Automated quantification of colonic crypt morphology using integrated microscopy and optical coherence tomography

Article (PDF Available)inJournal of Biomedical Optics 13(5):054055 · September 2008with21 Reads
DOI: 10.1117/1.2993323 · Source: PubMed
Abstract
Colonic crypt morphological patterns have shown a close correlation with histopathological diagnosis. Imaging technologies such as high-magnification chromoendoscopy and endoscopic optical coherence tomography (OCT) are capable of visualizing crypt morphology in vivo. We have imaged colonic tissue in vitro to simulate high-magnification chromoendoscopy and endoscopic OCT and demonstrate quantification of morphological features of colonic crypts using automated image analysis. 2-D microscopic images with methylene blue staining and correlated 3-D OCT volumes were segmented using marker-based watershed segmentation. 2-D and 3-D crypt morphological features were quantified. The accuracy of segmentation was validated, and measured features are in agreement with known crypt morphology. This work can enable studies to determine the clinical utility of high-magnification chromoendoscopy and endoscopic OCT, as well as studies to evaluate crypt morphology as a biomarker for colonic disease progression.
Automated quantification of colonic crypt morphology
using integrated microscopy and optical coherence
tomography
Xin Qi
Yinsheng Pan
Zhilin Hu
Wei Kang
Case Western Reserve University
Department of Biomedical Engineering
Cleveland, Ohio 44106
Joseph E. Willis
Case Western Reserve University
Department of Pathology
Cleveland, Ohio 44106
Kayode Olowe
Michael V. Sivak Jr.
Case Western Reserve University
Department of Medicine
Cleveland, Ohio 44106
Andrew M. Rollins
Case Western Reserve University
Department of Biomedical Engineering
and
Department of Medicine
Cleveland, Ohio 44106
Abstract. Colonic crypt morphological patterns have shown a close
correlation with histopathological diagnosis. Imaging technologies
such as high-magnification chromoendoscopy and endoscopic optical
coherence tomography OCT are capable of visualizing crypt mor-
phology in vivo. We have imaged colonic tissue in vitro to simulate
high-magnification chromoendoscopy and endoscopic OCT and
demonstrate quantification of morphological features of colonic
crypts using automated image analysis. 2-D microscopic images with
methylene blue staining and correlated 3-D OCT volumes were seg-
mented using marker-based watershed segmentation. 2-D and 3-D
crypt morphological features were quantified. The accuracy of seg-
mentation was validated, and measured features are in agreement
with known crypt morphology. This work can enable studies to deter-
mine the clinical utility of high-magnification chromoendoscopy and
endoscopic OCT, as well as studies to evaluate crypt morphology as a
biomarker for colonic disease progression.
© 2008 Society of Photo-Optical
Instrumentation Engineers. DOI: 10.1117/1.2993323
Keywords: colonic crypt morphological patterns; image processing; optical coher-
ence tomography
.
Paper 08159R received May 15, 2008; revised manuscript received Jul. 18, 2008;
accepted for publication Jul. 22, 2008; published online Oct. 9, 2008.
1 Introduction
Colorectal cancer is the second leading cause of cancer-
related death in the United States. In 2007, over 153,760 new
cases were diagnosed and over 52,180 deaths resulted from
colorectal cancer.
1
As these cancers have a long development
phase from inception to cancer,
2
theoretically all these cancers
are preventable with accurate universal screening.
Screening colonoscopy substantially reduces the risk of
colorectal cancer in long-term follow-up.
3,4
The identification
and removal of polyps has reduced the rate of mortality from
colon cancer.
5
However, the decrease in long-term incidence
of colorectal cancer varies greatly in these patients. Reasons
for these differences have been attributed to variations in
study populations, colonoscopy technique,
6
and the presence
of hard-to-detect cancer precursor lesions.
7
Colonic mucosa contains numerous pits, the crypts of Lie-
berkühn, with roundish openings arranged in a regular pattern
on the surface of normal colorectal mucosa.
8
Prior studies
have shown that shapes of colonic crypts change with disease
state and show characteristic patterns.
911
It has also been
shown that crypt patterns and histopathological diagnosis are
well correlated.
1115
For example, aberrant crypt foci ACF
were first described by Bird et al.
16
in 1987 in mice given
carcinogens. Aberrant colonic crypts have larger lumens than
normal crypts and have a thickened epithelium that stains
darker with dyes compared to surrounding crypts.
17,18
The lu-
mens of the aberrant crypts often appear slit-shaped rather
than circular. Aberrant crypts usually occur in groups of four
or more hence, ACF and are readily identified on the mu-
cosal surface of the colon by microscopy. There is compelling
evidence that ACF are biomarkers for colorectal
carcinoma.
1923
ACF are more numerous in patients with co-
lon cancer.
1921
It is also known that the orientation of crypts
changes with the progression of the disease.
2428
Normal co-
lonic crypts are oriented parallel to each other. Aberrant crypts
and crypts in cancerous tissue are oriented less parallel to
each other.
High-resolution endoscopic methods have been demon-
strated to visualize colonic crypts in humans, including high-
magnification chromoendoscopy, high-magnification narrow
band imaging, and endoscopic confocal microscopy. Chro-
moendoscopy is a technique that utilizes tissue stains, such as
methylene blue, applied topically to the gastrointestinal mu-
cosa to better characterize lesions.
29,30
Narrow band imaging
NBI may be regarded as the optical analog of chromoendo-
1083-3668/2008/135/054055/11/$25.00 © 2008 SPIE
Address all correspondence to: Andrew M. Rollins, Departments of Biomedical
Engineering and Medicine, Case Western Reserve University, 10900 Euclid
Ave., Cleveland, OH 44106; Tel: 216-368-1917; Fax: 216-368-0847; E-mail:
rollins@case.edu.
Journal of Biomedical Optics 135, 054055 September/October 2008
Journal of Biomedical Optics September/October 2008
Vol. 135054055-1
scopy, without the use of staining agents, which is accom-
plished by narrowing the bandwidth of spectral transmittance
of the red/green/blue optical filters used in the sequential-
frame imaging method for videoendoscopic imaging.
3133
Both chromoendoscopy and NBI can be used in conjunction
with magnification endoscopy, which uses optical systems to
magnify endoscopic images, allowing one to see the minute
colonic surface structures.
3335
Although these high-
magnification methods can visualize the morphology of the
crypt openings at the mucosal surface, they cannot visualize
the three-dimensional 3-D structure of the crypts.
Endoscopic confocal microscopy produces high-resolution
images of the gastrointestinal epithelium and is also sensitive
to fluorescence. Furthermore, confocal imaging is capable of
sectioning in depth and can therefore visualize 3-D structures.
In vivo confocal fluorescence microscopy has been demon-
strated in the upper
36,37
and lower
3739
gastrointestinal GI
tracts. However, the depth of crypts range from
100 to
1000
m,
26
while the in-depth observation range of endo-
scopic confocal microscopy is limited to a few hundred
micrometers.
36,38
Therefore, this method cannot always visu-
alize the entire depth of crypts.
Endoscopic optical coherence tomography EOCT pro-
vides subsurface, high-resolution real-time imaging of GI
mucosa.
4042
EOCT provides depth sectioning to 1.5 mm
from the mucosal surface, overcoming the limitations of mag-
nification endoscopy and endoscopic confocal microscopy for
visualizing the 3-D structure of crypts. Furthermore, in vivo
EOCT imaging of colon polyps has shown good correlation
with the pathology observed in these polyps after their
removal.
43
3-D EOCT imaging of the colon in an in vivo
animal model rabbit has recently been demonstrated, includ-
ing a discussion of crypt visualization and basic manual quan-
tification of crypt size.
44
3-D EOCT imaging of human co-
lonic crypts has not been demonstrated to date, but the
potential is clear.
44,45
It has been shown that image analysis can quantify colonic
crypt structures and improve interpretation of crypt patterns.
Quantification of crypt structures in histopathological
24,25,4649
and chromoendoscopic
28,50
images has been recently demon-
strated. In this work, we demonstrate that colonic crypt struc-
tures can be imaged and quantified by automated algorithms
using 3-D optical coherence tomography and by microscopy
with methylene blue staining.
2 Methods and Materials
Under a protocol approved by the Institutional Review Board
of University Hospitals Case Medical Center, we studied 79
samples of resected fresh colon tissues obtained from colec-
tomies 59 patients, including 31 female,
19 to 86 years old
using a microscope-integrated bench-top OCT scanner de-
scribed elsewhere.
