Early detection of cervical cancer can be achieved through visual analysis of cell anomalies. The established
PAP smear achieves a sensitivity of 50-90%, most false negative results are caused by mistakes in the preparation
of the specimen or reader variability in the subjective, visual investigation. Since cervical cancer is caused by
human papillomavirus (HPV), the detection of HPV-infected cells opens new perspectives for screening of precancerous
abnormalities. Immunocytochemical preparation marks HPV-positive cells in brush smears of the
cervix with high sensitivity and specificity.
The goal of this work is the automated detection of all marker-positive cells in microscopic images of a
sample slide stained with an immunocytochemical marker. A color separation technique is used to estimate the
concentrations of the immunocytochemical marker stain as well as of the counterstain used to color the nuclei.
Segmentation methods based on Otsu's threshold selection method and Mean Shift are adapted to the task of
segmenting marker-positive cells and their nuclei.
The best detection performance of single marker-positive cells was achieved with the adapted thresholding
method with a sensitivity of 95.9%. The contours differed by a modified Hausdorff Distance (MHD) of 2.8 μm.
Nuclei of single marker positive cells were detected with a sensitivity of 95.9% and MHD = 1.02 μ;m.
DNA Image Cytometry is a method for non-invasive cancer diagnosis which measures the DNA content of
Feulgen-stained nuclei. DNA content is measured using a microscope system equipped with a digital camera as
a densitometer and estimating the DNA content from the absorption of light when passing through the nuclei.
However, a DNA Image Cytometry measurement is only valid if each nucleus is only measured once.
To assist the user in preventing multiple measurements of the same nucleus, we have developed a unique
digital identifier for the characterization of Feulgen-stained nuclei, the so called Nucleus Fingerprint. Only nuclei
with a new fingerprint can be added to the measurement. This fingerprint is based on basic nucleus features,
the contour of the nucleus and the spatial relationship to nuclei in the vicinity. Based on this characterization,
a classifier for testing two nuclei for identity is presented.
In a pairwise comparison of ≈40000 pairs of mutually different nuclei, 99.5% were classified as different. In
another 450 tests, the fingerprints of the same nucleus recorded a second time were in all cases judged identical.
We therefore conclude that our Nucleus Fingerprint approach robustly prevents the repeated measurement of
nuclei in DNA Image Cytometry.
KEYWORDS: Cameras, High dynamic range imaging, Sensors, Contrast transfer function, Bandpass filters, Signal to noise ratio, Multispectral imaging, RGB color model, Optical filters, Light sources
Capturing natural scenes with high dynamic range content using conventional RGB cameras generally results
in saturated and underexposed and therefore compromising image areas. Furthermore the image lacks color
accuracy due to a systematic color error of the RGB color filters. The problem of the limited dynamic range
of the camera has been addressed by high dynamic range imaging1, 2 (HDRI): Several RGB images of different
exposures are combined into one image with greater dynamic range. Color accuracy on the other hand can be
greatly improved using multispectral cameras,3 which more accurately sample the electromagnetic spectrum.
We present a promising combination of both technologies, a high dynamic range multispectral camera featuring
a higher color accuracy, an improved signal to noise ratio and greater dynamic range compared to a similar low
dynamic range camera.
Cytopathological cancer diagnoses can be obtained less invasive than histopathological investigations. Cells
containing specimens can be obtained without pain or discomfort, bloody biopsies are avoided, and the diagnosis
can, in some cases, even be made earlier. Since no tissue biopsies are necessary these methods can also be used
in screening applications, e.g., for cervical cancer. Among the cytopathological methods a diagnosis based on
the analysis of the amount of DNA in individual cells achieves high sensitivity and specificity. Yet this analysis
is time consuming, which is prohibitive for a screening application. Hence, it will be advantageous to retain, by
a preceding selection step, only a subset of suspicious specimens. This can be achieved using highly sensitive
immunocytochemical markers like p16ink4a for preselection of suspicious cells and specimens.
We present a method to fully automatically acquire images at distinct positions at cytological specimens
using a conventional computer controlled microscope and an autofocus algorithm. Based on the thus obtained
images we automatically detect p16ink4a-positive objects. This detection in turn is based on an analysis of the
color distribution of the p16ink4a marker in the Lab-colorspace. A Gaussian-mixture-model is used to describe
this distribution and the method described in this paper so far achieves a sensitivity of up to 90%.
Compared to histopathological methods cancer can be detected earlier, specimens can be obtained easier and
with less discomfort for the patient by cytopathological methods. Their downside is the time needed by an expert
to find and select the cells to be analyzed on a specimen. To increase the use of cytopathological diagnostics,
the cytopathologist has to be supported in this task.
DNA image cytometry (DNA-ICM) is one important cytopathological method that measures the DNA content
of cells based on the absorption of light within Feulgen stained cells. The decision whether or not the patient has
cancer is based on the histogram of the DNA values. To support the cytopathologist it is desirable to replace
manual screening of the specimens by an automatic selection of relevant cells for DNA-ICM. This includes
automated acquisition and segmentation of focused cells, a recognition of cell types, and a selection of cells to
be measured. As a step towards automated cell type detection we show the discrimination of cell types in serous
effusions on a selection of about 3, 100 manually classified cells. We present a set of 112 features and the results
of feature selection with ranking and a floating-search method combined with different objective functions. The
validation of the best feature sets with a k-nearest neighbor and a fuzzy k-nearest neighbor classifier on a disjoint
set of cells resulted in classification rates of 96% for lymphocytes and 96.8% for the diagnostically relevant cells
(mesothelial+ cells), which includes benign and malign mesothelial cells and metastatic cancer cells.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.