination to said sample through a restricted solid angle.
27. The device of claim 26 wherein said radiation source comprises separate
optics means for restricting solid angle of said illuminating radiation
prior to illumination of said sample.
28. The device of claim 25 wherein said optics means for restricting said
solid angle comprises collimation optics for collimating said illuminating
radiation.
29. The device of claim 25 wherein said device further comprises
collimating means for collimating radiation transmitted by said sample.
30. The device of claim 29 in which said collimating means comprises
collimation optics which are located such that the ratio of said radiation
directly transmitted by said sample to radiation scattered by said sample
is higher than if said device is not used.
31. The device of claim 25 wherein said radiation source provides
illumination across a wide area at least comparable in diameter or width
to the thickness of said sample.
32. The device of claim 25 wherein said radiation source comprises a laser.
33. The device of claim 25 wherein said radiation source provides broad
spectrum illumination.
34. The device of claim 33 wherein said detector comprises a plurality of
detector units, at least two of said detector units having overlapping
spectral response.
35. The device of claim 25 wherein said selected region of the spectrum is
between 700-1500 nm.
36. The device of claim 25 wherein said restricted solid angle of
illumination is achieved using a converging lens selected and located on
the illumination side of said sample, with the focal point of said
converging lens located on the detector side of said sample.
37. The device of claim 25 wherein limiting the solid angle viewed by said
detector is achieved, at least in part, by locating an aperture between
said sample and said detector to enhance the ratio of directly transmitted
radiation to scattered radiation reaching said detector.
38. The device of claim 26 wherein restricting the solid angle illuminating
said sample is achieved, at least in part, by locating a lens that causes
a slight divergence of said beam between said radiation source and said
sample.
39. The device of claim 25 further comprising testing means for testing
radiation scattered outside said limited angle encompassed by said
detector.
40. The device of claim 25 wherein said sample is a portion of a mammalian
body.
41. The device of claim 40 wherein said sample is a portion of a human
body.
42. The device of claim 41 wherein said selected substance is selected from
the group consisting of glucose, glucose identifying substances, and
mixtures thereof.
43. The device of claim 41 wherein said selected substance is selected from
the group consisting of hemoglobin, deoxyhemoglobin and mixtures thereof.
44. The device of claim 43 wherein said concentration measurement is used
to derive a hematocrit.
45. The device of claim 43 wherein said concentration measurement is used
to derive oxygen saturation of the blood.
46. The device of claim 45 comprising an optical element which polarizes
said radiation prior to illumination of said sample.
47. The device of claim 46 wherein said detector further comprises an
analyzer which segregates polarized radiation directly transmitted by said
sample from depolarized radiation scattered from said scattering sites
within said sample.
48. The device of claim 46 further comprising testing means which collects
and tests depolarized radiation scattered from said sample.
49. A method of measuring the concentration in a tissue of a selected
substance which absorbs radiation in a particular region of the spectrum,
said tissue containing a plurality of sites which scatter radiation in
said particular region, said method comprising the steps of:
illuminating said tissue with radiation from a radiation source which
provides radiation in said particular region of the spectrum, said
radiation source being selected and located to provide illumination of
said tissue; and
collecting radiation which is transmitted from said tissue with a detector,
said detector being selected and located such that said detector collects
radiation only from an area having a diameter or width at least comparable
in size to the thickness.
50. A device for measuring the concentration in a tissue of a selected
substance which absorbs radiation in a particular region of the spectrum,
said tissue containing a plurality of sites which scatter radiation in
said particular region, said device comprising:
a radiation source for illuminating said tissue with radiation from said
particular region of the spectrum, said radiation source being selected
and located to provide illumination of said tissue; and
a detector for collecting radiation which is transmitted by said tissue,
said detector being selected and located such that said detector collects
radiation only from an area having a diameter or width at least comparable
in size to the thickness of said tissue. |
|
 |
|
 |
Claims |
|
|
|
Description |
|
|
BACKGROUND OF THE INVENTION
The present invention relates to improvements in optical systems and their
uses for the measurement of concentration and temperature in scattering
media, and the related discrimination of subsurface features. More
particularly, the invention provides methods and apparatus which minimize
the ratio of diffusely scattered radiation to directly transmitted
radiation reaching the detector(s) in optical concentration measurement
and imaging apparatus. The methods and apparatus of the invention have
special applicability to non-invasive testing, particularly for
concentration measurements of materials such as glucose and hemoglobin in
blood.