51
The time-domain system utilized a high-
power broadband source centered around
1310 nm with a
bandwidth of
70
m. Axial and transverse resolution was
8
m FWHM coherence length in tissue and 16
m 1 / e
2
spot diameter, respectively. With optical power of 5mWin-
cident on the sample, measured sensitivity was
116 dB. Im-
ages were acquired at 8 frames per second with 500 axial
lines A-scans per image. Data included 14 normal samples
from 12 patients, 44 samples of normal-appearing tissue ad-
jacent
5–15 cm to cancer NA兲共from 32 patients, and 21
malignant samples from 18 patients. Cancer was diagnosed
by routine clinical pathology, and tissue was classified as
“normal” or “normal adjacent to cancer” by the surgeon. The
clinical target for high-resolution colon imaging is to detect
abnormal variation of crypt structure in fields of apparently
normal colonic tissues where premalignancy may be sus-
pected but not detectable by conventional endoscopy. While
malignant colonic tissue is readily detectable by endoscopy
and therefore not a clinical diagnostic target for OCT imaging
nor for high-magnification chromoendoscopy, we include im-
ages of malignant samples in our descriptive results in order
to demonstrate that our observations match what is already
known about the architecture of malignant colon. We ex-
cluded the malignant samples from the data set for automatic
quantification of colonic crypt morphology.
2.1 Sample Preparation and Image Acquisition
Preparation of colon tissue was meant to simulate approxi-
mately in vivo imaging with high-magnification chromoendo-
scopy. First, the surface of the freshly resected colon speci-
men was washed using 10% acetylcysteine, a mucolytic
agent, and washed by water to remove the coating mucous
and to prepare the tissue for dye staining. Subsequently, tissue
was stained with 10% methylene blue to enhance the contrast
of crypts under the microscope. After dye staining, several
microscopic images were taken under different magnifications
using a commercial CCD camera PixelLink model PL-
A642. 3-D OCT image volumes were obtained in the same
region of interest ROI and registered with the microscopic
images. For each 3-D OCT volume, 9 times frame averaging
was used to reduce the speckle noise.
2.2 Microscopic Image and OCT En Face View
Segmentation
Each 2-D color microscopic image was transformed to a gray-
level image. Then, uneven illumination was corrected by mor-
phological opening with a large structure element,
52
and back-
ground noise was removed using a
6 6 median filter.
Subsequently, the morphological gradient image was calcu-
lated in order to capture the edges of the crypts.
53
A modified
gradient image was created by thresholding using a threshold
value calculated by Otsu’s method.
54
In the modified gradient
image, gradient intensities lower than the threshold were set
to zero, and intensities higher than threshold were unchanged.
The crypt edges produce the strongest gradient values, so this
method prevents oversegmentation by removing gradients
representing structures other than the crypt edges.
Watershed segmentation is subject to oversegmentation be-
cause it is sensitive to every local minimum and maximum.
This can be avoided by providing foreground markers, which
are locations known to be inside of crypt lumens, and back-
ground markers, which are locations known to be outside of
crypt lumens. These were automatically computed by mor-
phological reconstruction.
55,56
Watershed transformation using
immersion simulation
57
was executed on the superimposed
foreground and background markers and modified gradient
image to segment the crypts within the ROI. Because strands
of connective tissue within intervening nonpitted mucosa in-
dicated by white arrows in Figs. 3 and 4 have similar contrast
Qi et al.: Automated quantification of colonic crypt morphology
Journal of Biomedical Optics September/October 2008
Vol. 135054055-2
characteristics as crypts, they can lead to inclusion of “false
crypts” among the set of segmented crypts in an ROI. There-
fore, crypts having a ratio of major axis length to minor axis
length larger than three were rejected as false crypts. Finally,
the crypts touching the edges of the ROI were removed prior
to subsequent morphological feature extraction.
In the OCT volume data, each en face plane is analogous
to a gray-level microscopic image. Therefore, we adapted the
2-D microscopic image segmentation described above to seg-
ment the crypts within each en face plane of the 3-D OCT
volumes, resulting in segmented crypts in three dimensions,
which could be visualized using volume rendering.
To validate the automated segmentation results, one repre-
sentative microscopic image and one representative OCT vol-
ume were manually segmented, and the manual segmentation
results were quantitatively compared with the marker-
controlled watershed segmentation results. The contours of
crypts within one micrograph of normal colonic tissue were
manually traced and compared with the contours resulting
from marker-controlled watershed segmentation of the same
image. In the same way, the contours of the crypts within each
en face plane of one OCT volume were manually traced and
compared with the contours resulting from marker-controlled
watershed segmentation of the same volume. The precision
ratio PR and the match ratio MR
58
between the manually
segmented contours and the marker-controlled watershed seg-
mented contours were calculated to quantify the comparisons.
The PR is defined as
PR=1−N
diff
/ N
M
, where N
diff
is the
number of pixels that differ between the manually determined
contour and the marker-controlled watershed segmented con-
tour and
N
M
is the number of pixels in the manual contour.
The MR is defined as
MR=1−Area
M
Area
W
/ Area
M
,
where
Area
M
and Area
W
denote the areas enclosed by manu-
ally segmented contours and the areas covered by marker-
controlled watershed segmented contours, respectively.
2.3 Morphological Feature Extraction
Within the ROI of the 2-D microscopic images, the following
features of each segmented crypt were extracted: the area, the
major axis length, minor axis length, and eccentricity of the
best-fitting ellipse, and the solidity. Solidity is the ratio of the
area of the crypt to the area of its convex hull and is a mea-
sure of the degree to which the crypt is concave or convex. In
addition, the density of crypts within the ROI was calculated
as the number of crypts divided by the area of the ROI.
The 3-D OCT image volumes were visualized by volume
rendering and were used to quantify crypt orientation. In order
to quantify the crypt orientation, it is useful to extract the
“skeleton” of each crypt, the curve representing the center of
the lumen of the crypt in three dimensions. The conventional
method for skeleton extraction, morphological thinning via
distance transformation, led to unwanted branches
59
instead of
a single curve, which complicated quantification of the orien-
tation Fig. 7. Therefore, we developed a method to estimate
the skeleton of each crypt by linking the centroids of each
crypt in the ROI at each en face plane of the segmented 3-D
OCT volume.
Skeleton estimation by centroid-linking was carried out in
two steps. First, the centroids of the segmented crypts in each
en face plane of the OCT volume were calculated. Second, the
centroids of each crypt were linked with the centroids of the
same crypts in adjacent en face planes, resulting in strings of
coordinates representing the skeletons of each crypt. This
linking step was initiated from a plane
500
m below the
surface of the tissue, approximately the location of the focus
of the scanner. This initial plane was selected using the vol-
ume rendering to include as many crypts as possible. Then,
for each crypt identified in the initial plane, the linking pro-
ceeded deep to the initial plane until either the crypt termi-
nated or the data ended. Then the linking was carried out from
the initial plane up to the tissue surface.
Because of noise in the images, occasionally crypts were
not segmented in single en face planes, resulting in “missing
layers.” Also, occasionally false crypts were identified due to
oversegmentation. In order to link discontinuous crypts
caused by missing layers, if an overlying crypt was not found
in the plane adjacent to the current plane, subsequent planes
were searched until an overlaying crypt was found. This crypt
centroid was linked and skeleton coordinates were assigned to
missing layers by linear interpolation. If an overlaying crypt
was not found within a maximum gap of
30
m, the crypt
was assumed to be terminated. False crypts were only prob-
A
1mm
A’
1mm
B’
B
C’
C
DD
Fig. 1 Examples of the appearance of normal colonic mucosa Aand
A
, ACF BandB
, moderately differentiated adenocarcinoma C
and C
, and mucinous adenocarcinoma arising from a tubulovillous
adenoma DandD
. The first column shows micrographs of the
tissue samples under methylene blue staining. The second column
shows cross-sectional OCT images of the same samples at the loca-
tions indicated by the yellow lines in the micrographs. Color online
only.
Qi et al.: Automated quantification of colonic crypt morphology
Journal of Biomedical Optics September/October 2008
Vol. 135054055-3
lematic if they occurred in the initial plane or if they overlaid
a true crypt. False crypts in the initial plane were mitigated,
because any crypt of
60
m in extent was removed from
analysis. If more than one overlaid segmented crypt was
found in adjacent planes, then the centroids of the crypts hav-
ing the largest ratio of overlaid area were linked. The other
crypt was assumed to be false.
