U.S. patent application number 15/118384 was filed with the patent office on 2017-06-15 for intraoperative guidance system for tumor surgery.
This patent application is currently assigned to The Florida International University Board of Trustees. The applicant listed for this patent is The Florida International University Board of Trustees. Invention is credited to Wei-Chiang LIN, Yinchen SONG.
Application Number | 20170164837 15/118384 |
Document ID | / |
Family ID | 53800747 |
Filed Date | 2017-06-15 |
United States Patent
Application |
20170164837 |
Kind Code |
A1 |
LIN; Wei-Chiang ; et
al. |
June 15, 2017 |
INTRAOPERATIVE GUIDANCE SYSTEM FOR TUMOR SURGERY
Abstract
The current invention pertains to a system and methods of
identifying a site in a tissue of a patient as neoplastic or
normal. The system comprises a source of electromagnetic signals;
an optical probe which delivers the electromagnetic signals to a
working end of the probe; a spectrometer which acquires diffuse
reflectance electromagnetic signals returned from the site probed
by the optical probe. The spectrometer processes the diffuse
reflectance signals to produce a diffuse reflectance spectra which
is transmitted to a system controller programmed to analyze the
diffuse reflectance spectra to calculate hemoglobin concentration,
hemoglobin oxygenation, and/or diffuse reflectance intensity of
signals having wavelength of about 700 nm. These parameters are
used to identify the site as neoplastic or normal. The system of
current invention can be used in identifying neoplastic sites in
brain in an intraoperative manner, for example, during a pediatric
brain surgery.
Inventors: |
LIN; Wei-Chiang; (Coral
Gables, FL) ; SONG; Yinchen; (Lebanon, NH) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
The Florida International University Board of Trustees |
Miami |
MI |
US |
|
|
Assignee: |
The Florida International
University Board of Trustees
Miami
FL
|
Family ID: |
53800747 |
Appl. No.: |
15/118384 |
Filed: |
February 11, 2015 |
PCT Filed: |
February 11, 2015 |
PCT NO: |
PCT/US2015/015444 |
371 Date: |
August 11, 2016 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61938376 |
Feb 11, 2014 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G16H 30/40 20180101;
A61B 5/0075 20130101; A61B 5/14553 20130101; A61B 90/13 20160201;
A61B 2034/2055 20160201; A61B 34/20 20160201; A61B 2576/026
20130101 |
International
Class: |
A61B 5/00 20060101
A61B005/00; A61B 90/13 20060101 A61B090/13; A61B 34/20 20060101
A61B034/20; A61B 5/1455 20060101 A61B005/1455 |
Goverment Interests
GOVERNMENT SUPPORT
[0001] The subject invention was made with government support under
a research project supported by National Institute of Health under
Grant No. R15CA173617-01A1. The government has certain rights in
this invention.
Claims
1-65. (canceled)
66. A system for identifying a site in a tissue of a patient as
neoplastic or normal, the system comprising: a) a source of
electromagnetic signals; b) a probe operably connected to the
source of electromagnetic signals, for delivering the
electromagnetic signals to a tissue site; c) a spectrometer which
acquires diffuse reflectance electromagnetic signals returned from
the tissue site and produces a diffuse reflectance spectra from the
returned diffuse reflectance electromagnetic signals; and d) a
system controller operably connected to the spectrometer, wherein
the system controller: i) analyzes the diffuse reflectance spectra
to calculate the relative ratio of absorption coefficient
.sub..mu.a and reduced scattering cooefficient .mu..sub.s between
450 nm and 650 nm (.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.)); ii)
analyzes .mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) with respect to
those from normal reference tissue sites to identify the tissue
site as neoplastic or normal; and iii) determines the site in the
tissue of the patient in the tissue of the patient as neoplastic
if:
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.)>1+x.times..mu..sub.a(.lamda.)/-
.mu..sub.s(.lamda.).sub.N.sub._.sub.std or
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.)<1-x.times..mu..sub.a(.lamda.)/-
.mu..sub.s(.lamda.).sub.N.sub._.sub.std, wherein x is an adjustable
cutoff for .mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) and
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.).sub.N.sub._.sub.std is
standard deviation of .mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) from
a plurality of known noiinal sites.
67. The system of claim 66, wherein the tissue is brain.
68. The system of claim 66, wherein the spectrometer acquires the
diffuse reflectance electromagnetic signals returned from the site
at a rate of about 33 Hz.
69. The system of claim 66, wherein the the site is smaller than
about 5 mm in diameter and thinner than about 2 mm in
thickness.
70. A system for identifying a site in a tissue of a patient as
neoplastic or normal, the system comprising: a) a source of
electromagnetic signals; b) a probe operably connected to the
source of electromagnetic signals, for delivering the
electromagnetic signals to a tissue site; c) a spectrometer which
acquires diffuse reflectance electromagnetic signals returned from
the tissue site and produces a diffuse reflectance spectra from the
returned diffuse reflectance electromagnetic signals; and d) a
system controller operably connected to the spectrometer, wherein
the system controller: i) analyzes the diffuse reflectance spectra
to calculate the absolute hemoglobin oxygenation (SatO.sub.2); ii)
analyzes SatO.sub.2 with respect to those from normal reference
tissue sites to identify the tissue site as neoplastic or normal;
and iii) determines the site in the tissue of the patient as
neoplastic if: SatO.sub.2>1+y.times.SatO.sub.2 N.sub._.sub.std
or SatO.sub.2<1-y.times.SatO.sub.2 N.sub._.sub.std, wherein y is
an adjustable cutoff for SatO.sub.2 and SatO.sub.2 N.sub._.sub.std
is standard deviation of SatO.sub.2 from the plurality of known
normal sites.
71. The system of claim 70, wherein the tissue is brain.
72. The system of claim 70, wherein the spectrometer acquires the
diffuse reflectance electromagnetic signals returned from the site
at a rate of about 33 Hz.
73. The system of claim 70, wherein the the site is smaller than
about 5 mm in diameter and thinner than about 2 mm in
thickness.
74. A system for identifying a site in a tissue of a patient as
neoplastic or normal, the system comprising: a) a source of
electromagnetic signals; b) a probe operably connected to the
source of electromagnetic signals, for delivering the
electromagnetic signals to a tissue site; c) a spectrometer which
acquires diffuse reflectance electromagnetic signals returned from
the tissue site and produces a diffuse reflectance spectra from the
returned diffuse reflectance electromagnetic signals; and d) a
system controller operably connected to the spectrometer, wherein
the system controller: i) analyzes the diffuse reflectance spectra
to calculate the absolute diffuse reflectance intensity of signals
having wavelength of about 700 nm (Rd(700)); ii) analyzes Rd(700)
with respect to those from normal reference tissue sites to
identify the tissue site as neoplastic or normal; and iii)
determines the site in the tissue of the patient as neoplastic if:
Rd(700)>1+z.times.Rd(700).sub.N.sub._.sub.std or
Rd(700)<1-z.times.Rd(700).sub.N.sub._.sub.std wherein z is an
adjustable cutoff and Rd(700) and Rd(700).sub.N.sub._.sub.Std is
standard deviation for diffuse reflectance intensity of signals
having wavelength of about 700 nm from the plurality of known
normal sites.
75. The system of claim 74, wherein the tissue is brain.
76. The system of claim 74, wherein the spectrometer acquires the
diffuse reflectance electromagnetic signals returned from the site
at a rate of about 33 Hz.
77. The system of claim 74, wherein the the site is smaller than
about 5 mm in diameter and thinner than about 2 mm in
thickness.
78. A system for identifying a site in a tissue of a patient as
neoplastic or normal, the system comprising: a) a source of
electromagnetic signals; b) a probe operably connected to the
source of electromagnetic signals, for delivering the
electromagnetic signals to a tissue site; c) a spectrometer which
acquires diffuse reflectance electromagnetic signals returned from
the tissue site and produces a diffuse reflectance spectra from the
returned diffuse reflectance electromagnetic signals; and d) a
system controller operably connected to the spectrometer, wherein
the system controller: i) analyzes the diffuse reflectance spectra
to calculate one or more of: a) absolute hemoglobin concentration
([Hb]), b) absolute hemoglobin oxygenation (SatO.sub.2), c)
absolute diffuse reflectance intensity of signals having wavelength
of about 700 nm (Rd(700)), and d) relative ratio of absorption
coefficient .mu..sub.a and reduced scattering cooefficient
.mu..sub.s between 450 nm and 650 nm
(.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.)); ii) analyzes [Hb],
SatO.sub.2, Rd(700), and/or .mu..sub.a(.lamda.)/.mu..sub.s(.lamda.)
with respect to those from normal reference tissue sites to
identify the tissue site as neoplastic or normal; and iii)
determines the site in the tissue of the patient as neoplastic if:
a) any one of [Hb], SatO.sub.2, Rd(700), and
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) was
>1.+-.u.times.(stand_deviation), or b) any one of [Hb],
SatO.sub.2, Rd(700), and .mu..sub.a(.lamda.)/.mu..sub.s(.lamda.)
was <1.+-.u.times.(stand_deviation), wherein u is an adjustable
cutoff value for [Hb], SatO.sub.2, and Rd(700) and
(stand_deviation) is the standard deviation for the selected
parameter from the plurality of known normal sites.
79. The system of claim 78, wherein the tissue is brain.
80. The system of claim 78, wherein the spectrometer acquires the
diffuse reflectance electromagnetic signals returned from the site
at a rate of about 33 Hz.