Recent literature is replete with articles describing attempts at
performing noninvasive testing using optical measurements (e.g., infrared
systems). Part of this expansion has been fueled by the spread of acquired
immunodeficiency disease syndrome (AIDS), and the associated fear among
public and health care personnel of AIDS. AIDS and other diseases such as
hepatitis are born in the blood and can be spread by improper practice of
invasive procedures. In addition, the diabetic population has also been
anxiously awaiting non-invasive test instruments for many years. Many
diabetics must test their blood glucose levels four or more time a day.
The modem battery powered instruments for home use require a finger prick
to obtain the sample. The extracted blood samples are placed on a
chemically-treated carder which is inserted into the instrument to obtain
a glucose reading. This finger prick is painful and can be objectionable
when required frequently. In addition, although the price has dropped
considerably on these instruments, the cost for the disposable and the
discomfort and health risk associated with having open bleeding is
undesirable.
Accordingly, a number of groups have recently tried to make non-invasive
instruments for testing a variety of analytes, particularly glucose. A
recent trend in non-invasive testing has been to explore the use of the
near infrared spectral region, primarily 700-1100 nm because this is the
spectral response range of the silicon detectors typically used in the
prior art. A wider wavelength range to .about.1800 nm can be accessed by
the addition of germanium and/or InGaAs detectors, and useful measurements
can be made into the 2500 nm range with InSb or other detectors. The
region below .about.1400 nm is the most useful in transmission, as tissue
is transparent enough there to allow high enough photon flux for accurate
detection. Above 1400 nm, the strong absorption of water limits the
penetration depth of tissue, so that useful measurements are typically
made in reflectance geometry. Below 1100 nn, the penetration of the light
is sufficient that the signal modulation during the arterial pulse can be
measured comfortably in both transmission and reflectance geometries.
Above 1400 nm, such pulsatile measurements are extremely difficult in
transmission due to low intensity, and similarly difficult in reflectance
because the light does not penetrate deeply enough to sample the pulsatile
capillary beds.
Most of the non-invasive testing work has been carried out using classic
spectrophotometric methods, such as a set of narrow wavelengths sources,
or scanning spectrophotometers which scan wavelength by wavelength across
a broad spectrum. The data obtained from these methods are spectra which
then require substantial data processing to eliminate background;
accordingly, the papers are replete with data analysis techniques utilized
to glean the pertinent information. Examples of this type of testing
includes the work by Clarke, see U.S. Pat. No. 5,054,487; and primarily
the work by Rosenthal et al., see e.g., U.S. Pat. No. 5,028,787. Although
the Clarke work uses reflectance spectra and the Rosenthal work uses
primarily transmission spectra, both rely on obtaining near infrared
spectrophotometric data.
The major successful application of noninvasive testing is the measurement
of hemoglobin oxygen saturation with pulse oximetry. The most common
method compares the percentage modulation of the intensity of light
traversing a body part at two wavelengths chosen so that the ratio of
their respective modulations is a relatively strong function of oxygen
saturation. The observed change in this ratio is relatively large because
the two hemoglobin species involved have both high enough concentrations
and specific absorptions that they dominate the creation of the pulsatile
signal components at the wavelengths of interest. As a result, the ratio
of modulations can be attributed substantially to the two hemoglobins
alone, and only needs to be measured to the order of 0.1% in order to
achieve clinically significant detection limits with acceptable
universality of calibration.