Crypt orientation was quantified from the skeletons by
measuring the straightness of each crypt skeleton and how
parallel they were to each other. In order to quantify the
straightness of each crypt, each crypt skeleton was fit to a line
by calculating the coefficients of the first principal component
of the skeleton. The goodness of fit, assessed by the coeffi-
cient of determination also referred to as the
R
2
value, was
taken as our measure of the straightness of that crypt. The
R
2
value for the fit indicates the fraction of the variation in the
skeleton of crypt that is explained by the line i.e., the closer
that
R
2
value is to 1, the straighter the crypt is. In order to
quantify how parallel the crypts in a ROI were to each other,
the direction vector of the line fitted to the skeleton of each
crypt was compared to the mean direction vector of all of the
crypts in the ROI. The comparison was quantified by calcu-
lating the dot product of each crypt vector and the mean vec-
tor, resulting in a value representing the parallelism. If this
value is close to 1, the crypts within an ROI are more parallel
to each other. Within an ROI, the straightness and parallelism
measures were calculated to represent crypt orientation in the
region.
Running MATLAB version 7.4 on a PC with a
1.83 GHz
CPU and 2GBRAM, the entire image analysis algorithm ran
in
28 s including 15 s of preprocessing, 12 s of segmentation,
and
1s of morphological feature extraction for one micro-
graph. It ran in
1814 s including 632 s of preprocessing,
1033 s of segmentation, and 149 s of morphological feature
extraction, for a 3-D OCT volume 599 slices.
3 Results
3.1 Descriptive Observations
Observed characteristics of normal colonic tissue under the
microscope and OCT include round, small crypts that are uni-
formly distributed,
8
straight, and oriented parallel to each
other. Observed characteristics of ACF under the microscope
and OCT are larger and more often elongated crypts com-
pared with normal surrounding crypts
17,18
and crypt orienta-
tion that is less straight and less parallel. These observations
are consistent with previously reported observations;
8,2426
however, to our knowledge the 3-D orientations of colonic
crypts have not been quantified previously.
25,26
Within the 14 samples of normal colonic tissues, no ACF
were found. Within the 44 samples of apparently normal tis-
sue adjacent to cancer, four ACF were found. ACF are an
C
BA
1mm
1mm
F
E
D
G
I
H
Fig. 2 Images of a typical aberrant crypt focus: A the micrograph with methylene blue staining; B the corresponding OCT B-scan image at the
yellow line location; C the maximum intensity projection of the 3-D OCT volume; and D the OCT en face views at depth of 129 E 258, F
430, G 602, H 774, and I 946
m from the surface of the tissue, respectively. En face planes were filtered with a 3 3 median kernel to reduce
speckle noise. Color online only.
Qi et al.: Automated quantification of colonic crypt morphology
Journal of Biomedical Optics September/October 2008
Vol. 135054055-4
endoscopic diagnosis, not a histopathological diagnosis, based
on observation of the surface appearance of the tissue. Here,
the observation of ACF was based on the documented typical
appearance of ACF and confirmed by a GI pathologist. Within
the 21 samples of malignant tissue, two types of cancer were
observed, moderately differentiated adenocarcinoma and mu-
cinous adenocarcinoma arising from a tubulovillous adenoma.
In the 19 samples of moderately differentiated adenocarci-
noma, dramatically distorted crypt patterns were observed. In
the two samples of moderately differentiated mucinous adeno-
carcinoma arising from a tubulovillous adenoma, small, ob-
lique, tubulovillous structures were observed. Examples of
each of these mucosa types are shown in Fig. 1. Figure 1A is
a micrograph of normal colonic mucosa with methylene blue
staining; Fig. 1
A
is an OCT cross-sectional image, taken at
the location indicated by the yellow line in Fig. 1A. Simi-
larly, Figs. 1B and 1
B
demonstrate an ACF found in a
sample of apparently normal tissue adjacent to a tumor. Fig-
ures 1C and 1
C
demonstrate moderately differentiated
adenocarcinoma, and Figs. 1D and 1
D
demonstrate mod-
erately differentiated mucinous adenocarcinoma arising from
a tubulovillous adenoma.
3.2 Quantitative Results
Images of a typical aberrant crypt focus appear in Fig. 2.
Figure 2A is the micrograph with methylene blue staining;
Fig. 2B is the corresponding OCT B-scan image at the lo-
cation indicated by the yellow line. Figure 2C is the maxi-
mum intensity projection of the 3-D OCT volume. Figures
2D2I are the OCT en face views at depths of 129, 258,
430, 602, 774, and
946
m from the surface of the tissue,
respectively. En face planes were filtered with a
3 3 median
kernel to smooth the contours of segmented crypt boundaries.
Figures 3 and 4 show segmentation results of representative
NA and ACF samples. Figure 3A shows the microscopic
image with methylene blue of a sample of NA colonic tissue,
while Fig. 3B shows the corresponding OCT volume. Figure
3C shows the crypts segmented within the 2-D micrograph;
Fig. 3D shows the crypts segmented in three dimensions
within the OCT volume. Similarly, Fig. 4 shows microscopic
and OCT images of an ACF as well as the crypts segmented
in two and three dimensions.
The results of the segmentation validation experiments are
shown in Figs. 5 and 6. Figure 5 shows the results of the
manual and marker-controlled watershed segmentation of one
microscopic image of a normal colon. Figure 5A is the mi-
croscopic image. The yellow contours in Fig. 5B are the
manually traced contours of each crypt. Figure 5C shows the
contours of crypts segmented automatically. Figure 6 shows
the results of the manual and marker-controlled watershed
segmentation on one OCT volume. Figure 6A shows the
volume rendering of the OCT image volume of a sample in-
cluding an aberrant crypt focus. Figure 6B shows the vol-
ume rendering overlaid with the manually segmented crypts,
while Fig. 6C shows the volume rendering overlaid with the
automatically segmented crypts. Only selected crypts were
manually segmented, and these were quantitatively compared
with the marker-controlled watershed segmentation of the
same selected crypts. Comparison between the manual and
automatic segmentation of this OCT image volume at various
AB
1mm 1mm
C D
Fig. 3 Segmentation results of representative NA sample: A the mi-
croscopic image with methylene blue and B corresponding OCT
volume of a sample of NA colonic tissue; C the crypt segmented
within the 2-D micrograph and D the crypts segmented in 3-D
within the OCT volume. The crypts touching the edges of the ROI
were removed. White arrows indicate intervening nonpitted mucosa.
Yellow arrows indicate missed crypts. Color online only.
A
B
1mm 1mm
C D
Fig. 4 Segmentation results of representative ACF sample: A the mi-
croscopic image with methylene blue and B corresponding OCT
volume of a sample of colonic tissue including an ACF; C the crypts
segmented within the 2-D micrograph and D the crypts segmented
in 3-D within the OCT volume. The crypts touching the edges of the
ROI were removed. White arrows indicate intervening nonpitted mu-
cosa. Yellow arrows indicate missed crypts. Color online only.
Qi et al.: Automated quantification of colonic crypt morphology
Journal of Biomedical Optics September/October 2008
Vol. 135054055-5
depths from the surface of tissue resulted in the PR and MR
values shown in Table 1.
Figure 7 shows the results of skeleton extraction from 3-D
segmented crypts using the conventional method, morpho-
logical thinning via distance transformation, and our centroid-
linking method. Using both methods, a skeleton of each seg-
mented crypt was extracted. Figures 7A and 7B
demonstrate the skeletons extracted from each crypt within a
3-D OCT volume of a representative sample of NA tissue
using the conventional method and our method, respectively.
Similarly, Figs. 7C and 7D demonstrate the skeletons ex-
tracted from each crypt within a 3D OCT volume of a sample
including an ACF. Skeletons of individual crypts can be
clearly observed in the figure insets eight times magnified.
Note that the unwanted branches on the skeletons extracted by
morphological thinning are not present on the skeletons ex-
tracted by centroid-linking. Also, the skeletons extracted by
centroid-linking do not include any false crypts shorter than
60
m. To compare the suitability of the skeletons extracted
by both methods for the purpose of quantifying the straight-
ness and parallelism of the segmented crypts, the metrics were
computed from a representative sample of NA tissue and a
sample including ACF. These results are summarized in Table
2. To evaluate the effect of noise reduction by nine-times
frame averaging and
3 3 median filtering on the measure-
ment of crypt orientation, straightness and parallelism of the
segmented crypts were computed from the same two tissue
samples. These results are summarized in Table 3.