81. The system of claim 78, wherein the the site is smaller than
about 5 mm in diameter and thinner than about 2 mm in
thickness.
82. A system for identifying a site in a tissue of a patient as
neoplastic or normal, the system comprising: a) a source of
electromagnetic signals; b) a probe operably connected to the
source of electromagnetic signals, for delivering the
electromagnetic signals to a tissue site; c) a spectrometer which
acquires diffuse reflectance electromagnetic signals returned from
the tissue site and produces a diffuse reflectance spectra from the
returned diffuse reflectance electromagnetic signals; and d) a
system controller operably connected to the spectrometer, wherein
the system controller: i) analyzes the diffuse reflectance spectra
to calculate two or more of: a) absolute hemoglobin concentration
([Hb]), b) absolute hemoglobin oxygenation (SatO.sub.2), c)
absolute diffuse reflectance intensity of signals having wavelength
of about 700 nm (Rd(700)), and d) relative ratio of absorption
coefficient .mu..sub.a and reduced scattering cooefficient
.mu..sub.s between 450 nm and 650 nm
(.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.)); ii) analyzes [Hb],
SatO.sub.2, Rd(700), and/or .mu..sub.a(X)/.mu..sub.s(k) with
respect to those from normal reference tissue sites to identify the
tissue site as neoplastic or normal; and iii) determines the site
in the tissue of the patient as neoplastic if: a) any two of [Hb],
SatO.sub.2, Rd(700), and .mu..sub.a(.lamda.)/.mu..sub.s(.lamda.)
was >1.+-.v.times.(stand_deviation), or b) any two of [Hb],
SatO.sub.2, Rd(700), and .mu..sub.a(.lamda.)/.mu..sub.s(.lamda.)
was <1.+-.v.times.(stand_deviation), wherein v is an adjustable
cutoff value for any two of [Hb], SatO.sub.2, Rd(700), and
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) and (stand_deviation) is
the standard deviation for the selected parameters from the
plurality of known normal sites.
83. The system of claim 82, wherein the tissue is brain.
84. The system of claim 82, wherein the spectrometer acquires the
diffuse reflectance electromagnetic signals returned from the site
at a rate of about 33 Hz.
85. The system of claim 82, wherein the the site is smaller than
about 5 mm in diameter and thinner than about 2 mm in
thickness.
86. A system for identifying a site in a tissue of a patient as
neoplastic or normal, the system comprising: a) a source of
electromagnetic signals; b) a probe operably connected to the
source of electromagnetic signals, for delivering the
electromagnetic signals to a tissue site; c) a spectrometer which
acquires diffuse reflectance electromagnetic signals returned from
the tissue site and produces a diffuse reflectance spectra from the
returned diffuse reflectance electromagnetic signals; and d) a
system controller operably connected to the spectrometer, wherein
the system controller: i) analyzes the diffuse reflectance spectra
to calculate: a) absolute hemoglobin concentration ([Hb]), b)
absolute hemoglobin oxygenation (SatO.sub.2), c) absolute diffuse
reflectance intensity of signals having wavelength of about 700 nm
(Rd(700)), and d) relative ratio of absorption coefficient
.mu..sub.a and reduced scattering cooefficient .mu..sub.s between
450 nm and 650 nm (.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.)); ii)
analyzes [Hb], SatO.sub.2, Rd(700), and/or
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) with respect to those from
normal reference tissue sites to identify the tissue site as
neoplastic or normal; and iii) determines the site in the tissue of
the patient as neoplastic if: a) all of [Hb], SatO.sub.2, Rd(700),
and .mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) were
>1.+-.w.times.(stand_deviation), or b) all of [Hb], SatO.sub.2,
Rd(700), and .mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) were
<1.+-.w.times.(stand_deviation), wherein w is an adjustable
cutoff value for all of [Hb], SatO.sub.2, and Rd(700) and
(stand_deviation) is the standard deviation for the selected
parameters from the plurality of known normal sites.
87. The system of claim 86, wherein the tissue is brain.
88. The system of claim 86, wherein the spectrometer acquires the
diffuse reflectance electromagnetic signals returned from the site
at a rate of about 33 Hz.
89. The system of claim 86, wherein the the site is smaller than
about 5 mm in diameter and thinner than about 2 mm in thickness.
Description
BACKGROUND OF INVENTION
[0002] Brain tumors continue to be the leading cause of death among
all forms of pediatric cancer. In a report issued by the United
States Central Brain Tumor Registry, based upon data collected from
2004 to 2007 across 48 states, more than 4000 new pediatric cases
of brain tumor were predicted for 2011:.sup.[1] The three most
common pediatric brain tumors are astrocytoma (50%),
medulloblastoma (20%), and ependymoma (10%). Dysembryoplastic
neuroepithelial tumors, although benign, are congenital tumors that
are strongly associated with seizures and occur frequently in
children. For each of these pathologic entities, gross total
surgical resection, when feasible, is the optimal initial treatment
option..sup.[2,3] Strong correlations between the extent of tumor
resection and clinical outcomes, such as 10 year progression-free
survival, have been reported by different groups:.sup.[4-8]
[0003] Currently, frameless stereotaxy is used intraoperatively to
guide tumor resection. Frameless stereotactic navigation systems
enable neurosurgeons to relate the position of a tracking probe to
structures present in pre-operative computed tomography (CT) or
magnetic resonance imaging (MRI) studies. .sup.[9-10] However,
CT/MRI studies may not delineate exact brain tumor margins,
especially when infiltrating, due to their limited sensitivity.
Previous investigations have demonstrated that tumor cells are
often found in the brain outside of tumor margins defined by CT and
MRI studies:.sup.[11] Furthermore, the accuracy of surgical
navigation systems is often diminished during surgical procedures,
due to shifts in brain structures..sup.[11-12] While this issue may
be overcome using intraoperative MRI (iMRI), this technology is
very expensive to install and operate and its image acquisition
time is long.
[0004] A popular alternative for intraoperative brain tumor
detection is ultrasonography. While affordable and convenient, its
reliability in detecting tumors is less than that of iMRI. Gerganov
et al. compared ultrasonography with iMRI and pointed out that
ultrasound could be used to detect more-confined, deeply-located
remnants of low- or high-grade tumors with high accuracy, though
its accuracy is limited when detecting superficial remnants. In
addition, ultrasound is less likely to differentiate tumor from
peritumoral edema, which is also hyperechoic.
[0005] Due to the limitations of these existing techniques,
neurosurgeons typically rely on visual inspection and/or on-site
pathology to guide tumor resection. Visual inspection,
unfortunately, can be subjective and sometimes inconclusive;
on-site pathology can be time-consuming and often inaccurate based
upon frozen sections alone. Thus, significant residual tumor cells
frequently are left behind after surgical resection.
[0006] Optical spectroscopy uses light absorption and scattering to
quantify tissue biochemical composition and morphological
characteristics. It has the advantage of providing non-destructive,
automated tissue characterization in real-time, without removing
tissue. To date, optical spectroscopy has been widely used to study
pathological and physiological features at the tissue and cell
levels in vivo and in vitro.
[0007] Among the spectroscopic techniques used in biomedicine,
fluorescence spectroscopy has been of particular interest, because
quantities of two intrinsic fluorophores, NAD(P)H and flavin
proteins, are strongly associated with metabolic states of tissue.
In addition, fluorescence spectroscopy can be used to study
abnormal blood vessel proliferation and fibrosis in tissue, because
Type-I collagen is another prominent biological fluorophores.
[0008] Diffuse reflectance spectroscopy is another optical
spectroscopy technique commonly used in optical tissue
characterization. It allows detection of tissue structure and
biochemical composition by monitoring the optical properties (e.g,
absorption and reduced scattering coefficients) of tissue.
.sup.[13]
[0009] One common utility of diffuse reflectance spectroscopy is
the detection of tissue hemodynamics, based on the fact that oxy-
and deoxy-hemoglobin possess unique absorption features..sup.[14]
Understanding and assessing Hb oxygenation and concentration
provides valuable insights into the condition of tissue. It is not
surprising, then, that regional Hb oxygenation and Hb concentration
are highly-sought information in tissue injury and disease
diagnosis. .sup.[15-20]
[0010] Optical diagnostic technologies provide a potential
complementary solution for intraoperative brain tumor demarcation,
as demonstrated by several research groups. Optical technologies
have also been used in conjunction with exogenesis tumor labels,
such as ALA, for intraoperative guidance of brain tumor resection.
.sup.[21-24] Clinical trials conducted with the ALA-based
fluorescence brain tumor demarcation system demonstrate significant
improvement in completeness of brain tumor removal in adult
patients with high-grade brain tumors. However, the applicability
of this guidance system to pediatric populations remains uncertain,
as the safety of ALA in children has not yet been evaluated.
Furthermore, the ALA-based guidance system is most efficient only
for high-grade brain tumors, because ALA can efficiently cross the
compromised blood brain barrier that exists in these
tumors..sup.[25] Whether the blood brain barrier also breaks down
in pediatric brain tumors remains unknown.
[0011] To make optical diagnostic technology applicable within an
in vivo or intraoperative environment, a contact probe often is
employed to achieve remote sensing from in vivo tissue in the
sterile operating field. The probe typically is held by the hand of
an operator. One common problem with such a practice is motion
artifacts in the acquired data. Unintentional hand movements or
tremors alter the pressure of the probe against the target tissue.