The optical system in typical pulse oximeters have two or more LED emitters
placed side-by-side on one side of a finger, and a single detector
receiving the radiation on the other side of the finger. Some more recent
systems have the detector on the same side of the tissue as the emitters,
with baffles preventing the direct illumination of the detector by the
sources. As the sources are physically small and optically displaced from
each other and the detector, the light from each detector enters the
tissue at slightly different locations, and therefore travel different
paths through tissue to the detector.
Despite its relatively large signal levels, pulse oximetry has well-known
difficulties such as the selection of an adequately vascular sampling site
on each individual and variability of the results with motion of the site
and breathing by the patient, as well as sensitivity to changes in blood
pressure, heart rate, temperature, and tissue hydration. Disturbances such
as motion and breathing artifacts typically appear as statistical
outriders, i.e., as measurements which fall well off the "average"
calibration curve of the instrument obtained from a group of individuals
breathing controlled gas mixtures to vary their oxygen saturation.
The calibration of a pulse oximeter is subject to these same error sources;
it is not uncommon to find site-to-site variations on the same individual,
with results that suggest that the calibration curve even varies, for
example, with the absolute magnitude of the pulsatile signal modulation.
The effort to obtain a meaningful universal calibration is clearly at odds
with intra- and inter-individual physiological variations.
Despite recent efforts to improve the measurement S/N by increasing source
intensities and lowering detector noise, as well as increasing the number
of detectors, the frequency of outriders and the universality of
calibration have not improved substantially. Thus it is clear that while
the light traversing the tissue is being measured more precisely, the
site- and physiologically-induced variability has not been improved
significantly below the 0.1% level needed for the measurement of oxygen
saturation.
While these physical and physiological interferences are marginally
acceptable for oxygen saturation measurements, they set a lower limit of
detectivity that is too high for other clinical analytes such as glucose
and cholesterol for which the combination of concentration and specific
absorption requires optical measurements to be made 100-1000 times more
precise than for the hemoglobins used in pulse oximetry. The hemoglobins,
which in themselves are difficult to calibrate in the presence of these
site- and physiologically specific limitations comprise a major background
interference for the measurement of such trace constituents as glucose.
The optical systems employed for these lower concentration analytes
naturally drew on the experience of pulse oximetry, and typically employ
similar arrangements of a plurality of slightly displaced LED's to extend
the wavelengths sampled, or which use fiber optics to carry light to and
from the sources and/or spectrometers which perform the separation of the
signal into the different wavelengths employed. Displacement of the
sources and wide numerical apertures for the light entering and leaving
the tissue enhance the likelihood that different detected wavelengths will
have sampled different portions of the medium. Many of the physiological
interferences to accurate measurement are mediated by differences in the
mean paths traced by light of different wavelength in traversing the
intervening tissue between light source›s! and detector›s!. These path
variations are produced, in part, by light scattering in the tissue, which
varies with the wavelength of the light and which makes photons follow a
jagged overall path from scattering to scattering. The detected signals
are a complicated function of both the scattering and the total absorption
of all constituents along the longer total path of the light. Thus, the
present optical systems used for noninvasive measurement allow and perhaps
even encourage light of different wavelengths to travel different paths
through the tissue, sampling lateral and axial tissue inhomogeneities
differently.
This situation violates a fundamental premise of all optical non-invasive
measurement methods; namely, that the light intensity which is measured in
the individual detection channels can be attributed to the analyte and not
to any difference in tissue sampling. Tissue inhomogeneity produces
wavelength-dependent spreading of the light which ultimately reaches the
detectors, and in the extreme of high scattering and large inhomogeneity,
the mixture of detector signals becomes an uncontrollable and
uncalibratable average response to the physiological and biochemical
conditions at the sampled site.
In addition the existing noninvasive art has employed spectrophotometric
methods which limit the intensity of light detected in the individual
resolution elements, and which also apply the method in a way which uses
the available spectral information inefficiently. These methods were
conceived primarily for accurate determination of narrow band spectral
structures rather than for discriminating the presence of weak broadband
features in strong broadband backgrounds that characterize the noninvasive
measurement problem for constituents such as glucose. The multivariate
analysis mathematics required to separate the analyte signature from
strongly overlapping interferent signatures also introduces an error
propagation penalty that compounds the intensity limitation by increasing
the impact of detector noise on the calculated measure of concentration.
