Marina E. Plissiti

Automated Detection of Cell Nuclei in Pap Smear Images Using Morphological Reconstruction and Clustering

In this paper, we present a fully automated method for cell nuclei detection in Pap smear images. The locations of the candidate nuclei centroids in the image are detected with morphological analysis and they are refined in a second step, which incorporates a priori knowledge about the circumference of each nucleus. The elimination of the undesirable artifacts is achieved in two steps: the application of a distance-dependent rule on the resulted centroids; and the application of classification algorithms. In our method, we have examined the performance of an unsupervised (fuzzy C-means) and a supervised (support vector machines) classification technique. In both classification techniques, the effect of the refinement step improves the performance of the clustering algorithm. The proposed method was evaluated using 38 cytological images of conventional Pap smears containing 5617 recognized squamous epithelial cells. The results are very promising, even in the case of images with high degree of cell overlapping.

Combining shape, texture and intensity features for cell nuclei extraction in Pap smear images

In this work, we present an automated method for the detection and boundary determination of cells nuclei in conventional Pap stained cervical smear images. The detection of the candidate nuclei areas is based on a morphological image reconstruction process and the segmentation of the nuclei boundaries is accomplished with the application of the watershed transform in the morphological color gradient image, using the nuclei markers extracted in the detection step. For the elimination of false positive findings, salient features characterizing the shape, the texture and the image intensity are extracted from the candidate nuclei regions and a classification step is performed to determine the true nuclei. We have examined the performance of two unsupervised (K-means, spectral clustering) and a supervised (Support Vector Machines, SVM) classification technique, employing discriminative features which were selected with a feature selection scheme based on the minimal-Redundancy-Maximal-Relevance criterion. The proposed method was evaluated on a data set of 90 Pap smear images containing 10,248 recognized cell nuclei. Comparisons with the segmentation results of a gradient vector flow deformable (GVF) model and a region based active contour model (ACM) are performed, which indicate that the proposed method produces more accurate nuclei boundaries that are closer to the ground truth.

An automated method for lumen and media-adventitia border detection in a sequence of IVUS frames

In this paper, we present a method for the automated detection of lumen and media-adventitia border in sequential intravascular ultrasound (IVUS) frames. The method is based on the use of deformable models. The energy function is appropriately modified and minimized using a Hopfield neural network. Proper modifications in the definition of the bias of the neurons have been introduced to incorporate image characteristics. A simulated annealing scheme is included to ensure convergence at a global minimum. The method overcomes distortions in the expected image pattern, due to the presence of calcium, employing a specialized structure of the neural network and boundary correction schemas which are based on a priori knowledge about the vessel geometry. The proposed method is evaluated using sequences of IVUS frames from 18 arterial segments, some of them indicating calcified regions. The obtained results demonstrate that our method is statistically accurate, reproducible, and capable to identify the regions of interest in sequences of IVUS frames.

Overlapping Cell Nuclei Segmentation Using a Spatially Adaptive Active Physical Model

A method for the segmentation of overlapping nuclei is presented, which combines local characteristics of the nuclei boundary and a priori knowledge about the expected shape of the nuclei. A deformable model whose behavior is driven by physical principles is trained on images containing a single nuclei, and attributes of the shapes of the nuclei are expressed in terms of modal analysis. Based on the estimated modal distribution and driven by the image characteristics, we develop a framework to detect and describe the unknown nuclei boundaries in images containing two overlapping nuclei. The problem of the estimation of an accurate nucleus boundary in the overlapping areas is successfully addressed with the use of appropriate weight parameters that control the contribution of the image force in the total energy of the deformable model. The proposed method was evaluated using 152 images of conventional Pap smears, each containing two overlapping nuclei. Comparisons with other segmentation methods indicate that our method produces more accurate nuclei boundaries which are closer to the ground truth.