Ultimately, these movements are incorporated into the data as noise
or artifacts.
[0012] A couple of earlier reports have suggested that probe
pressure does not significantly affect the fluorescence intensity
of the cervix..sup.[26] A similar finding was noted in a study in
which Raman spectroscopy was used to detect pre-cancerous lesions
within the gastrointestinal tract..sup.[27-28] However, several
other studies have shown that excessive probe contact pressure,
resulting from hand movements, can lead to strong alterations in
the hemodynamic and metabolic characteristics of local
tissue..sup.[29-32] This issue must be addressed in order to
maximize the efficiency of an optical diagnostic technology for
intraoperative surgical guidance.
BRIEF SUMMARY
[0013] The current invention provides a real-time system that is
capable of intraoperatively detecting the margins of tumors, for
example, brain tumors, with high sensitivity to aid surgeons in
their objective to safely and completely resect tumors without
removing normal tissue. The system of the current invention can be
used to identify and remove tumors from any tissue, for example,
brain.
[0014] For example, the current invention provides an optical
guidance system that can differentiate brain tumors from normal
brain at the resection front based upon distinct intrinsic
morphological, biochemical, and physiological attributes of
neoplastic cells compared to normal cells.
[0015] The current invention provides a system for identifying a
site in the tissue of a patient as neoplastic or normal. In one
embodiment, the system of the current invention comprises:
[0016] a) a source of electromagnetic signals;
[0017] b) an optical probe coupled with the source of
electromagnetic signals, wherein the optical probe delivers the
electromagnetic signals to a working end of the probe;
[0018] c) a spectrometer which acquires diffuse reflectance
electromagnetic signals returned from the tissue site probed by the
working end of the probe and wherein the spectrometer processes the
diffuse reflectance signals to produce diffuse reflectance spectra
of the returned diffuse signals;
[0019] d) a system controller having a processor coupled with the
spectrometer, wherein the system controller is programmed to:
[0020] i) analyze the diffuse reflectance spectra to calculate one
or more of: [0021] a) absolute hemoglobin concentration ([Hb]),
[0022] b) absolute hemoglobin oxygenation (SatO.sub.2), [0023] c)
diffuse reflectance intensity of signals having wavelength of about
700 nm (such as, for example, Rd(700)), [0024] and [0025] ii)
analyze [Hb], SatO.sub.2, and/or Rd(700) relative to those from
normal reference tissue sites to identify the tissue site as
neoplastic or normal. In one embodiment, the sites are from the
same patient.
[0026] The invention can be used to positively impact the
management of brain tumors, for example, pediatric brain tumors,
because it provides new means by which neurosurgeons can
objectively optimize the outcomes of brain tumor surgeries, thereby
improving the prognoses of patients, and reducing the emotional and
financial burdens endured by patients and their families. Moreover,
the current invention can be used to enhance other studies for
characterizing the spatial and temporal physiological
characteristics of epileptic cortex in vivo and, hence, aid in
epilepsy surgery, for example, pediatric epileptic surgery.
[0027] The surgical guidance system of the current invention can
make a significant impact to brain tumor surgery because it
provides speedy and objective assessment to the brain tissue in the
operating room, which allows surgeons to make an informed decision
about the extent of tumor resection.
[0028] This system can be designed for specific use in any type of
brain tumor patients, for example, adult patients and pediatric
patients. It may be used in other tumor resection procedures with a
modified tissue differentiation algorithm.
[0029] The current invention can be used to identify a site in a
tissue, for example, brain, of a patient as neoplastic or normal.
The current invention also provides methods of using the system to
intraoperatively identify a site as neoplastic or normal and
surgically remove the site from the tissue of the patient, for
example, brain, if the site is identified as neoplastic.
BRIEF DESCRIPTION OF DRAWINGS
[0030] FIGS. 1A and 1B. FIG. 1A shows the distribution of tumor and
normal tissue sites, in accordance with their corresponding
nSatO.sub.2, n[Hb], and nRd(700). The open circles represent normal
cortex tissue and the solid circles represent brain tumor tissue.
FIG. 1B shows the Receiver Operating Characteristics (ROCs) from
the discrimination methods based on single parameters (nSatO.sub.2,
n[Hb], and nRd(700)) as well as all three parameters combined.
[0031] FIG. 2. Diffuse reflectance signal as a function of r, f(r),
at four different wavelengths. Here individual f(r) was generated
using a Monte Carlo (MC) simulation model for photon migration.
Optical properties used in the simulation are those of human cortex
at the four specified wavelengths; the illumination used is a
pencil beam.
[0032] FIGS. 3A and 3B. FIG. 3A shows theprototype of a hybrid
imaging spectroscopy system with the new diffuse reflectance
acquisition scheme embedded. The Hastings lenses are used to relay
the image from the image plane of the camera lens (mimicking a
surgical microscope) to the CCD sensor of the camera and the
detection plane of the diffuse reflectance measurement port. The
optical fiber attached to the diffuse reflectance measurement port
enables us to selectively measure the diffuse reflectance signal
from an arbitrary point within the field of view of the imaging
system. According to Eq. (1) and Eq. (2) (see, Example 2), this
measured signal can be converted to diffuse reflectance if it is
normalized to the illumination power at the point of detection.
Multiple fibers may be attached to the diffuse reflectance
measurement port to enable simultaneous investigation of multiple
points on the subject. FIG. 3B shows the system attached to the
beam splitter of a surgical microscope, which will be used in brain
tumor surgery.
[0033] FIG. 4. Total diffuse reflectance R.sub.TA as a function of
reduced albedo. The solid line is calculated using a MC simulation
model for photon migration, the dash-line using the analytical
model by Farrell et al..sup.[36]
[0034] FIGS. 5A, 5B, and 5C. Cortical vasculature analysis using
static photos. FIG. 5A shows the original picture of the exposed
cortical surface. FIG. 5B shows the processed image with all blood
vessels, including arteries, arterioles, veins, and venules
highlighted. Note specular reflection from the cortical surface
results in certain noticeable artifacts in the processed image.
Hence, blood vessel density analysis should not be performed in
these regions. FIG. 5C shows the Representative vessel density
characteristics determined by the pixel density histogram. The
upper panel is from a 50.times.50 pixel area within the zone of
resection, and the lower panel is from an area outside the zone of
resection. The vessel identification algorithm used here highlights
the edge of the vessels, and the intensities of the highlighted
pixels vary in accordance with the edge definition of the vessels
(i.e., the sharper the edge, the higher the intensity). Currently,
we are exploring other methodologies to quantify blood vessel
density within areas of interest, as well as to separate
veins/venules from arteries/arterioles more objectively.
DETAILED DISCLOSURE
[0035] The current invention provides a system for identifying a
site in the tissue of a patient as neoplastic or normal with high
sensitivity and specificity. In general, embodiments of the hybrid
spectroscopy and imaging system of the subject invention can
include:
[0036] a) a source of electromagnetic signals;
[0037] b) an optical probe coupled with the source of
electromagnetic signals, wherein the optical probe delivers the
electromagnetic signals to a working end of the probe;
[0038] c) a spectrometer which acquires diffuse reflectance
electromagnetic signals returned from the tissue site probed by the
working end of the probe and wherein the spectrometer processes the
diffuse reflectance signals to produce diffuse reflectance spectra
of the returned diffuse signals;
[0039] d) a system controller having a processor coupled with the
spectrometer, wherein the system controller is programmed to:
[0040] i) analyze the diffuse reflectance spectra to calculate one
or more of: [0041] a) absolute hemoglobin concentration (e.g.,
[Hb]), [0042] b) absolute hemoglobin oxygenation (e.g.,
SatO.sub.2), [0043] c) absolute diffuse reflectance intensity of
signals having wavelength of about 700 nm (e.g., Rd(700)), [0044]
and [0045] ii) analyze [Hb], SatO.sub.2, and/or Rd(700) relative to
those from normal reference tissue to identify the site as
neoplastic or normal.
[0046] In a more specific embodiment, the hybrid spectroscopy and
imaging system of the subject invention utilizes:
[0047] (1) an imaging device such as a surgical microscope;
[0048] (2) a source of electromagnetic signals that provides a
broadband, large area of illumination;
[0049] (3) an image relay lens mounted at a secondary camera port
of the imaging device in (1);
[0050] (4) an optical fiber adaptor placed at the imaging plane of
the relay lens;
[0051] (5) one or more optical fibers with a small core diameter
that are connected to the optical fiber adaptor to collect total
diffuse reflectance electromagnetic signals;
[0052] (6) a spectrometer which acquires total diffuse reflectance
electromagnetic signals collected by the optical fibers of (5) and
produces total diffuse reflectance spectra of the returned diffuse
signals;
[0053] (7) a system controller having a process coupled with the
spectrometer, wherein the system controller is programmed to:
[0054] a. Analyze the total diffuse reflectance spectra to
calculate one or more of: [0055] i. absolute hemoglobin oxygenation
(e.g.,, SatO2), [0056] ii. absolute total diffuse reflectance
intensity of signals having wavelength between 650 nm and 800 nm
(e.g., Rd (700)), [0057] iii. relative ratio of absorption
coefficient n.sub.a and reduced scattering cooefficient .mu..sub.s
between 450 nm and 650 nm (e.g.,
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.)),
[0058] and [0059] b. Analyze SatO2, Rd(700), and/or
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) relative to those from
normal reference tissue sites of the same patient to identify the
site as neoplastic or normal.