Improvements that enhance the solution of problems of interference in
broadband spectra, by obtaining different raw data, are described in U.S.
Pat. No. 5,321,265 (the "Block 265 Patent"). This patent sets forth a
different approach in non-invasive testing as compared with the prior
instruments and methods. As noted, substantially all workers in the
non-invasive testing field prior to the Block '265 Patent were using
classic spectrophotometric instrumentation and substantial processing in
an attempt to resolve the low resolution features from the background.
However, the spectra of analytes such as glucose in a human body are not
discrete high resolution features which spectrophotometric instruments
were originally designed to measure but rather have a few low resolutions
features with much of the information contained in subtle variations of
the detected intensity as a function of wavelength. As such, these spectra
appear more like reflectance spectra of colored objects in the visible
region. The Block '265 Patent teaches the use of an analog of human color
perception to obtain meaningful data by means of methods and apparatus
which utilize overlapping, broadbeam detectors to mimic the spectral
response characteristics of the human retinal cones, but translated into
the near-infrared. This approach, which is radically different than
classic spectrophotometric measurements, provides advantageous effects in
determining the concentration of glucose and other similar materials in an
aqueous solution and is particularly advantageous for use with scattering
media such as tissue where it also provide the added advantage of higher
light flux at the detectors so that the intrinsic shotnoise limitation as
a percentage of the total signal intensities is reduced.
U.S. patent applications Ser. Nos. 08/130,257, 08/182,572 and 08/333,758,
the disclosures of which are incorporated herein by reference, all
disclose improvements in the basic techniques and apparatus described in
the Block '265 Patent. These improvements include the concepts of
congruent illumination and detection of light emerging from the sampled
tissue site, pulsatile processing, modulation of illumination sources as a
means of eliminating unwanted radiation, the use of non-overlapping broad
beam radiation as well as overlapping radiation, and a number of
variations thereon. These applications make it clear, in part, that a
variety of techniques are useful (and in some instances may be necessary)
to deal with the problems encountered in non-invasive measurement of
analyte concentration in tissue or other scattering media. Many of these
problems arise from the fact that scattering media exhibit higher
effective path lengths than their physical dimensions because of
scattering by the samples themselves. In fact, the samples, such as human
tissue, act as if they are formed of a plurality of scattering sites or
centers in the sample. Techniques such as the congruent illumination and
congruent detection described in these patent applications equalize the
acceptance angles and distances traveled by light of different wavelengths
outside the scattering media. Technically, this is achieved by locating
all the illumination sources and/or detectors so that the path lengths and
angles between the media and the detector(s)/source(s) are equal, so that
the detectors or radiation sources act as if they were optically
superimposed.
However, the desired congruency of detected light is degraded within the
observed media because the multiple scatterings of light spread the light
beam to adjacent regions in a way which is strongly wavelength dependent.
If the scattering media is inhomogeneous, the result of this spreading is
to mix light from these adjacent structures in relative amounts which are
dependent on wavelength. One object of the present invention is to reduce
this disturbing effect by refining the launch and detection optics to
limit their angular acceptance ranges.
It has long been known that a certain portion of the illuminating radiation
survives transit across a turbid sample without being either scattered or
absorbed, while a much larger portion is scattered in all directions. The
more scattering a particular photon undergoes the longer the integrated
path it follows, and the longer the time that elapses before it emerges
from of the sample. Some groups have attempted to reduce the deleterious
effects of scattering by using pulsed sources and time gating the
detection so as to view the sample only in light which has undergone few
scatterings. What is measured is a "snapshot" of the sample in light
starting at the time of flight for an unscattered beam, and extending long
enough in time to obtain sufficient signal for the desired analysis
without including much scattered radiation. When the time gate is short,
"ballistic" or "snakelike" photons which have undergone no or few
scatterings along their path are selected, and shadowgram images similar
to those commonly obtained with x-ray's can be obtained.