The six morphological features were calculated within the
segmented ROI of the 14 normal colonic tissue samples, the
40 NA tissue samples, and the 4 ACF. The dominant value of
each feature in one sample was taken to be the peak of the
distribution of values within that sample. The first six col-
umns within Table 4 show the mean and standard deviation of
the dominant values of the six morphological features of all
samples of three types. The first two rows within Fig. 8 dis-
play the histograms of each morphological feature, separated
into the three sample types. Table 5 shows the correlation
coefficients for each pair of the six morphological features.
Because some features were correlated, the first two principal
components PCs of the six features were calculated. The
mean and standard deviations of the first two PCs are shown
in the last two columns within Table 4. Also, the histograms
of the first two PCs are displayed in the third row of Fig. 8.
4 Discussion and Conclusion
From our methylene blue-stained micrographs, samples of
normal mucosa and normal-appearing mucosa adjacent to
cancer showed predominantly round crypts distributed uni-
formly, while ACF stained dark and had crypts with large,
irregular lumens. Previous studies using high-magnification
chromoendoscopy have reported equivalent
observations.
11,60,61
From the micrographs, topology of the tis-
sue surface and crypt orientation are not readily apparent.
However, because it records 3-D image data with depth reso-
lution, OCT provides visualization of these features. This dif-
ference is illustrated clearly in Fig. 2, where the crypt open-
ings of the ACF can be clearly visualized in the micrograph
A, but the OCT cross section B and volume rendering C
demonstrate more clearly the topology of the slightly raised
ACF and the less-parallel orientation of the crypts. From the
OCT data, we observed that normal crypts were generally
very straight and parallel to each other. Crypts within ACF
were observed to be somewhat less straight and significantly
less parallel to each other compared with the orientation of
normal crypts. Crypts within moderately differentiated adeno-
carcinoma showed oblique orientation which was different
AC
B
1mm
Fig. 5 Results of the manual and marker-controlled watershed segmentation of one microscopic image of a normal colon: A the microscopic
image, B the yellow contours of crypts traced manually, and C the contours of crypts segmented automatically. Color online only.
C
A
B
1mm
Fig. 6 Results of manual and marker-controlled watershed segmentation of one OCT volume: A volume rendering of the OCT image volume of
a sample including an aberrant crypt focus, B the volume rendering overlaid with the manually segmented selected crypts, and C the volume
rendering overlaid with the automatically segmented crypts.
Qi et al.: Automated quantification of colonic crypt morphology
Journal of Biomedical Optics September/October 2008
Vol. 135054055-6
from the orientation of normal and ACF crypts, which were
more perpendicular to the tissue surface. Although there was
variability from sample to sample, generally contrast in the
OCT volumes was sufficient to visualize and segment crypts
to a depth of at least
600
m. However, we observed that
crypts were observable deeper in fresher samples, up to
1mm
in depth. This gives us confidence that measurement of crypt
orientation by OCT is feasible in vivo. Figure 2 displays OCT
en face images of an ACF at various depths down to
946
m
from the tissue surface. Two previous studies observed crypt
orientation without quantification from histopathology
slides, and our observations are consistent with those
findings.
18,25
It should be noted, however, that our observa-
tions were obtained from images of intact, unfixed tissue.
These observations confirm that the observed characteristics
of colonic crypts reported here are consistent with known
crypt morphology, suggesting that the analysis methods de-
scribed here will be applicable to high-magnification chro-
moendoscopy and endoscopic OCT images obtained in
patients.
Previously, several demonstrations of quantitative analysis
of colonic tissue have been reported from histological sec-
tions. Image-feature quantification methods based on texture
analysis,
24,50,62,63
fractals,
48
or frequency analysis
28
have been
shown, but these methods do not directly quantify colonic
crypts morphology. Mophometric methods of analyzing histo-
logical sections have also been demonstrated,
25,4649,63,64
in-
cluding automated segmentation of cellular and crypt
features.
47
However, the image properties of photographs of
histological sections are vastly different than those of chro-
moendoscopy and OCT, so those methods are not directly
applicable to chromoendoscopy and OCT data. A few demon-
strations of quantification of crypt morphology have been re-
ported using high-magnification chromoendoscopy,
11,6567
but
to our knowledge, automated segmentation of colonic crypts
from chromoendoscopic or OCT imaging has not been previ-
ously reported. Our methods reported here demonstrate the
feasibility of automatically quantifying crypt morphometry
from 2-D micrographs with methylene blue staining and from
3-D OCT volumes. This is significant, because these methods
are applicable to the high-potential endoscopic microscopy
techniques of high-magnification chromoendoscopy and
EOCT.
The marker-based watershed segmentation method pre-
sented here achieved accurate segmentation of colonic crypts
from 2-D micrographs with methylene blue stain as well as
Table 1 Comparison between the manuual and automic segmentation of the micrograph shown in Fig. 5 and the OCT image shown in Fig. 6 at
various depths from the surface of tissue.
Micrograph
mean±
standard
deviation
OCT volume at various depths from the tissue surface
m
mean±standard deviation
413 456 499 542 858 628 671 714
PR 0.69±
0.09
0.77±
0.10
0.73±
0.10
0.76±
0.07
0.67±
0.13
0.64±
0.15
0.63±
0.17
0.61±
0.20
0.59±
0.17
MR 0.73±
0.13
0.70±
0.10
0.71±
0.13
0.78±
0.08
0.60±
0.12
0.52±
0.17
0.51±
0.19
0.48±
0.22
0.47±
0.23
Table 2 Calculated straightness and parallelism values of a represen-
tative sample of NA tissue and a sample including ACF, from skeletons
extracted by morphological thinning via distance transformation and
by centroid-linking.
Morphological thinning Centroid-linking
Straightness Parallelism Straightness Parallelism
NA 0.88±0.08 0.91±0.25 0.94±0.04 0.97±0.05
ACF 0.83±0.03 0.79±0.24 0.82±0.05 0.75±0.15
B
A
DC
1mm
1mm
Fig. 7 Results of skeleton extraction from 3-D segmented crypts using
the conventional method, morphological thinning via distance trans-
formation, and our centroid-linking method. Skeletons extracted from
each crypt within a 3-D OCT volume of a representative sample of
NA using A the conventional method and B our method, respec-
tively. Skeletons extracted from each crypt within a 3-D OCT volume
of a sample including an ACF using C the conventional method and
D our method, respectively.
Qi et al.: Automated quantification of colonic crypt morphology
Journal of Biomedical Optics September/October 2008
Vol. 135054055-7
from 3-D OCT volumes. It can be observed in Figs. 3 and 4
that crypts were generally accurately segmented, few false
crypts were included, and few crypts were missed. To validate
the accuracy of automatic segmentation, we manually seg-
mented one representative microscopic image and one repre-
sentative OCT volume to compare with automatic segmenta-
tion. One representative image was considered adequate,
because the accuracy of the segmentation did not depend on
the shape of the crypts but on image contrast and resolution,
which did not vary with tissue type. To quantify segmentation
accuracy, we compared the manual and automated segmenta-
tion results using the PR and MR,
58
as summarized in Table 1.
Rejection of false crypts using a 3:1 ratio of major to minor
axes was generally successful; however, some of these con-
nective tissue artifacts having a ratio of
3:1 were re-
tained. For OCT data, these remaining false crypts were usu-
ally rejected during the centroid-linking process. Automated
segmentation of the micrographs with methylene blue
achieved a PR and MR of
0.70, which is considered accept-
able as indicating good agreement between the manual and
automatic segmentation.
58
Segmentation of the en face OCT
image also achieved high values of PR and MR. Best agree-
ment with manual segmentation was achieved at the plane of
OCT volume corresponding to the focus of the OCT scanner,
which is the location of best transverse image resolution
500
m below the surface of the tissue. This is indicated
by both the high mean values and low standard deviations of
PR and MR at this depth. Shallower than
500
m, the accu-
racy of the OCT segmentation was generally better than that
of the micrograph. Deeper than
500
m, however, the accu-
racy decreases steadily. From the results summarized in Table
1, we observe that the accuracy of the automated segmenta-
tion of OCT images is influenced by two major factors, the
transverse resolution, which peaks at the focal plane, and the
image SNR, which decreases monotonically from the tissue
surface into the tissue. We also observe that, at least near the
surface, segmentation of OCT slices result in better PR and
MR compared with segmentation of the corresponding crypts
in the micrograph. This is likely due to the less-distinct crypt
lumen borders within the micrographs as compared to the
OCT slices. Also, OCT images are unaffected by uneven
staining.