[0060] The system of the current invention can be utilized to
identify a tissue site as neoplastic or normal in a variety of
tissues, for example, brain, eyes, pineal gland, pituitary gland,
thyroid gland, parathyroid glands, thorax, heart, lungs, esophagus,
thymus gland, pleura, adrenal glands, appendix, gall bladder,
urinary bladder, large intestine, small intestine, kidneys, liver,
pancrease, spleen, stoma, prostate gland, testes, ovaries, or
uterus.
[0061] In one embodiment of the invention, the system of the
current invention is utilized to identify a site in the brain as
neoplastic or normal.
[0062] The system of the current invention can also be configured
to identify a site in the brain as epileptic or non-epileptic based
on the SatO.sub.2, Rd(700), and
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.).
[0063] In certain embodiments, of the hybrid spectroscopy imaging
system of the subject invention, the system controller is
programmed to analyze the total diffuse reflectance spectra to
calculate SatO2 and Rd(700), and analyze SatO2and Rd(700) to
identify the tissue site as neoplastic or normal. In certain other
embodiments, the system controller is programmed to analyze the
diffuse reflectance spectra to calculate SatO.sub.2 and
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.), and analyze SatO.sub.2 and
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) to identify the site as
neoplastic or normal. In further embodiments, the system controller
is programmed to analyze the diffuse reflectance spectra to
calculate Rd(700) and .mu..sub.a(.lamda.)/.mu..sub.s(.lamda.), and
analyze Rd(700) and .mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) to
identify the site as neoplastic or normal.
[0064] In one embodiment, the system controller is programmed to
analyze the total diffuse reflectance spectra to calculate
SatO.sub.2, Rd(700), and .mu..sub.a(.lamda.)/.mu..sub.s(.lamda.),
and analyze SatO.sub.2, Rd(700), and
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) to identify the site as
neoplastic or normal.
[0065] For the purposes of this invention, if a site in the tissue
of a patient is identified as neoplastic, it indicates that the
site contains cancerous cells; whereas, if a site in the tissue of
a patient is identified as normal it indicates that the site is
free from cancerous cells.
[0066] An electromagnetic signal refers to a wave of energy having
a wavelength within the electromagnetic spectrum. An example of an
electromagnetic wave is a light wave having a wavelength of 600
nm.
[0067] Electromagnetic signals refer to a group of waves having one
or more frequencies within the electromagnetic spectrum. An example
of electromagnetic signals is a beam of white light, which
comprises a plurality of electromagnetic waves of different
wavelengths.
[0068] The source of electromagnetic signals produces the
electromagnetic signals comprising a plurality of electromagnetic
waves of different wavelengths. For example, the electromagnetic
signals comprising electromagnetic waves having wavelengths ranging
from about 650 nm to about 800 nm refers to the electromagnetic
signals comprising a plurality of waves, each wave having a
wavelength within the range of about 650 nm to about 800 nm and the
electromagnetic signals contain at least some waves of each
wavelength falling within the range of about 650 to about 800
nm.
[0069] In certain embodiments of the invention, the electromagnetic
signals comprise electromagnetic waves having wavelengths ranging
from about 500 nm to about 900 nm, about 600 nm to about 850 nm, or
about 650 nm to about 800 nm. In one embodiment of the invention,
the electromagnetic signals comprise electromagnetic waves having
wavelengths ranging from about 650 nm to about 800 nm.
[0070] The electromagnetic signals utilized by embodiments of the
subject invention can have several sources, including, but not
limited to: [0071] the illumination light of an imaging device such
as a surgical microscope; [0072] an image relay lens mounted at a
secondary camera port of an imaging device such as a surgical
microscope; [0073] an optical fiber adaptor placed at the imaging
plane of the relay lens, where multiple fibers can be connected to
the adaptor; [0074] one or more optical fibers with a small core
diameter that are connected to the optical fiber adaptor to collect
total diffuse reflectance electromagnetic signals.
[0075] A spectrometer detects total diffuse reflectance
electromagnetic signals collected by the optical fibers and
produces total diffuse reflectance spectra of the returned diffuse
signals.
[0076] An optical probe coupled with the source of electromagnetic
signals transfers the electromagnetic waves to the working end of
the probe. In one embodiment of the invention, the optical probe
used to transfer the electromagnetic waves from the electromagnetic
source to the working end of the optical probe is a fiber optic
probe.
[0077] The working end of the optical probe is the end of the
optical probe from which the electromagnetic signals emerge from
the optical probe. This end of the optical probe delivers the
electromagnetic signals.
[0078] For the purposes of this invention "probing a site with the
optic probe" is intended to mean that the electromagnetic signals
are delivered to the site by holding the working end of the optic
probe in direct physical contact with the site.
[0079] A spectrometer acquires diffuse reflectance electromagnetic
signals returned from the site probed by the working end of the
optical probe. The spectrometer processes the diffuse reflectance
signals to produce the diffuse reflectance spectra of the returned
diffuse signals. In one embodiment of the invention, a set of 400
diffuse reflectance spectra is produced by the spectrometer.
[0080] In certain embodiments of the invention, the spectrometer
acquires the diffuse reflectance electromagnetic signals returned
from the site at a rate of about 5 Hz or higher, about 5 Hz to
about 60 Hz, about 30 Hz to about 40 Hz, or about 60 Hz or higher.
In one embodiment of the invention, the spectrometer acquires the
diffuse reflectance electromagnetic signals returned from the site
at about 33 Hz.
[0081] For the purposes of this invention, a system controller
programmed to perform certain tasks indicates that the system
controller is provided with a set of coded instructions that
enables the system controller to perform a desired sequence of
operations. In one embodiment of the invention, the system
controller is a computer. The set of coded instructions that
enables the system controller to perform a desired sequence of
operations is called a software program or software.
[0082] The system controller having a processor coupled with the
spectrometer is programmed to analyze the diffuse reflectance
spectra to calculate (1) absolute hemoglobin oxygenation (e.g.,
SatO2), (2) absolute total diffuse reflectance intensity of signals
having a wavelength between 650 nm and 800 nm (e.g., Rd(700)), (3)
relative ratio of absorption coefficient p.sub.a and reduced
scattering coefficient .mu..sub.s between 450 nm and 650 nm (e.g.,
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.)). The methods of
calculation are known to those with ordinary skill in the art.
[0083] For example, methods of calculation can be found in
published scientific articles..sup.[34-35] Additional methods for
calculating Rd(700) and SatO.sub.2 using diffuse reflectance
spectra are well known to a person of ordinary skill in the
art.
[0084] In certain embodiments of the invention, the system
controller is programmed to identify the site as neoplastic if:
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.)>1+x.times..mu..sub.a(.lamda.)-
/.mu..sub.s(.lamda.).sub.N.sub._.sub.std or
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.)<1-x.times..mu..sub.a(.lamda.)/-
.mu..sub.s(.lamda.).sub.N.sub._.sub.std,
[0085] wherein x is an adjustable cutoff for
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) and
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.).sub.N.sub._.sub.std is
standard deviation of .mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) from
a plurality of known normal sites.
[0086] In certain other embodiments, the system controller is
programmed to identify the site as neoplastic if:
SatO.sub.2>1+y.times.SatO.sub.2 N.sub._.sub.std or
SatO.sub.2<1-y.times.SatO.sub.2 N.sub._.sub.std,
[0087] wherein y is an adjustable cutoff for SatO.sub.2 and
SatO.sub.2 N.sub._.sub.std is standard deviation of SatO.sub.2 from
the plurality of known normal sites.
[0088] In further embodiments, the system controller is programmed
to identify the site as neoplastic if:
Rd(700)>1+z.times.Rd(700).sub.N.sub._.sub.std or
Rd(700)<1-z.times.Rd(700).sub.N.sub._.sub.std,
[0089] wherein z is an adjustable cutoff and Rd(700) and
Rd(700).sub.N.sub._.sub.std is standard deviation for Rd(700) from
the plurality of known normal sites.
[0090] In even further embodiments, the system controller is
programmed to identify the site as neoplastic if:
[0091] a) any one of SatO.sub.2, and Rd(700), and
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) is
>1+u.times.(stand_deviation), or
[0092] b) any one of SatO.sub.2, and Rd(700), and
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) is
<1-u.times.(stand_deviation),
[0093] wherein u is an adjustable cutoff value for SatO.sub.2, or
Rd(700), and .mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) and
(stand_deviation) is the standard deviation for the selected
parameter from the plurality of known normal sites.
[0094] In additional embodiments, the system controller is
programmed to indentify the site as neoplastic if:
[0095] a) any two of SatO.sub.2, Rd(700), and
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) is
>1+v.times.(stand_deviation), or
[0096] b) any two of SatO.sub.2, Rd(700), and
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) is
<1-v.times.(stand_deviation), wherein u is an adjustable cutoff
value for SatO.sub.2, Rd(700), and
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) and (stand_deviation) is
the standard deviation for the selected parameters from the
plurality of known normal sites. In certain other embodiments, the
system controller is programmed to identify the site as neoplastic
if:
[0097] a) all of SatO.sub.2, Rd(700), and
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) are
>1+w.times.(stand_deviation), or
[0098] b) all of SatO.sub.2, Rd(700), and
p.sub.a(.lamda.)/.mu..sub.s(.lamda.) are
<1-w.times.(stand_deviation),
[0099] wherein u is an adjustable cutoff value for SatO.sub.2,
Rd(700), and .mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) and
(stand_deviation) is the standard deviation for the selected
parameters from the plurality of known normal sites.