This approach, however, requires complicated apparatus, and in addition to
the intensity limitation from the short time-gate after each pulse of the
light source, adds a further limitation on the number of detected photons
because the duty-cycle of the pulsed source is low compared to the
continuous source of the present invention. Other research groups such as
Wist et al., IEEE Transactions on Medical Imaging, 12 (4) 751-757 (1993),
have demonstrated that shadowgram-type images can be obtained by severely
restricting the angular acceptance range of detected photons about the
forward direction, essentially demonstrating that doing so limits the
detection to "ballistic" or "snakelike" photons. The Wist et al. apparatus
generates a geometrically narrow beam which is raster-scanned across the
sample, at a first wavelength, and then generates new images at changed
wavelengths. The work of this group, however, also demonstrates a severe
limitation on the total flux of transmitted photons which make it
inapplicable to the detection of trace constituents in scattering media.
Other workers such as Schmitt et. al., SPIE 1641, 150-161, (1992), have
demonstrated advantages for using collimated input and output light on in
vitro phantoms that simulate some of the light scattering properties of
turbid media, but the transmitted intensity limitation of their system
when it was applied to real in vivo measurements made it necessary to
change the system design away from this collimated approach. One
difficulty appears to be that their in vitro system was designed to
"approximate the plane-parallel conditions under which ›the theoretical!
photon diffusion model was derived," rather than addressing the
characteristics of the in vivo sample. Schmitt's collimated system was
designed to approximate a "collimated beam of infinite extent" by
establishing a finite incident beam of light traversing tissues and
confining the collimated detection to a small central region on the exit
side, apparently in order to eliminate unwanted edge effects. In addition,
the narrowband sources and detector used limited the transmitted
intensity.
The failure of Schmitt's design was that insufficient photon flux was
available at the detector, so that this system was abandoned for his in
vivo work. Instead, Schmitt's in vivo apparatus employed a fiber optic
that launched light into the tissue at its large (.about.50 degree)
numerical aperture, and an integrating detector on the opposite side of
the tissue receiving light through almost the whole hemisphere. Even then,
as noted in his article, the system had inadequate light intensity for the
measurement he was attempting. His work thus vividly illustrates the light
transmission limitations of real tissue that characterizes the prior art.
Thus, it is a specific object of the present invention to balance the light
collection efficiency and spatial resolution of the optical sampling
system viewing scattering media to simultaneously achieve high detected
light intensity and equality of response, as a function of wavelength, to
inhomogeneous inclusions within the media. This is accomplished by
selecting optical configurations of sources, detectors, and intervening
optical elements to minimize the effect of tissue inhomogeneities on the
relative changes in signal strengths in each of the different detectors
due to the presence of analyte.
It is a further object of the invention to achieve this balance in a way
which improves the repeatability of the measurements from site-to-site on
a given individual in the presence of disturbances such as motion,
breathing, hydration, and the like, with the Ultimate objective to achieve
universal calibratability of the measurement across subject in the
presence of such disturbances.
A related object of the invention is to provide a method of non-invasive
concentration measurement in a scattering media which increases the ratio
of direct collimated radiation to diffusely scattered radiation reaching
the detector, while maintained high integrated light intensity at the
detectors.
Another object in the invention is provide an apparatus for non-invasive
concentration measurements which maximizes the ratio of direct collimated
radiation to diffusely scattered radiation while maintaining high
integrated light intensity.
A further object of the invention is to facilitate the use of tighter
collimation by increasing the number of photons received in the individual
detector resolution elements through broadening their wavelength
acceptance range.
A similar objective of the invention is to facilitate the use of tighter
collimation by increasing the number of photons received by individual
detector resolution elements through increasing their surface area while
maintaining their congruency.