As shown in Fig. 7, the conventional skeleton extraction
method using 3-D morphological thinning via distance trans-
formation resulted in many branches and artifacts, which are
undesirable for quantifying orientation of crypts. The
centroid-linking skeleton extraction method demonstrated
here resulted in crypt skeletons without branches and also
rejected artifactual false crypts. Table 2 shows that centroid-
linking resulted in much narrower distributions of measured
straightness and parallelism of crypts, as well as a clearer
separation between normal and ACF tissues, as compared
with the measurements of the skeletons extracted by 3-D mor-
phological thinning. Table 3 shows that although noise reduc-
tion by nine times of frame averaging and
3 3 median fil-
tering slightly improves separation between NA and ACF
tissues from crypt orientation measures, the improvement is
not significant. This is because skeleton extraction by
centroid-linking does not strongly depend on contour rough-
ness of the segmented crypt boundaries. From this we are
encouraged that the method does not require frame averaging,
Table 3 Straightness S and parallelism P values of a representative sample of NA tissue and a sample including ACF from skeletons extracted
by centroid-linking using single frames and nine-times frame averaging, with and without 3 3 median filtering.
Without median filtering 3 3 With median filtering 3 3
S 1-frame S 9-frame P 1-frame P 9-frame S 1-frame S 9-frame P 1-frame P 9-frame
NA 0.94±
0.04
0.94±
0.02
0.96±
0.04
0.98±
0.02
0.94±
0.03
0.94±
0.04
0.98±
0.04
0.97±
0.05
ACF 0.84±
0.08
0.82±
0.05
0.79±
0.13
0.75±
0.15
0.85±
0.06
0.82±
0.05
0.77±
0.15
0.75±
0.15
Table 4 Mean and standard deviation of the dominant values of the morphological features and first two PCs of those six features within the
segmented ROI of the 14 normal colonic tissues, 40 NA colonic tissues, and 4 ACF within colonic tissues.
Area
m
2
Density
# /mm
2
Eccentricity Solidity Straightness Parallelism PC 1 PC 2
Normal
14
3786
±1991
58
±20
0.72
±0.07
0.9
±0.04
0.88
±0.05
0.82
±0.10
0.93
±0.74
−0.33
±1.14
NA
40
5033
±3294
46
±22
0.76
±0.07
0.87
±0.04
0.87
±0.07
0.8
±0.08
0.04
±1.28
0.17
±1.17
ACF
4
13166
±1634
18
±2
0.78
±0.02
0.81
±0.03
0.74
±0.04
0.65
±0.07
−3.67
±0.58
−0.53
±0.77
Qi et al.: Automated quantification of colonic crypt morphology
Journal of Biomedical Optics September/October 2008
Vol. 135054055-8
which reduces effective imaging rate, and is amenable to in
vivo EOCT application.
We have shown that the six morphological features of
crypts summarized in Table 4 can be automatically quantified
from micrographs with methelyne blue and/or OCT images.
Our measurements of the area and density of crypts within
normal tissue and NA tissues are in agreement with previous
studies.
26,49
To our knowledge, eccentricity, solidity, straight-
ness, and parallelism of crypts have not previously been quan-
tified. The purpose of this work is to demonstrate crypt mor-
phology quantification methods, not to prove that these
methods can be used to classify tissue types. However, we can
make some observations from the distributions of feature val-
ues represented by the histograms shown in Fig. 8. The dis-
tributions representing normal samples and NA samples ap-
pear generally very similar. However, NA may have broader
distributions, for example, of area, solidity, parallelism, and
eccentricity. The distribution of values representing ACF ap-
pear to have different means than normal/NA, except in the
case of eccentricity. The distributions of area, density,
straightness, and parallelism, and to a lesser extent solidity,
show strong potential for separation of ACF from normal/NA
tissues. The distributions of eccentricity for the three types of
tissue appear to be similar to each other and broad compared
to distributions of other parameters. This may be due to lack
of control for the obliqueness of the image slices compared to
the orientation of the crypts and the tissue surface topology. A
more accurate measure of crypt eccentricity could be achieved
by making use of 3-D crypt orientation and surface topology
information from the OCT data.
Some crypt morphology features were highly correlated. In
particular, the mean luminal area and density of crypts within
an ROI are highly correlated, with a correlation coefficient of
−0.65 as shown in Table 5. The straightness and parallelism
of crypts within an ROI are also correlated, with a correlation
coefficient of 0.49. It is intuitive and expected that these met-
rics should be correlated. Because of the correlation among
these features, principal-component analysis was applied to
these six features. The first principal component accounted for
56% variance of the six measured features. Together, the first
two principal components accounted for 62% variance. From
Fig. 8, we can observe that the distributions of the first PC for
normal and NA tissues appear similar, with the NA distribu-
tion possibly being broader. Furthermore, the distribution of
the first PC for ACF tissues appears to be significantly differ-
ent than normal and NA tissues. The distributions of the sec-
ond principal component do not appear to be particularly in-
formative. This is not surprising, as the second PC does not
account for much additional variance in the data. Because of
the small sample size of ACF four samples, no generalizable
conclusion may be drawn, but these observations indicate that
quantified morphological features of colonic crypts may have
the potential to aid tissue classification for screening for early
cancer biomarkers such as ACF. The observation that ACF
occurred within NA tissue but not within normal tissue is
consistent with the observation that the distributions of mor-
phological feature values for NA tissue are broader than dis-
tributions for normal tissue and are shifted toward the ACF
distributions. These observations taken together are consistent
with the so called field effect, which refers to the idea that
colon cancer arises in a wide field of mucosa that has previ-
ously undergone molecular changes.
60,68,69
These observations
suggest that subtle changes in crypt morphology may accom-
pany molecular changes and that tissues progressing toward
disease may be detectible by high-resolution endoscopic im-
aging.
The methods described here are likely to be useful for
analysis of data acquired in vivo with little modification. In
high-magnification chromoendoscopy, the resolution is simi-
lar to our benchtop images; however, the image quality is
expected to vary more due to less-consistent removal of mu-
1
5
0
0
0
1
2
0
0
0
9
0
0
0
6
0
0
0
3
0
0
0
16
8
0
1
0
0
8
0
6
0
4
0
2
0
16
8
0
0
.
8
8
0
.
8
0
0
.
7
2
0
.
6
4
0
.
5
6
10
5
0
0
.
9
6
0
.
9
0
0
.
8
4
0
.
7
8
16
8
0
0
.
9
6
0
.
9
2
0
.
8
8
0
.
8
4
0
.
8
0
0
.
7
6
0
.
7
2
8
4
0
1
.
0
0
.
9
0
.
8
0
.
7
0
.
6
10
5
0
1
.
6
0
.
0
-
1
.
6
-
3
.
2
10
5
0
2
.
4
1
.
2
0
.
0
-
1
.
2
-
2
.
4
10
5
0
Area
Frequency
Density E ccentricity
Solidity Straightnes s Parallelism
PC1 PC2
AC F
NA
Normal
Types
H
i
stograms
Fig. 8 Histogram of each morphological feature and the first two PCs
of those features within the segmented ROI of the 14 normal colonic
tissues, 40 NA colonic tissues, and 4 ACF within colonic tissues.
Table 5 Correlation coefficients between each morphological feature.
Area Density Eccentricity Solidity Straightness
Density −0.65
Eccentricity 0.07 −0.26
Solidity −0.42 0.35 −0.28
Straightness 0.31 0.24 0.04 0.15
Parallelism −0.34 0.24 0.02 0.03 0.49
Qi et al.: Automated quantification of colonic crypt morphology
Journal of Biomedical Optics September/October 2008
Vol. 135054055-9
cus and staining. From subjective experience, we are confi-
dent that the shapes of crypts observable in vivo are not dif-
ferent than we observed in the data presented here. Similarly,
image quality of endoscopic OCT data is expected to be less
consistent. However, the morphological features we analyzed
parallelism and straightness extracted from 3-D OCT vol-
umes are rotation invariant, so absolute tissue orientation or
gross orientation variations should not affect local orientation
quantification of crypts in vivo. For both technologies, auto-
mated image-quality feedback to the operator would help to
make the quality of images acquired in vivo more consistent.