[0100] For the purposes of this invention, the term "adjustable
cutoff value" for a particular parameter is a pre-determined value
that is used to identify whether a site is neoplastic or normal.
The adjustable cutoff value can be between about 0.5 to about 4.0,
about 0.75 to about 3.0, about 1.0 to about 2.0, or about 1.25 to
about 1.75. In one embodiment of the invention, the adjustable cut
off value for a particular parameter is about 1. In another
embodiment of the invention, the adjustable cut off value for a
particular parameter is about 2.
[0101] The current invention also provides a method of using the
system to identify a site in the tissue, for example, brain tissue,
of a patient as neoplastic or normal. The method of using the
system of the current invention comprises the steps of:
[0102] a) delivering the electromagnetic signals to the site and
surrounding areas (i.e., large illumination area),
[0103] b) acquiring total diffuse reflectance electromagnetic
signals at the camera port of an imaging system, such as a surgical
microscope, wherein the spectrometer detects total diffuse
reflectance electromagnetic signals collected by the optical fibers
and produces total diffuse reflectance spectra of the returned
diffuse signals;
[0104] c) transmitting the total diffuse reflectance spectra from
the spectrophotometer to the system controller having the processor
coupled with the spectrometer, wherein the system controller is
programmed to:
[0105] a) analyze the total diffuse reflectance spectra to
calculate one or more of: [0106] i) absolute hemoglobin oxygenation
(e.g., SatO2), [0107] ii) absolute total diffuse reflectance
intensity of signals having wavelengths between 650 nm and 800 nm
(e.g., Rd(700)), [0108] iii) relative ratio of absorption
coefficient .mu..sub.a and reduced scattering coefficient
.mu..sub.s between 450 nm and 650 nm (e.g.,
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.)),
[0109] and
[0110] d) analyze SatO2, Rd(700), and/or
.mu..sub.a(.lamda.)/.mu..sub.s(.lamda.) relative to those from
normal reference sites of the same patient to identify the site as
neoplastic or normal.
[0111] In certain embodiments of the invention, the method is
performed intraoperatively, e.g., the method is performed during
the course of a surgery. In certain other embodiments of the
invention, the method of identifying a site as neoplastic or normal
further comprises removing the site from the brain of the patient
if the site is identified as neoplastic.
[0112] In certain embodiments of the invention, the site is smaller
than about 10 mm in diameter and about 5 mm in thickness. In
certain other embodiments, the site is smaller than about 5 mm in
diameter and about 2 mm in thickness. In further embodiments of the
invention, the site is about 4 mm, about 3 mm, about 2 mm, or about
1 mm in diameter and about 5 mm, 4 mm, 3 mm, 2 mm, and 1 mm in
thickness.
[0113] The method of the current invention can be performed on an
adult patient or a pediatric patient, i.e. a child. In certain
embodiments the child is less than 10, 9, 8, 7, 6, 5, 4, 3, 2, or 1
years of age.
[0114] The current invention also provides a procedure of
determining the margins of a tumor in an organ. The procedure
comprises identifying a plurality of sites in the organ as
neoplastic or normal according to the methods and systems of the
current invention, and providing the information about the
plurality of sites to an apparatus which is programmed to represent
the margins of the tumor in the organ in a graphical manner.
[0115] The organ can be selected from brain, eyes, pineal gland,
pituitary gland, thyroid gland, parathyroid glands, thorax, heart,
lungs, esophagus, thymus gland, pleura, adrenal glands, appendix,
gall bladder, urinary bladder, large intestine, small intestine,
kidneys, liver, pancrease, spleen, stoma, prostate gland, testes,
ovaries, or uterus. In one embodiment of the invention, the methods
of the current invention are used to determine the margins of a
brain tumor, for example, in an adult or a pediatric patient.
EXAMPLE 1
Detecting Brain Tumors by Using Diffuse Reflectance Signals,
Hemoglobin Oxygenation, and Hemoglobin Concentration
[0116] Both in vitro and in vivo study results show that diffuse
reflectance signals in the long wavelength region (e.g. between 650
nm and 800 nm) are efficient at differentiating a brain tumor from
normal brain tissue..sup.[33] This spectral feature reflects the
structural characteristics of a tissue. The current invention
demonstrates that adding regional hemodynamic characteristics (e.g.
hemoglobin oxygenation (nSatO.sub.2) and hemoglobin concentration
(n[Hb]) further enhances the accuracy of differentiation.
[0117] Twelve pediatric patients, between one and 18-years old,
were recruited to participate in a clinical study. Diffuse
reflectance spectra were obtained from the in vivo brains of
participants during craniotomy procedures for tumor resection,
using a fiber optic based spectroscopy system..sup.[44] During
spectral data acquisition, the optical probe was held by the
neurosurgeon and was in direct contact with the brain tissue.
Optical measurements were taken from areas away from the resection
zone (i.e., normal sites) and from areas within the resection zone
(tumor sites). A set of 400 diffuse reflectance spectra Rd(.lamda.)
was acquired from each investigation site at a rate of 33 Hz.
Following each spectral acquisition sequence, a baseline
measurement Rd(.lamda.).sub.base was conducted by turning off the
electromagnetic signals source. At least three normal sites and
three brain tumor sites were investigated during each patient
study.
[0118] A specimen was collected from each investigated site within
the resection zone for histopathological evaluation to identify the
type of brain tumor. Biopsy samples were immediately fixed in a 10%
formalin solution after resection. The specimens were then prepared
for sectioning and hematoxylin and eosin staining. Processed slides
were reviewed by a neuropathologist who was blinded to study
results and other clinical information.
[0119] The baseline measurement Rd(.lamda.).sub.base from each
investigated site was first subtracted from the corresponding
diffuse reflectance spectral set Rd(.lamda.) to remove unwanted
ambient light influences. Then, instrumentally-induced spectral
alterations were eliminated by dividing the spectra by a
calibration spectrum Rd.sub.cal(.lamda.). Note that the calibration
spectrum was measured from a diffuse reflectance standard
(FGS-20-02c, Avian Technologies, NH) using the same spectroscopic
system. Mathematically speaking, the entire spectral calibration
process can be described using the following equation:
[Rd(.lamda.)-Rd(.lamda.).sub.base]/Rd.sub.cal(.lamda.).
[0120] Three parameters extracted from each calibrated diffuse
reflectance spectra were used to characterize the tissue site of
investigation. These indicative parameters were n[Hb], nSatO.sub.2,
and nRd(700) (e.g., the diffuse reflectance intensity at 700 nm).
The methods for estimating n[Hb] and nSatO.sub.2 using diffuse
reflectance spectra can be found in published scientific
articles..sup.[60,61] Since 400 continuous measurements were
acquired from each investigated site, the mean value of each
indicative parameter across the 400 measurements was used as a
representative value. In order to reduce inevitable biological
variations among all studied subjects, the indicative parameters
from each patient were normalized to the average of the indicative
parameters from the normal tissue sites of the same patient, which
yielded n[Hb], nSatO.sub.2, and nRd(700). In addition, the
distributions of n[Hb], nSatO.sub.2, and nRd(700) from all normal
sites of the studied patients were assessed, and their standard
deviations, denoted as n[Hb].sub.N std, nSatO2.sub.N.sub._.sub.std
and nRd(700).sub.N.sub._.sub.std, respectively, were
calculated.
[0121] The normalized indicative parameters described in the
previous section were used to establish a classification system to
differentiate brain tumor from normal brain (cortex). The
classification criterion for each parameter is described as
follows: (1) If n[Hb]>1+x.times.n[Hb].sub.N.sub._.sub.std or
n[Hb]<1-x.times.n[Hb].sub.N.sub._.sub.std, where x is an
adjustable cutoff, the site was classified as neoplastic;
otherwise, the site was classified as normal. Similarly, the tumor
classification criteria for the two other parameters were
designated (2) nSatO.sub.2>1+x.times.nSatO.sub.2 N.sub._.sub.std
or nSatO.sub.2<1-x.times.nSatO.sub.2 N.sub._.sub.std and (3)
nRd(700)>1+x.times.nRd(700).sub.N.sub._.sub.std or
nRd(700)<1-x.times.nRd(700).sub.N.sub._.sub.std,
respectively.
[0122] The accuracies of these classification algorithms were
quantified in terms of sensitivity and specificity, as well as
through a receiver operating characteristic (ROC) curve. Moreover,
a multi-variant classification criterion was established by
combining the three classification criteria mentioned above. Using
these combined criteria, a tissue site was classified as neoplastic
if any one of the indicative parameters was outside the
1.+-.x.times.(stand_deviation) boundaries. Here, the cutoff x
values for all three parameters were identical. As with the single
parameters, the performance of this multi-variant classification
algorithm was characterized in terms of sensitivity, specificity,
and ROC.