Yet another object of the invention is to further facilitate the use of
tighter collimation by the use of overlapping broadband detector
resolution elements in an analog of human color perception to combine
increased photon flux with more efficient separation of similar broad
analyte and interferent spectral features.
Consequently, it is a specific object of this invention to select optical
configurations of sources, detectors, and intervening optical elements to
minimize the effect of tissue inhomogeneities on the relative changes in
signal strengths in each of the different detectors due to the presence of
analyte.
It is a still further object of this invention to adjust the optical
interface to take maximum advantage of the natural spreading
characteristics of the light distribution patterns in tissue in maximizing
the S/N of the determination.
These and other objects which features the invention will be apparent from
the detailed description and the drawing.
SUMMARY OF THE INVENTION
The present invention features methods and apparatus for measuring
concentration in a sample which contains a plurality of radiation
scattering sites, and for measuring the distribution of concentration
and/or temperature within a sample when employed with imaging detectors.
The methods and apparatus can also be utilized for discrimination of
subsurface features through shadowgram generation. This procedure is also
useful for detection of temperature inhomogeneities. The methods and
apparatus of the invention employ means for restricting the solid angle of
illumination and/or collection, e.g., by collimation of the radiation to
minimize the amount of scattered radiation collected, employ polarization
of the illuminating radiation to differentiate scattered from unscattered
radiation, or a combination of the two. The methods and apparatus of the
invention provide more reproducible measurements on scattering media and
are particularly well suited to non-invasive testing of tissue for
materials such as glucose and drugs of abuse.
More particularly, the present invention provides a method of measuring the
concentration in a sample of a selected substance which absorbs radiation
in a particular region of the spectrum. The sample containing the
substance of interest also contains a plurality of sites which scatter
radiation in the same particular region of the spectrum. The method has
the steps of illuminating the sample with broad geometric area
illumination within a particular region of the spectrum (preferably using
broad spectrum radiation) where the substance of interest has absorption,
with the illumination and detection solid angles restricted, and with both
said solid angles extending over a geometrically wide surface
cross-section. The term "broad spectrum illumination" as used herein means
and implies that the wavelength of the illumination covers a substantial
portion of the region of the spectrum in which there is absorption by the
selected substance. Normally, the illumination is greater than 50 nm wide,
and if the substance of interest has absorbance at several wavelengths in
the particular region, it preferably is wide enough to cover all of the
absorption bands.
After leaving the sample, radiation which is transmitted or reflected from
the sample is collected with a detector, the detector being selected and
located such that the each resolution element of the detector collects
radiation only from a limited solid angle extending over a relatively wide
area of the viewed surface. The restricted solid angle illumination also
extends over a relatively wide area of the illuminated surface. The term
"wide" as used herein implies a beam width that is comparable or larger in
size than the thickness of the tissue being viewed. The term "comparable"
means and implies that the width of the beam or viewing area is at least
half the thickness of the sample or tissue. That thickness is itself
preferably restricted to be not much deeper than several "natural" 1/e
penetration lengths, the depth over which the diffuse radiation photon
density falls to a few percent of its maximum value near the entrance
surface of the medium. For typical tissues viewed in the 700-1400 nm
wavelength range, these preferred thicknesses are of the order of a few
mm, and the surface areas through which the light enters and exits are
both in the 5-10 mm range.
The terms "restricted solid angle" and "limited solid angle" as used herein
imply that the type of detector or illuminating radiation, which may
include some form of filtering and/beam focusing apparatus, limits the
angle over which the illumination or detection occurs. Preferably, the
illumination and or detection is restricted to a solid angle of about
10.degree. or less from the central illumination beam axis. This
configuration maximizes the ratio of directly transmitted radiation to
scattered radiation collected by the detector from the sample. Preferably,
the restricted solid angle of illumination is achieved by collimating the
radiation from the radiation source prior to illuminating the sample, most
preferably with collimating optics such as described herein.
Alternatively, a laser or another source which provides restricted
divergence illumination may be used without the necessity of some type of
collimating optics.
Similarly, the preferred detector limits the solid angle of t | | |