Also, for both technologies, real-time analysis would be most
useful. We believe that this can be achieved by programming
the algorithms using parallel processing techniques.
In conclusion, we have shown that an automated image
analysis algorithm can quantify colonic crypt morphology
from high-resolution images approximating high-
magnification chromoendoscopy and EOCT. These computer-
ized algorithms can provide quantitative and objective mea-
sures, they are suitable for specific and repetitive image
reading tasks, and they can potentially reduce inter- and in-
traobserver variability when human readers evaluate such im-
ages. This work can enable studies to determine the clinical
utility of high-magnification chromoendoscopy and EOCT as
well as studies to evaluate colonic crypt morphology as a
biomarker for colonic disease progression.
Acknowledgments
The authors acknowledge the contributions of David L. Wil-
son, Ph.D., Theresa P. Pretlow, Ph.D., Gerard Isenberg, M.D.,
Jeffry Katz, M.D., Metini Janyasupab, Christine Lemyre,
Lateefa Russell, Wendi Barrett, and Sunny McClellan Morton.
This work was supported in part by the National Institutes of
Health Grants No. CA114276 and No. CA110943.
References
1. A. Jemal, R. Siegel, E. Ward, T. Murray, J. Xu, and M. J. Thun,
“Cancer statistics, 2007,” Ca-Cancer J. Clin. 57, 43–66 2007.
2. S. Jones, W. Chen, G. Parmigiani, F. Diehl, N. Beerenwinkel, T.
Antal, A. Traulsen, M. A. Nowak, C. Siegel, V. E. Velculescu, K. W.
Kinzler, B. Vogelstein, J. Willis, and S. D. Markowitz, “Comparative
lesion sequencing provides insights into tumor evolution,” Proc. Natl.
Acad. Sci. U.S.A. 10511, 4283–4288 2008.
3. S. J. Winawer, A. G. Zauber, M. N. Ho, M. J. O’Brien, L. S. Gottlieb,
S. S. Sternberg, J. D. Waye, M. Schapiro, J. H. Bond, J. F. Panish, F.
Ackroyd, M. Shike, R. C. Kurtz, L. Hornsby-Lewis, H. Gerdes, E. T.
Stewart, and T. N. P. S. Workgroup, “Prevention of colorectal cancer
by colonoscopic polypectomy,” N. Engl. J. Med. 32927, 1977–1981
1993.
4. F. Citarda, G. Tomaselli, R. Capocaccia, S. Barcherini, M. Crespi,
and T. I. M. S. Group, “Efficacy in standard clinical practice of
colonoscopic polypectomy in reducing colorectal cancer incidence,”
Gut 48, 812–815 2001.
5. S. Winawer, A. G. Zauber, M. N. Ho, M. J. O’Brien, L. S. Gottlieb,
S. S. Sternberg, J. D. Waye, M. Schapiro, J. H. Bond, J. F. Panish,
F. Ackroyd, M. Shike, R. C. Kurtz, L. Hornsby-Lewis, H. Gerdes,
and E. T. Stewart, “Prevention of colorectal cancer by colonoscopic
polypectomy.,” The National Polyp Study Workgroup, N. Engl. J.
Med. 32927, 1977–1981 1993.
6. R. L. Barclay, J. J. Vicari, A. S. Doughty, J. F. Johanson, and R. L.
Greenlaw, “Colonoscopic withdrawal times and adenoma detection
during screening colonoscopy,” N. Engl. J. Med. 35524, 2533–2541
2006.
7. R. M. Soetikno, T. K. Kaltenbach, R. V. Rouse, W. Park, A. Mahesh-
wari, T. Sato, S. Matsui, and S. Friedland, “Prevalence of nonpoly-
poid flat and depressed colorectal neoplasms in asymptomatic and
symptomatic adults,” JAMA, J. Am. Med. Assoc. 2999, 1027–1035
2008.
8. S. S. Poulsen, K. C. Christensen, M. Petri, and S. Jarnum, “Stereomi-
croscopic examination of stained rectal biopsies,” Scand. J. Gastro-
enterol. 13, 605–608 1978.
9. S. Tada, M. Iida, T. Yao, T. Matsumoto, M. Tsuneyoshi, and M.
Fujishima, “Stereomicroscopic examination of surface morphology in
colorectal epithelial tumors,” Hum. Pathol. 24, 1243–1252
1993.
10. S. Kudo, S. Tamura, T. Nakajima, H. Yamano, H. Kusaka, and H.
Watanabe, “Diagnosis of colorectal tumorous lesions by magnifying
endoscopy,” Gastrointest. Endosc. 44, 8–14 1996.
11. S. Kudo, C. A. Rubio, C. R. Teixeira, H. Kashida, and E. Kogure,
“Pit pattern in colorectal neoplasia: endoscopic magnifying view,”
Endoscopy 33, 367–373 2001.
12. A. Iso, S. Shimizu, M. Tada, and K. Kawai, “Usefulness of real-time
band enhancement in differential dianosis of colorectal tumors,” Dig.
Endosc. 7, 397–400 1995.
13. E. Jaramillo, M. Watanabe, R. Befrits, E. Ponce, C. Rubio, and P.
Slezak, “Small, flat colorectal neoplasms in long-standing ulcerative
colitis detected by high-resolution electronic video endoscope,” Gas-
trointest. Endosc. 44, 15–22 1996.
14. T. Matsumoto, F. Kuroki, M. Mizuno, S. Nakamura, and M. Iida,
“Application of magnifying chromoscopy for the assessment of se-
verity in patients with mild to moderate ulcerative colitis,” Gas-
trointest. Endosc. 46, 400–405 1997.
15. S. Y. Tung, C. S. Wu, and M. Y. Su, “Magnifying colonoscopy in
differentiating neoplastic from non-neoplastic colorectal lesions,”
Am. J. Gastroenterol. 96, 2628–2632 2001.
16. R. P. Bird, “Observation and quantification of aberrant crypts in the
murine colon treated with a colon carcinogen: Preliminary findings,”
Cancer Lett. 372, 147–151 1987.
17. L. Roncucci, D. Stamp, A. Meline, J. B. Cullen, and W. R. Burce,
“Identification and quantification of aberrant crypt foci and microad-
enomas in the human colon,” Hum. Pathol. 22, 287–294 1991.
18. C. D. Gregorio, L. Losi, R. Fante, S. Modica, M. Ghidoni, M.
Pedroni, M. G. Tamassia, L. Gafa, P. de Leon, and L. Roncucci,
“Histology of aberrant crypt foci in the human colon,” Histopathol-
ogy 304, 328–334 1997.
19. T. P. Pretlow, B. J. Barrow, W. S. Ashton, M. A. O’Riordan, T. G.
Pretlow, J. A. Jurcisek, and T. A. Stellato, “Aberrant crypts: putative
preneoplastic foci in human colonic mucosa,” Cancer Res. 51,1564
1567 1991.
20. L. Roncucci, S. Modica, M. Pedroni, M. G. Tamassia, M. Ghidoni, L.
Losi, R. Fante, C. Di Gregorio, A. Manenti, L. Gafa, and M. Ponz De
Leon, “Aberrant crypt foci in patients with colorectal cancer,” Br. J.
Cancer 77, 2343–2348 1998.
21. T. Takayama, S. Katsuki, Y. Takahashi, M. Ohi, S. Nojiri, S. Saka-
maki, J. Kato, K. Kogawa, H. Miyake, and Y. Niitsu, “Aberrant crypt
foci of the colon as precursors of adenoma and cancer,” N. Engl. J.
Med. 339, 1277–1284 1998.
22. E. McLellan, A. Medline, and R. P. Bird, “Sequential analyses of the
growth and morphological characteristics of aberrant crypt foci: pu-
tative preneoplastic lesions,” Cancer Res. 51, 5270–5274 1991.
23. E. McLellan and R. P. Bird, “Effect of disulfiram on 1,2-
dimethylhydrazine- and azoxymethane-induced aberrant crypt foci,”
Carcinogenesis 12, 969–972 1991.
24. A. N. Esgiar, R. N. G. Naguib, B. S. Sharif, M. K. Bennet, and A.
Murray, “Microscopic image analysis for quantitative measurement
and feature identification of normal and cancerous colonic mucosa,”
IEEE Trans. Inf. Technol. Biomed. 23, 197–203 1998.