[0123] Comparing mean values, the average of nRd(700) from tumor
measurements was lower than from normal brain. However, this trend
was not observed in the n[Hb] and nSatO2 comparisons. Evaluation of
the histograms of n[Hb] and nSatO.sub.2 revealed a bi-modal
distribution characteristic in the tumor measurements. When all
three normalized indicative parameters from normal and tumor sites
were plotted in a three-dimensional Cartesian coordinate system
(FIG. 1A), the indicative parameters from normal cortex clearly
formed a cluster centered with all parameters equal to one;
conversely, parameters from tumors tended to scatter around the
normal group cluster. FIG. 1B shows the ROCs and corresponding
A.sub.ROC for all three individual classification criteria, as well
as the multi-variant classification criterion. All three individual
classification criteria performed satisfactorily, with
A.sub.ROC>0.8. The multi-variant criterion performed best, with
A.sub.ROC>0.9. The upper left point of the ROC of the
multi-variant criterion is the result of the cutoff x=2 for all
three parameters.
[0124] These thresholds also are plotted in FIG. 1A and form a cube
in the figure. If the data points inside the cube are classified as
normal and outside as tumor, this classifier yields a sensitivity
of 95.8% and specificity of 84.6% within the patient pool
studied.
EXAMPLE 2
Development of a Hybrid Imaging and Spectroscopy System for
Intraoperative Brain Tumor Demarcation
[0125] While conducting the in vivo study described above, several
difficulties associated with integrating a probe-based diagnostic
system into a brain tumor resection procedure were noted. First,
the ambient lights must be turned off to acquire accurate diffuse
reflectance spectra; but this is difficult when a surgical
microscope is used. Secondly, it may be difficult to maintain
constant probe contact pressure during each spectral acquisition
procedure, which usually lasts a few seconds. Finally, the probe
may not be able to reach deep-seated tumors when the access channel
is small. These shortcomings motivated the development of a new
modality to conveniently detect artifact-free diffuse reflectance
spectra from in vivo brain.
[0126] The diffuse reflectance characteristics of in vivo tissue
may be acquired using a spectral imaging system. .sup.[62-74] With
this approach, broad illumination and a wavelength selection
mechanism (e.g., a liquid crystal tunable filter) are employed.
Assuming that illumination is uniform and that the target is
homogenous, the diffuse reflectance signal, Rd, from a single point
(e.g., r.sub.d=0) on the target surface, acquired by a spectral
imaging system, can be described by the equation:
Rd ( r d = 0 ) = c .intg. r = 0 .infin. .intg. .theta. = 0 2 .pi. f
( r ) r d .theta. d r = c .intg. r = 0 .infin. f ( r ) 2 .pi. rdr
Eq . ( 1 ) ##EQU00001##
[0127] where c is a constant associated with the collection
geometry of the imaging system; r is the distance between the
source and the detection points; and f(r) is the diffuse
reflectance signal from the same material illuminated by a pencil
beam (e.g., point spread function).
[0128] According to this equation, the contribution from the
illumination point at r to the diffuse reflectance signal at the
point of detection is identified by f(r), which is wider in the
longer wavelength region (e.g., low reduced scattering coefficient
.mu.'.sub.s and absorption coefficient .mu..sub.a), as shown in
FIG. 2. In other words, the volume of investigation associated with
the diffuse reflectance signals measured through the
above-mentioned spectral imaging system increases significantly as
the wavelength increases. This explains why diffuse reflectance
spectra measured using a spectral imaging system differ greatly
from those obtained using a conventional probe-based spectroscopic
system, and how difficult it is to interpret them accurately.
.sup.[64]
[0129] Assuming that an I.sub.p [W] pencil beam is used to
illuminate a homogenous medium and that the resultant diffuse
reflectance is f(r) [W/m.sup.2], the total reflectance R.sub.TM [%]
can be calculated by
R TM = .intg. r = 0 .infin. f ( r ) 2 .pi. rdr / I p . Eq . ( 2 )
##EQU00002##
[0130] By comparing this formula with Eq. (1), the differences
between the two equations are the collection factor in Eq. (1) and
normalization to the illumination power of the pencil beam in Eq.
(2). This similarity, in turn, suggests that diffuse reflectance
signals measured from a surface point on an investigated subject
using a spectral imaging system should be considered as total
reflectance, instead of as a diffuse reflectance signal for an
arbitrary source-detect separation. Based on this concept, a new
detection scheme for diffuse reflectance spectroscopy was
developed, as shown in FIG. 2. The advantage of this design is that
it provides excellent spectral resolution in the measured diffuse
reflectance spectra, as well as the feasibility of continuous
spectral acquisition at a rate of 5 Hz or greater.
[0131] To convert total or diffuse reflectance signals to
physiological parameters (e.g., hemodynamic and structural
characteristics), a model for photon migration is required. Among
all the possible options, we found that both the analytical total
reflectance R.sub.TA model proposed by Farrell et al. .sup.[36] and
the look-up table model based on a Monte Carlo simulation model for
photon migration meet the requirements. It should be noted that
both models indicate that R.sub.TA is a function of reduced albedo
a'=, as shown in FIG. 4. In a prior study conducted by our research
group, we found that it is also possible to retrieve nSatO.sub.2
information through the profile of a calibrated diffuse reflectance
spectrum..sup.[60,61] We have modified the algorithm to make it
applicable to total reflectance spectra.
[0132] In summary, this Example provides a new measurement scheme
that can yield artifact-free total reflectance spectra
R.sub.TA(.lamda.), which can be easily integrated into the surgical
environment.
EXAMPLE 3
Validating the Efficiency of Diffuse Reflectance Spectroscopy in
Guiding Brain Tumor Resection
[0133] This Example shows that diffuse reflectance spectroscopy is
effective in detecting pediatric brain tumors intraoperatively. To
achieve this goal, the new diffuse reflectance measurement scheme,
described in the Examples 1 and 2, is incorporated into a surgical
microscope. The hybrid system is used to acquire artifact-free
diffuse reflectance spectra from in vivo brain during brain tumor
surgery, and the measured spectra are applied to the discrimination
algorithm. The diagnostic results yielded by the discrimination
algorithm are compared with the histological records, and the
accuracy of the new intraoperative diagnostic system is identified.
Details of this study are provided below.
Reflectance Standard Development
[0134] As indicated by Eq. (1) and Eq. (2), calculating total
reflectance R.sub.TM using Rd requires prior knowledge about the
power of the electromagnetic signals at the site of investigation,
I.sub.o. Quantitatively, R.sub.TM=Rd/(c.times.I.sub.0). The
strategy we propose to quantify J.sub.o involves a disposable
diffuse reflectance standard with known reflectivity (RF). To meet
the needs of the in vivo human studies, the standard should possess
the following properties. First of all, the material of the
standard should be highly biocompatible and able to withstand the
sterilization procedure (e.g., gas sterilization) used in the
hospitals. Secondly, the angular dependence of Rd of the standard
should be comparable to that of biological tissues. This will
ensure that the collection geometry factor c in Eq. (1) does not
introduce unwanted alterations to R.sub.TM. Thirdly, the standard
should have stable optical properties and be usable in a wet
environment. In order to satisfy all the above criteria, we will
test various materials that are used in the surgical field. The
reflectivity (RF) of the evaluated material and the angular
dependence of its Rd will be quantified, before and after
treatments (e.g., wetting and sterilization), using a
spectrophotometer and an optical goniometer. The material that
processes the same angular dependence of Rd of biological tissue is
used to fabricate disk-shape diffuse reflectance standards that are
5 mm in diameter and 2 mm thick. During an in vivo tissue
investigation, one reference standard is placed on top of the
investigated site, from which a reference diffuse reflectance
spectrum (Rd.sub.ref) is measured. Then, illumination power at the
site of investigation can be estimated by
c.times.I.sub.0=Rd.sub.ref/RF. Once the reference standard is
removed, another diffuse reflectance spectrum is acquired from the
target biological tissue (Rd.sub.sample). Total reflectance from
this investigated site is then calculated by
R.sub.TM=(Rd.sub.sample.times.RF)/Rd.sub.ref. During this entire
spectral acquisition procedure, the hybrid imaging system is not
moved, so as to maintain constant collection geometry (i.e., c in
Eq. (1)).
Spectral Interpretation Software Development
[0135] A software program processes and interprets
R.sub.TM(.lamda.). Briefly, R.sub.TM(.lamda.) from each
investigated site is used in conjunction with a spectral
interpretation algorithm derived from published
articles.sup.[60,61] or otherwise known to a person of ordinary
skill in the art to identify the absolute level of SatO.sub.2. In
addition, R.sub.TM(.lamda.) is converted to a'(.lamda.) using the
reference table shown in FIG. 4. This conversion can be limited to
the spectral range of 400 to 600 nm, if the volume of investigation
of the system is intended to be limited, e.g., <5 mm diameter.
The mean and standard deviation of nSatO.sub.2 and a'(.lamda.) from
all normal tissue sites is calculated and then used to create the
threshold in the tissue differentiation algorithm shown in Section
3.1 and FIG. 3A.
Evaluation of the Hybrid Imaging Spectroscopy System
[0136] The main purpose of this evaluation study is to quantify the
measurement accuracy of the new diffuse reflectance measurement
scheme, as well as the hybrid imaging system, for a wide range of
optical properties as well as collection geometry (e.g.,
measurement distance and observation angle). The evaluation study
is performed using tissue phantoms with known optical
properties:.sup.[60] These phantoms consist of a mixture of whole
blood (absorbers) and scatterers in a saline solution. The blood
samples are heparinized after their acquisition from healthy human
subjects. The scatterers added to the solution are 0.1, 0.5, 1.0,
and 5.0 .mu.m diameter microspheres (Polysciences Inc., Warrington
Pa.). From the phantoms, total reflectance spectra (R.sub.TM) is
acquired using the hybrid system. The absorption and scattering
properties derived from R.sub.TM are compared with those of the
corresponding phantoms, and estimation errors quantified using
statistical tools.