Watershed-based segmentation of cell nuclei boundaries in Pap smear images

In this work we present a fully automated method for the accurate detection of cell nuclei boundaries in conventional Pap smear images, based on the watershed transform. For the extraction of nuclei and cytoplasm markers, which are used as starting points for the flooding process, a morphological reconstruction step is initially performed in each image. The watershed transform is then applied in the color morphological gradient image, which shows the boundaries of the more pronounced nuclei. For the elimination of false positive findings, salient features of shape and intensity of the detected regions were calculated and a clustering step is performed. The method was evaluated with a data set of 19 images containing 3616 recognized cells nuclei. The performance of the method was evaluated in terms of the correct detection of the positions of the nuclei. Comparisons with the segmentation results of the gradient vector flow (GVF) deformable model showed that the segmentation of the watershed transform captures more accurately the boundaries of nuclei, leading to a better performance of the clustering algorithm.

Cervical cell classification based exclusively on nucleus features

In this work, we present a framework for the efficient classification of cervical cells in normal and abnormal categories, based on features extracted exclusively from the nucleus area and ignoring the contingent cytoplasm features. This task is very important, since the nuclei are the only distinguishable areas in complex Pap smear images, as these images present a high degree of cell overlapping and the exact borders of the cytoplasm areas are ambiguous. We have examined the ability of non-linear dimensionality reduction schemes to produce accurate representation of the features manifold, along with the definition of an efficient feature subset, and their influence on the classification performance. Two unsupervised classifiers were used and the results indicate that we can achieve high classification performance when only the nuclei features are used.

A Review of Automated Techniques for Cervical Cell Image Analysis and Classification

Cervical smear screening is the most popular method used for the detection of cervical cancer in its early stages. The most eminent screening test is the Pap smear, which is based on the staining of cervical cells, using the technique that was first introduced by George Papanicolaou (Science 1942). With this screening technique, precancerous conditions and abnormal changes in cells that may develop into cancer are recognized. The widespread use of this test in developed countries has significantly reduced the incidence and mortality of invasive cervical cancer. In the last years, many methods have been appeared in the literature, which aim at the automated determination of the cytoplasm and the nucleus in these images. In this context, sophisticated image processing techniques and feature extraction and classification methods have been developed by several researchers, in order to derive useful conclusions for the characterization of the contents of the Pap smear images. In this work, an overview of the published techniques related to cervical smear screening is presented, in order to provide an integrated essay of the state of the art methods in the specific scientific field. Special focus has been paid on two main concepts with great research interest: the cell image segmentation and the classification techniques proposed for the characterization Pap smear images.

Automated Detection of Cell Nuclei in PAP stained cervical smear images using Fuzzy Clustering

In this work we present an automated method for cell nuclei detection in PAP stained cervical smear images. The method is based on the detection of regional minima in the image, followed by a two phase clustering of the detected centroids. An empirical rule and the fuzzy C-means clustering algorithm are applied on the resulted centroids in order to reduce false positive findings. The number of classes in which the nuclei are classified is determined automatically for the dataset that is used. The proposed method is evaluated using cytological images of conventional PAP stained cervical smears, which contain 3085 recognized squamous epithelial cells.

Inflammatory Cell Extraction and Nuclei Detection in Pap Smear Images

The automated diagnosis of cervical cancer in Pap smear images is a difficult though extremely important procedure. In order to obtain reliable diagnostic information, the nuclei and their characteristics must be correctly identified and evaluated. However, the presence of inflammatory and overlapping cells in these images complicates the detection process. In this work, a segmentation algorithm is developed to extract the inflammatory cells and enable accurate nuclei detection. The proposed algorithm is based on the combination of gray level thresholding and the definition of a distance rule, which entails in the identification of inflammatory cells. The results indicate that our method significantly simplifies the nuclei detection process, as it reduces the number of inflammatory cells that may interfere.

Three-dimensional coronary artery reconstruction using fusion of intravascular ultrasound and biplane angiography

Abstract We have developed an efficient method for 3D reconstruction of coronary arteries. Our approach is based on the fusion of intravascular ultrasound (IVUS) and biplane angiography. The method includes an efficient algorithm for the automatic identification of the regions of interest in IVUS images and a novel methodology for the extraction of the catheter path from biplane angiographies. The estimation of IVUS frames relative twist and the computation of first IVUS frame absolute orientation is also investigated. To assess the performance of the method, a validation procedure is introduced. Several metrics are obtained to verify the reliability of our method in the description of coronary artery morphology.