25. H. Furukawa, S. Asai, T. Moriyasu, H. Ono, and T. Saiga, “Three-
dimensional reconstruction of the mucosa from sequential sections of
biopsy specimens for understanding diseasae progress in ulcerative
colitis,” Dig. Endosc. 12, 317–321 2000.
26. S. Tamura, Y. Furuya, T. Tadokoro, Y. Higashidani, Y. Yokoyama, K.
Araki, and S. Onishi, “Pit pattern and three-dimensional configura-
tion of isolated crypts from the patients with colorectal neoplasm,” J.
Gastroenterol. 3710, 798806 2002.
27. A. Ohta, K. Tominaga, and Y. Sakai, “Efficacy of magnifying
colonoscopy for the diagnosis of colorectal neoplasia: comparison
with histopathological findings,” Dig. Endosc. 16, 308–314 2004.
28. M. Haefner, L. Brunauer, H. Payer, R. Resch, F. Wrba, A. Gangl, A.
Vecsei, and A. Uhl, “Pit pattern classification of zoom-endoscopical
colon images using DCT and FFT,” in Proc. IEEE, 159–164 2007.
29. M. B. Kimmey, R. W. Martin, R. C. Gaggitt, K. Y. Wang, D. W.
Qi et al.: Automated quantification of colonic crypt morphology
Journal of Biomedical Optics September/October 2008
Vol. 135054055-10
Franklin, and F. E. Silverstein, “Histologic correlates of gastrointes-
tinal ultrasound images,” Gastroenterology 96, 433–441 1989.
30. D. P. Hurlstone, S. S. Cross, R. Slater, and S. Brown, “Detecting
diminutive colorectal lesions at colonoscopy: A randomised con-
trolled trial of pan-colonic versus targeted chromoscopy,” Gut 53,
376–380 2004.
31. K. Gono, T. Obi, M. Yamaguchi, N. Ohyama, H. Machida, Y. Sano,
S. Yoshida, Y. Hamamoto, and T. Endo, “Appearance of enhanced
tissue features in narrow-band endoscopic imaging,” J. Biomed. Opt.
9, 568–577 2004.
32. J. E. East, N. Suzuki, M. Stavrinidis, T. Guenther, H. J. W. Thomas,
and B. P. Saunders, “Narrow band imaging for colonoscopic surveil-
lance in hereditary non-polyposis colorectal cancer,” Gut 571,
65–70 2008.
33. J. E. East, N. Suzuki, A. von Herbay, and B. P. Saunders, “Narrow
band imaging with magnification for dysplasia detection and pit pat-
tern assessment in ulcerative colitis surveillance: A case with multiple
dysplasia associated lesions or masses,” Gut 5510, 1432–1435
2006.
34. M. V. J. Sivak, Gastroenterologic Endoscopy, W. B Saunders Com-
pany, New York, 2nd edition 2000.
35. D. P. Hurlstone, S. S. Cross, I. Adam, A. J. Shorthouse, S. Brown, D.
S. Sanders, and A. J. Lobo, “Efficacy of high magnification chromo-
scopic colonoscopy for the diagnosis of neoplasia in flat and de-
pressed lesions of the colorectum: A prospective analysis,” Gut 532,
284–290 2004.
36. H. Inoue, S. Kudo, and A. Shiokawa, “Technology insight: Laser-
scanning confocal microscopy and endocytoscopy for cellular obser-
vation of gastrointestinal tract,” Nat. Clin. Pract. Gastroenterol.
Hepatol. 21, 31–37 2005.
37. A. Polglase, W. McLaren, S. A. Skinner, R. Kiesslich, M. F. Neurath,
and P. Delaney, “A fluorescence confocal endomicroscope for in vivo
microscopy of the upper- and lower-GI tract,” Gastrointest. Endosc.
625, 686–695 2005.
38. R. Kiesslich, J. Burg, M. Vieth, J. Gnaendiger, M. Enders, P.
Delaney, A. Polglase, W. McLaren, D. Janell, S. Thomas, B. Nafe, P.
R. Galle, and M. F. Neurath, “Confocal laser endoscopy for diagnos-
ing intraepithelial neoplasias and colorectal cancer in vivo,” Gastro-
enterology 1273, 706–713 2004.
39. R. Kiesslich, A. Hoffman, A. Goetz, S. Biesterfeld, M. Vieth, P. R.
Galle, and M. F. Neurath, In vivo diagnosis of collagenous colitis by
confocal endomicoscopy,” Gut 55, 591–592 2006.
40. G. J. Tearny, M. E. Brezinski, B. E. Bouma, S. A. Boppart, C. Pitris,
J. F. Southern, and J. G. Fujimoto, In vivo endoscopic optical biopsy
with optical coherence tomography,” Science 276, 2037–2039
1997.
41. A. M. Rollins, R. Ung-arunyawee, A. Chak, R. C. K. Wong, K.
Kobayashi, M. V. Sivak, Jr., and J. A. Izatt, “Real-time in vivo-
imaging of human gastrointestinal ultrastructure by use of endoscopic
optical coherence tomography with a novel efficient interferometer
design,” Opt. Lett. 2419, 1358–1360 1999.
42. B. E. Bouma, G. J. Tearney, C. C. Compton, and N. S. Nishi, “High-
resolution imaging of the human esophagus and stomach in vivo us-
ing optical coherence tomography,” Gastrointest. Endosc. 514,
467–474 2000.
43. P. R. Pfau, M. V. Sivak, Jr., A. Chak, M. Kinnard, R. C. Wong, G. A.
Isenberg, J. A. Izatt, A. M. Rollins, and V. Westphal, “Criteria for the
diagnosis of dysplasia by endoscopic optical coherence tomography,”
Gastrointest. Endosc. 582, 196–202 2003.
44. D. C. Adler, Y. Chen, R. Huber, J. Schmitt, J. Connolly, and J. G.
Fujimoto, “Three-dimensional endomicroscopy using optical coher-
ence tomography,” Nat. Photonics 1, 709–716 2007.
45. S. H. Yun, G. J. Tearney, B. J. Vakoc, M. Shishkov, W. Y. Oh, A. E.
Desjardins, M. J. Suter, R. C. Chan, J. A. Evans, I. Jang, N. S.
Nishioka, J. F. de Boer, and B. E. Bouma, “Comprehensive volumet-
ric optical microscopy in vivo,” Nat. Med. 1212, 1429–1433
2006.
46. P. Schmitz-Moormann and G. W. Himmelmann, “Does quantitative
histology of rectal biopsy improve the differential diagnosis of
Crohn’s disease and ulcerative colitis in adults?,” Pathol. Res. Pract.
1834, 481–488 1988.
47. D. Thompson, P. H. Bartels, H. G. Bartels, P. W. Hamilton, and J. M.
Sloan, “Knowledge-guided segmentation of colorectal histopatho-
logic imagery,” Anal Quant Cytol. Histol. 154, 236–246 1993.
48. A. N. Esgiar, R. N. G. Naguib, B. S. Sharif, M. K. Bennet, and A.
Murray, “Fractal analysis in the detection of colonic cancer images,”
IEEE Trans. Inf. Technol. Biomed. 61, 54–58 2002.
49. P. W. Hamilton, D. C. Allen, P. C. H. Watt, C. C. Patterson, and J. D.
Biggart, “Classification of normal colorectal mucosa and adenocarci-
noma by morphometry,” Histopathology 11, 901–911 1987.
50. M. P. Tjoa and S. M. Krishnan, “Feature extraction for the analysis of
colon status from the endoscopic images,” Biomed. Eng. 29, 1–17
2003.
51. Z. Hu and A. M. Rollins, “Quasi-telecentric optical design of a
microscope-compatible OCT scanner,” Opt. Express 1317, 6407–
6415 2005.
52. R. C. Gonzalez and R. E. Woods, “Digital image processing,”
Addison-Wesley, New York, 1st edition 1992.
53. R. C. Gonzalez and R. E. Woods, “Digital image processing,” Pren-
tice Hall, New York, 3rd edition 2007.
54. N. Otsu, “A threshold selection method from gray-level histograms,”
IEEE Trans. Syst. Man Cybern. SMC-91, 62–66 1979.
55. L. Vincent, “Morphological grayscale reconstruction in image analy-
sis: applications and efficient algorithms,” IEEE Trans. Image Pro-
cess. 2
2, 176–201 1993.
56. P. Salembier, “Morphological multiscale segmentation for image cod-
ing,” Signal Process. 38, 359–386 1994.