In vivo Validation Study
[0137] We will verify the accuracy of the system developed for
intraoperative brain tumor detection in an in vivo study. Here,
histology is used as the gold standard for comparison. In addition,
the accuracy of the system is compared with that of the
intraoperative ultrasound and frameless stereotactic navigation
systems (Brainlab Inc., Westchester, Ill.).
[0138] Total reflectance spectra are measured from brain tumors and
normal brain in pediatric patients in vivo, using the hybrid
imaging spectroscopy system developed in Example 2.
[0139] The study protocol described below is followed. For each
patient studied, the optical investigation is performed prior to,
during, and after brain tumor removal. To establish the ranges of
optical/physiological parameters of normal brain tissue, five or
more independent total reflectance spectra R.sub.TM(.lamda.) are
acquired from normal brain tissue (e.g., at least 2 cm away from
the resection margin) prior to initiation of tumor resection.
During the process of tumor removal, R.sub.TM(.lamda.) are acquired
from both the tumor core and tumor margins. Additional
R.sub.TM(.lamda.) are acquired when the surgeon approaches the
resection margin to identify the sensitivity of the proposed
optical system in detecting tumor margins. Prior to spectral
acquisition, each investigated site will be rinsed gently with
saline to remove surface blood and other debris.
[0140] With the detection point (e.g., the center point of the
field of view of the surgical microscope) placed at the site of
investigation, total reflectance spectra R.sub.TM, Tissue(.lamda.)
is acquired continuously for at least 3 seconds to verify the
reproducibility of the spectral signals. Upon completing the
spectral measurements at the investigated site, the reflection
standard is placed on the site of investigation, and another set of
total reflectance spectra R.sub.TM, Ref(.lamda.) are acquired for
reference purposes. It should be noted that the surface of the
standard should be parallel to that of the investigated site, and
that the surgical microscope should not be moved during the
spectral acquisition procedure. Each investigated site is marked
and its tissue characteristics identified on the appropriate
pre-operative magnetic resonance (MR) images, as well as on
intraoperative ultrasound images, using frameless stereotaxy. Each
site is also categorized by the surgeon, by means of visual
inspection, either as normal, tumor margin, or tumor center.
Biopsies are taken from each investigated site, except when the
investigated site is situated outside the resection margins. All
biopsies are preserved in 10% formalin, and then processed and read
by a neuropathologist. During histological evaluation, each biopsy
sample is identified as either tumor or normal tissue, and the
volume fraction of tumor (e.g., 100%, >75%, >50%, >25%,
0%) identified.
Data Analysis
[0141] Acquired spectra are processed using the procedure proposed
in Example 2 to obtain the two indicative parameters representing
the tissue characteristics: nSatO.sub.2 and a'(.lamda.).
Subsequently, these parameters are used in conjunction with the
discrimination method described above to identify the
characteristics of the investigated tissue. These classification
results will be compared with their corresponding histological
records. In this comparison, we will use sensitivity, specificity,
and receiver operating characteristics (ROC) analysis to quantify
the accuracy of the current discrimination algorithm. Furthermore,
the level of alterations in the indicative parameters are
quantitatively compared with the tumor volume fraction identified
by histology via correlation analysis, which should yield insights
into the sensitivity as well as the detection limit of the total
reflectance spectroscopy system proposed in this Aim. Finally, the
accuracy of the optical guidance system is compared with those
based on the intraoperative ultrasound, the frameless stereotactic
navigation systems (MRI), and surgeons' visual inspection.
[0142] In this comparison, the histological identities of the
investigated sites are used as the gold standard, against which the
accuracy of the four modalities in detecting tumor margins is
calculated.
EXAMPLE 4
Identifying the Microscopic Flow and Vascular Architectural
Characteristics of Brain Tumors Thereby Distinguishing them from
Normal Brain
[0143] The primary objective of this Example is incorporating the
flow and vascular architectural characteristics of tumors to
enhance brain tumor demarcation. Flow and vascular architectural
characteristics can enhance the efficiency of intraoperative brain
tumor demarcation. .sup.[37-40]
System Upgrade
[0144] The methodology for measuring the microcirculation
characteristics employed in this
[0145] Example will be orthogonal polarization spectral
imaging..sup.[41-42] To implement this methodology into the
spectroscopy surgical microscope system mentioned in Example 3, the
following modifications are done.
[0146] First, microscope illumination is polarized using a linear
polarizer, which prevents any interference from specular reflection
in the recorded images. Second, a camera system is attached to the
beam splitter of the surgical microscope to which the point
detection system is attached (FIG. 2). The camera system is
designed in the way that it provides an additional 2.times.
magnification. The camera used is a high-speed and high-resolution
DSLR camera with video capture capability. The video frame rate
should exceed 60 frames per second. A linear polarizer is mounted
at the entrance of the camera system to eliminate specular
reflection from the brain surface. In addition, an optical filter
is used to select the band of diffuse reflection. For the purpose
of studying the microscopic flow and vasculature characteristics of
in vivo brain, diffuse reflectance images at 540 nm and 560 nm are
recorded. The selection of these two wavelengths is based on the
absorption characteristics of oxy- and deoxy hemoglobin. The
system, once set up, is calibrated using layered tissue phantoms to
identify its probing depth as a function of optical properties and
to ascertain its functionality.
Quantification of Microvascular Flow Characteristics
[0147] A software program analyzes the video frames recorded using
the imaging system to identify the flow characteristics of the
microcirculation (arterioles and venules). Shifts in intravascular
optical patterns over a defined time interval are used to calculate
velocity by spatial correlation, a method developed for orthogonal
polarization spectral imaging as well as intravital
microscopy..sup.[43-44] The overall data process time for this
analysis is limited to 10 seconds or less, so the surgeons can
receive the diagnostic feedback immediately in the operating room
and use it to make surgical decision.
Quantification of Microvascular Architectural Characteristics
[0148] Images at the resection front are taken at 540 and 560 nm
and then used to identify the characteristics of microvascular
architecture. An imaging process technique called edge detection is
applied to the image to highlight the blood vessels
(arteries/arterioles and veins/venules). Using the ratio of these
two images, the relative SatO.sub.2 and, hence, separate
arteries/arterioles from veins/venules are identified. To create
quantitative indices for these regional microvascular architectural
characteristics, the histogram of pixel intensity within a fixed
window of the processed image can be used (see FIGS. 5A, 5B, and
5C) to represent vessel density within the area of interest. In
addition, vessel length per area, the number of vessel segments per
area, and mean vessel diameter to quantify the architectural
features of the vasculature can be utilized. An image processing
program can automate these measurements. The overall data process
time for this analysis is limited to 5 seconds or less, so the
surgeons can receive the diagnostic feedback immediately in the
operating room and use it to make surgical decision.
In vivo Validation Study
[0149] The usefulness of microscopic flow and microvascular
architectural characteristics for intraoperative brain tumor/brain
tumor margin detection in vivo is established. Here, histology is
used as the gold standard. The study protocol and procedure are the
same as that described in Reflectance Standard Development in
Example 3. Video recording from the brain surface is performed
while the diffuse reflectance spectra are acquired. The recording
duration is 10 seconds for each site.
Data Analysis
[0150] The recorded video is processed using the two programs
developed to quantify microvascular flow and microvascular
architectural characteristics. This leads to the following
parameters for each investigated site: blood flow velocity in
arterioles and venules; blood flow kinetics (stop flow, spurt flow,
and continuous flow); mean vessel diameter; total length of the
vessels; and blood vessel density within the site of investigation.
These parameters are analyzed collectively with the regional
SatO.sub.2 and a'(.lamda.), derived from R.sub.TM(.lamda.).
[0151] First of all, the normal ranges of these parameters are
calculated using video recordings from normal cortex. In addition,
the distribution characteristics (e.g., normal distribution) of
these parameters from normal cortex are evaluated. Secondly, the
ranges of these parameters from brain tumors and their margins are
calculated and compared against those derived from normal cortex
using non-parametric or parametric statistical comparison methods,
as indicated. Parameters that demonstrate significant differences
between normal cortex and tumor/tumor margins are used to
reconstruct the discrimination algorithm mentioned in Example
1.
[0152] The efficiency of the new discrimination algorithm, in terms
of intraoperative brain tumor detection, is evaluated via ROC
analysis and compared against that of the original algorithm.
Furthermore, the level of alterations in the indicative parameters
are quantitatively compared with the tumor volume fraction
identified by histology via correlation analysis, which yields
insights into the sensitivity as well as the detection limit of the
imaging modalities proposed in this Example.
[0153] All patents, patent applications, provisional applications,
and publications referred to or cited herein, including those
listed in the "References" section, are incorporated by reference
in their entirety, including all figures and tables, to the extent
they are not inconsistent with the explicit teachings of this
specification.
[0154] It should be understood that the examples and embodiments
described herein are for illustrative purposes only and that
various modifications or changes in light thereof will be suggested
to persons skilled in the art and are to be included within the
spirit and purview of this application.