57. L. Vincent and S. Pierre, “Watersheds in digital spaces: an efficient
algorithm based on immersion simulations,” IEEE Trans. Pattern
Anal. Mach. Intell. 136, 583–598 1991.
58. Y. Huang and D. Chen, “Watershed segmentation for breast tumor in
2-D sonography,” Ultrasound Med. Biol. 305, 625–632 2004.
59. D. Reniers and A. Telea, “Skeleton-based hierarchical shape segmen-
tation,” IEEE Int. Conf. on Shape Modeling and Applications
SMI’07, 76952815, 2007.
60. R. E. Rudolph, J. A. Dominitz, J. W. Lampe, L. Levy, P. Qu, S. S. Li,
P. D. Lampe, M. P. Bronner, and J. D. Potter, “Risk factors for col-
orectal cancer in relation to number and size of aberrant crypt foci in
humans,” Cancer Epidemiol. Biomarkers Prev. 143,605608
2005.
61. K. Seike, K. Koda, K. Odea, C. Kosugi, K. Shimizu, M. Nishimura,
M. Shioiri, S. Takano, H. Ishikura, and M. Miyazaki, “Assessment of
rectal aberrant crypt foci by standard chromoscopy and its predictive
value for colonic advanced neoplasms,” Aust. Occup. Ther. J. 1016,
1362–1369 2006.
62. A. N. Esgiar, R. N. G. Naguib, B. S. Sharif, M. K. Bennet, and A.
Murray, “Automated feature extraction and identification of colon
carcinoma,” Anal Quant Cytol. Histol. 20, 297–301 1998.
63. M. Haefner, M. Liedlgruber, A. Puespoek, S. Thomas, R. Schoefl, F.
Wrba, A. Gangl, and A. Uhl, “Computer-assisted pit pattern analysis
of colonic lesions,” Gastrointest. Endosc. 635, AB247 2006.
64. H. Yaegashi, Y. Zhang, F. Tezuka, T. Takahashi, and M. Fukumoto,
“Computer-assisted 3D mapping and morphometry of dysplastic
zones in endoscopically resected colonic adenomas,” J. Pathol. 191,
143–149 2000.
65. Q. Huang, N. Fukami, H. Kashida, T. Takeuchi, E. Kogure, T. Kura-
hashi, E. Stahl, Y. Kudo, H. Kimata, and S. Kudo, “Interobserver and
intra observer consistency in the endoscopic assessment of colonic pit
patterns,” Gastrointest. Endosc. 604, 520–526 2004.
66. D. P. Hurlstone, S. S. Cross, I. Adam, A. J. Shorthouse, S. Brown, D.
S. Sander, and A. J. Lobo, “Endoscopic morphological anticipation of
submucosal invasion in flat and depressed colorectal lesions: Clinical
implications and subtype analysis of the kudo type V pit pattern using
high-magnification-chromoscopic colonoscopy,” Int. J. Colorectal
Dis. 65, 369–375 2004.
67. D. P. Hurlstone, M. Karajeh, D. S. Sanders, S. K. Drew, and S. S.
Cross, “Rectal aberrant crypt foci indentification using high-
magnification-chromoscopic colonoscopy: Biomarkers for flat and
depressed neoplasia,” Am. J. Gastroenterol. 100,172005.
68. A. K. Gupta, T. P. Pretlow, and R. E. Schoen, “Aberrant crypt foci:
What we know and what we need to know,” Nat. Clin. Pract. Gas-
troenterol. Hepatol. 55, 526–533 2007.
69. H. K. Roy, Y. Liu, R. K. Wali, Y. L. Kim, A. K. Kromine, M. J.
Goldberg, and V. Backman, “Four-dimensional elastic light-scattering
fingerprints as preneoplastic markers in the rat model of colon car-
cinogenesis,” Gastroenterology 1264, 1071–1081 2004.
Qi et al.: Automated quantification of colonic crypt morphology
Journal of Biomedical Optics September/October 2008
Vol. 135054055-11
    • "Colorectal cancer is the third most common form of cancer, and the second leading cause of cancer death in the United States [25]. OCT has strong translational potential for endoscopic imaging for identification of colonic pathologies262728 For FMI, besides the cathepsin-based contrast agents [23], Roney et al. recently demonstrated that the legume lectin Ulex europaeus agglutinin I (UEA-1) binds the surfaces of adenomatous polyps in specimens of colorectal cancer from the APC Min mouse model. The carbohydrate α-L-fucose, which is over-expressed on the surfaces of polyps, facilitates the bond with UEA-1. "
    [Show abstract] [Hide abstract] ABSTRACT: Optical coherence tomography (OCT) provides high-resolution, cross-sectional imaging of tissue microstructure in situ and in real time, while fluorescence molecular imaging (FMI) enables the visualization of basic molecular processes. There is a great deal of interest in combining these two modalities so that the tissue's structural and molecular information can be obtained simultaneously. This could greatly benefit biomedical applications such as detecting early diseases and monitoring therapeutic interventions. In this research, an optical system that combines OCT and FMI was developed. The system demonstrated that it could co-register en face OCT and FMI images with a 2.4 x 2.4 mm(2) field-of-view. The transverse resolutions of OCT and FMI of the system are both approximately 10 microm. Capillary tubes filled with fluorescent dye Cy 5.5 in different concentrations under a scattering medium are used as the phantom. En face OCT images of the phantoms were obtained and successfully co-registered with FMI images that were acquired simultaneously. A linear relationship between FMI intensity and dye concentration was observed. The relationship between FMI intensity and target fluorescence tube depth measured by OCT images was also observed and compared with theoretical modeling. This relationship could help in correcting reconstructed dye concentration. Imaging of colon polyps of the APC(min) mouse model is presented as an example of biological applications of this co-registered OCT/FMI system.
    Full-text · Article · Jan 2010
    • "For many diseases, changes in morphology may provide important information regarding disease outcome [1] [2]. Difficulties in quantifying differences in morphology of anatomical organs has limited the systematic use of these features for distinguishing between patients with different disease outcome. "
    [Show abstract] [Hide abstract] ABSTRACT: Three-dimensional (3D) morphometric features of anatomical objects may provide important information regarding disease outcome. In this paper we develop an integrated framework to quantitatively extract and analyze 3D surface morphology of anatomical organs. We consider two datasets: (a) synthetic dataset comprising 640 super quadratic ellipsoids, and (b) clinical dataset comprising 36 prostate MRI studies. Volumetric interpolation and shape model construction were employed to find a concise 3D representation of objects. For the clinical data, a total of 630 pairwise registrations and shape distance computations were performed between each of 36 prostate studies. Graph embedding was used to visualize subtle differences in 3D morphology by non-linearly projecting the shape parameters onto a reduced dimensional manifold. The medial axis shape model used to represent the shape of super quadratic ellipsoids was found to have a large Pearson's correlation coefficient R<sup>2</sup> = 0.805 with known shape parameters. For the prostate gland datasets, spherical glands were found to aggregate at one end of the manifold and elliptical glands were found to aggregate at the other extrema of the manifold. Our results suggest our framework might discriminate between objects with subtle morphometric differences.
    Full-text · Conference Paper · Jan 2010
    • "Since then imaging depth did only increase marginally, resolution was improved by less than a factor of ten, but imaging speed was boosted by more than three orders of magnitude, from less than 100 to more than 300,000 A-scans per second [2]. This enables not only the scan of larger tissue surfaces like esophagus [3], colon [4][5] or vessel [6], but also opens new application beyond diagnosis. A non-contact volumetric imaging with less than 15μm resolution, which is not possible by ultrasound or any other medical imaging modality, can guide microsurgery at the eye [7][8], in ENT [9] and in other medical disciplines [10]. "
    [Show abstract] [Hide abstract] ABSTRACT: p>Modern measurement equipment delivers more detailed data and faster data with each generation. These data can be used for different applications, one of them is doing real time display. Instead of saving all data during the measurements and analyze it afterwards, the data is displayed in real time and only especially selected parts of the data are saved for further work. Moving the screening part of the analysis to the human brain and pattern recognition avoids the saving of vast amounts of data and massive calculation power on computers afterwards and it dramatically improves the level of interaction with the measurement systems. This work starts first with a short look to the question what OCT is and how data is acquired. The different possibilities of volume rendering are presented in their basic ideas. Graphics hardware and algorithms are presented and discussed. Last the results of measurements taken by the system will be presented and discussed.</p
    Full-text · Article · Jun 2009
Show more