REFERENCES
[0155] 1. Statistical Report: Primary Brain and Central Nervous
System Tumors Diagnosed in the United States in 2004-2007. 2011;
Available from: http://www.cbtrus.org/2007-2008/2007-20081.html.
[0156] 2. Rutka, J. T. and J. S. Kuo, Pediatric surgical
neuro-oncology: current best care practices and strategies. J
Neurooncol, 2004. 69(1-3): p. 139-50. [0157] 3. Karajannis, M., J.
C. Allen, and E. W. Newcomb, Treatment of pediatric brain tumors. J
Cell Physiol, 2008. 217(3): p. 584-9. [0158] 4. Fernandez, C., et
al., Pilocytic astrocytomas in children: prognostic factors--a
retrospective study of 80 cases. Neurosurgery, 2003. 53(3): p.
544-53; discussion 554-5. [0159] 5. Pollack, I. F., et al.,
Low-grade gliomas of the cerebral hemispheres in children: an
analysis of 71 cases. J Neurosurg, 1995. 82(4): p. 536-47. [0160]
6. Hirsch, J. F., et al., Benign astrocytic and oligodendrocytic
tumors of the cerebral hemispheres in children. J Neurosurg, 1989.
70(4): p. 568-72. [0161] 7. Laws, E. R., Jr., et al., Neurosurgical
management of low-grade astrocytoma of the cerebral hemispheres. J
Neurosurg, 1984. 61(4): p. 665-73. [0162] 8. Mercuri, S., A. Russo,
and L. Palma, Hemispheric supratentorial astrocytomas in children.
Long-term results in 29 cases. J Neurosurg, 1981. 55(2): p. 170-3.
[0163] 9. Maciunas, R. J., Approaches to Frame-Based and Frameless
Stereotaxis. Minimally Invasive Therapy of the Brain, ed. A. D.
Salles and R. Lufkin 1997, New York: Thieme Medical Publishers,
Inc. [0164] 10. Maciunas, R. J., Pitfalls, in Image-guided
neurosurgery: clinical applications of surgical navigation, G. H.
Barnett, D. W. Roberts, and R. J. Maciunas, Editors. 1998, Quality
Medical Publishing, Inc.: St. Louis, Mo. p. 43-60. [0165] 11.
Dorward, N., et al., Postimaging brain distortion: magnitude,
correlates, and impact on neuronavigation. J Neurosurg, 1998.
88(4): p. 656-662. [0166] 12. Hill, D., et al., Measurement of
intraoperative brain surface deformation under a craniotomy.
Neurosurgery, 1998. 43(3): p. 514-526. [0167] 13. Cheong, W. -F.,
Summary of optical properties, in Optical-thermal response of
laser-irradiated tissue, A. Welch and M. v. Gemert, Editors. 1995,
Plenum Press: New York. p. 233-301. [0168] 14. Prahl, S. Optical
absorption of hemoglobin. [Internet] 1999; Available from:
http://omlc.ogi.edu/spectra/hemoglobin/index.html. [0169] 15.
Siegemund, M., J. van Bommel, and C. Ince, Assessment of regional
tissue oxygenation. Intensive Care Med, 1999. 25(10): p. 1044-60.
[0170] 16. Benaron, D. A., et al., Continuous, noninvasive, and
localized microvascular tissue oximetry using visible light
spectroscopy. Anesthesiology, 2004. 100(6): p. 1469-75. [0171] 17.
Verdant, C. and D. De Backer, How monitoring of the
microcirculation may help us at the bedside. Curr Opin Crit Care,
2005. 11(3): p. 240-4. [0172] 18. Knotzer, H. and W. R. Hasibeder,
Microcirculatory function monitoring at the bedside--a view from
the intensive care. Physiol Meas, 2007. 28(9): p. R65-86. [0173]
19. Mallia, R., et al., Oxygenated hemoglobin diffuse reflectance
ratio for in vivo detection of oral pre-cancer. J Biomed Opt, 2008.
13(4): p. 041306. [0174] 20. Wang, H. W., et al., Diffuse
reflectance spectroscopy detects increased hemoglobin concentration
and decreased oxygenation during colon carcinogenesis from normal
to malignant tumors. Opt Express, 2009. 17(4): p. 2805-17. [0175]
21. Floeth, F. W., et al., Comparison of (18)F-FET PET and 5-ALA
fluorescence in cerebral gliomas. Eur J Nucl Med Mol Imaging, 2011.
38(4): p. 731-41. [0176] 22. Pichlmeier, U., et al., Resection and
survival in glioblastoma multiforme: an RTOG recursive partitioning
analysis of ALA study patients. Neuro Oncol, 2008. 10(6): p.
1025-34. [0177] 23. Valdes, P. A., et al., Quantitative
fluorescence in intracranial tumor: implications for ALA-induced
PpIX as an intraoperative biomarker. J Neurosurg, 2011. 115(1): p.
11-7. [0178] 24. Veiseh, M., et al., Tumor paint: a
chlorotoxin:Cy5.5 bioconjugate for intraoperative visualization of
cancer foci. Cancer Res, 2007. 67(14): p. 6882-8. [0179] 25. Ennis,
S. R., et al., Transport of 5-aminolevulinic acid between blood and
brain. Brain Res, 2003. 959(2): p. 226-34. [0180] 26. Nath, A., et
al., Effect of probe pressure on cervical fluorescence spectroscopy
measurements. J Biomed Opt, 2004. 9(3): p. 523-33. [0181] 27. Shim,
M. G., et al., In vivo near-infrared Raman spectroscopy:
demonstration of feasibility during clinical gastrointestinal
endoscopy. Photochem Photobiol, 2000. 72(1): p. 146-50. [0182] 28.
Rivoire, K., et al., The effects of repeated spectroscopic pressure
measurements on fluorescence intensity in the cervix. Am J Obstet
Gynecol, 2004. 191(5): p. 1606-17. [0183] 29. Lim, L., et al.,
Probe pressure effects on human skin diffuse reflectance and
fluorescence spectroscopy measurements. J Biomed Opt, 2011. 16(1):
p. 011012. [0184] 30. Reif, R., et al., Analysis of changes in
reflectance measurements on biological tissues subjected to
different probe pressures. J Biomed Opt, 2008. 13(1): p. 010502.
[0185] 31. Ruderman, S., et al., Analysis of pressure, angle and
temporal effects on tissue optical properties from
polarization-gated spectroscopic probe measurements. Biomed Opt
Express, 2010. 1(2): p. 489-499. [0186] 32. Ti, Y. and W.C. Lin,
Effects of probe contact pressure on in vivo optical spectroscopy.
Opt Express, 2008. 16(6): p. 4250-62. [0187] 33. Lin, W. C., et
al., Optical spectroscopy for in-vitro differentiation of pediatric
neoplastic and epileptogenic brain lesions. J Biomed Opt, 2009.
14(1): p. 014028. [0188] 34. Chen, P., B. Fernald, and W. Lin,
Estimation of regional hemoglobin concentration in biological
tissues using diffuse reflectance spectroscopy with a novel
spectral interpretation algorithm. Physics in Medicine and Biology,
2011. 56(13): p. 3985-4000. [0189] 35. Chen, P. C. and W. C. Lin,
Spectral-profile-based algorithm for hemoglobin oxygen saturation
determination from diffuse reflectance spectra. Biomed Opt Express,
2011. 2(5): p. 1082-96. [0190] 36. Farrell, T. J., M. S. Patterson,
and B. Wilson, A diffusion theory model of spatially resolved,
steady-state diffuse reflectance for the noninvasive determination
of tissue optical properties in vivo. Med Phys, 1992. 19(4): p.
879-88. [0191] 37. Pennings, F. A., C. Ince, and G. J. Bouma,
Continuous real-time visualization of the human cerebral
microcirculation during arteriovenous malformation surgery using
orthogonal polarization spectral imaging. Neurosurgery, 2006.
59(1): p. 167-71; discussion 167-71. [0192] 38. Uhl, E., et al.,
Intraoperative detection of early microvasospasm in patients with
subarachnoid hemorrhage by using orthogonal polarization spectral
imaging. Neurosurgery, 2003. 52(6): p. 1307-15; disacussion 1315-7.
[0193] 39. Jain, R. K., et al., Angiogenesis in brain tumours. Nat
Rev Neurosci, 2007. 8(8): p. 610-22. [0194] 40. Vajkoczy, P. and M.
D. Menger, Vascular microenvironment in gliomas. J Neurooncol,
2000. 50(1-2): p. 99-108. [0195] 41. Groner, W., et al., Orthogonal
polarization spectral imaging: a new method for study of the
microcirculation. Nat Med, 1999. 5(10): p. 1209-12. [0196] 42.
Cerny, V., Z. Turek, and R. Parizkova, Orthogonal polarization
spectral imaging. Physiol Res, 2007. 56(2): p. 141-7. [0197] 43.
Torp, H., K. Kristoffersen, and B. A. J. Angelsen, Autocorrelation
Techniques in Color-Flow Imaging--Signal Model and Statistical
Properties of the Autocorrelation Estimates. Ieee Transactions on
Ultrasonics Ferroelectrics and Frequency Control, 1994. 41(5): p.
604-612. [0198] 44. Slaaf, D. W., et al., A Versatile Incident
Illuminator for Intravital Microscopy. International Journal of
Microcirculation-Clinical and Experimental, 1987. 6(4): p.
391-397.
* * * * *
References