Browsing by Subject "Medical image analysis."

Now showing 1 - 5 of 5

Open Access
Histopathological image classification using salient point patterns
(2011) Çığır, Celal
Over the last decade, computer aided diagnosis (CAD) systems have gained great importance to help pathologists improve the interpretation of histopathological tissue images for cancer detection. These systems offer valuable opportunities to reduce and eliminate the inter- and intra-observer variations in diagnosis, which is very common in the current practice of histopathological examination. Many studies have been dedicated to develop such systems for cancer diagnosis and grading, especially based on textural and structural tissue image analysis. Although the recent textural and structural approaches yield promising results for different types of tissues, they are still unable to make use of the potential biological information carried by different tissue components. However, these tissue components help better represent a tissue, and hence, they help better quantify the tissue changes caused by cancer. This thesis introduces a new textural approach, called Salient Point Patterns (SPP), for the utilization of tissue components in order to represent colon biopsy images. This textural approach first defines a set of salient points that correspond to nuclear, stromal, and luminal components of a colon tissue. Then, it extracts some features around these salient points to quantify the images. Finally, it classifies the tissue samples by using the extracted features. Working with 3236 colon biopsy samples that are taken from 258 different patients, our experiments demonstrate that Salient Point Patterns approach improves the classification accuracy, compared to its counterparts, which do not make use of tissue components in defining their texture descriptors. These experiments also show that different set of features can be used within the SPP approach for better representation of a tissue image.
Open Access
Local object patterns for tissue image representation and cancer classification
(2013) Olgun, Gülden
Histopathological examination of a tissue is the routine practice for diagnosis and grading of cancer. However, this examination is subjective since it requires visual interpretation of a pathologist, which mainly depends on his/her experience and expertise. In order to minimize the subjectivity level, it has been proposed to use automated cancer diagnosis and grading systems that represent a tissue image with quantitative features and use these features for classifying and grading the tissue. In this thesis, we present a new approach for effective representation and classification of histopathological tissue images. In this approach, we propose to decompose a tissue image into its histological components and introduce a set of new texture descriptors, which we call local object patterns, on these components to model their composition within a tissue. We define these descriptors using the idea of local binary patterns. However, we define our local object pattern descriptors at the component-level to quantify a component, as opposed to pixel-level local binary patterns, which quantify a pixel by constructing a binary string based on relative intensities of its neighbors. To this end, we specify neighborhoods with different locality ranges and encode spatial arrangements of the components within the specified local neighborhoods by generating strings. We then extract our texture descriptors from these strings to characterize histological components and construct the bag-of-words representation of an image from the characterized components. In this thesis, we use two approaches for the selection of the components: The first approach uses all components to construct a bag-ofwords representation whereas the second one uses graph walking to select multiple subsets of the components and constructs multiple bag-of-words representations from these subsets. Working with microscopic images of histopathological colon tissues, our experiments show that the proposed component-level texture descriptors lead to higher classification accuracies than the previous textural approaches.
Open Access
Multilevel cluster ensembling for histopathological image segmentation
(2011) Şimşek, Ahmet Çağrı
In cancer diagnosis and grading, histopathological examination of tissues by pathologists is accepted as the gold standard. However, this procedure has observer variability and leads to subjectivity in diagnosis. In order to overcome such problems, computational methods which use quantitative measures are proposed. These methods extract mathematical features from tissue images assuming they are composed of homogeneous regions and classify images. This assumption is not always true and segmentation of images before classification is necessary. There are methods to segment images but most of them are proposed for generic images and work on the pixel-level. Recently few algorithms incorporated medical background knowledge into segmentation. Their high level feature definitions are very promising. However, in the segmentation step, they use region growing approaches which are not very stable and may lead to local optima. In this thesis, we present an efficient and stable method for the segmentation of histopathological images which produces high quality results. We use existing high level feature definitions to segment tissue images. Our segmentation method significantly improves the segmentation accuracy and stability, compared to existing methods which use the same feature definition. We tackle image segmentation problem as a clustering problem. To improve the quality and the stability of the clustering results, we combine different clustering solutions. This approach is also known as cluster ensembles. We formulate the clustering problem as a graph partitioning problem. In order to obtain diverse and high quality clustering results quickly, we made modifications and improvements on the well-known multilevel graph partitioning scheme. Our method clusters medically meaningful components in tissue images into regions and obtains the final segmentation. Experiments showed that our multilevel cluster ensembling approach performed significantly better than existing segmentation algorithms used for generic and tissue images. Although most of the images used in experiments, contain noise and artifacts, the proposed algorithm produced high quality results.
Open Access
Object-oriented testure analysis and unsupervised segmentation for histopathological images
(2012) Tosun, Akif Burak
The histopathological examination of tissue specimens is essential for cancer diagnosis and grading. However, this examination is subject to a considerable amount of observer variability as it mainly relies on visual interpretation of pathologists. To alleviate this problem, it is very important to develop computational quantitative tools, for which image segmentation constitutes the core step. The segmentation algorithms in literature commonly use pixel-level color/texture descriptors that they define on image pixels for quantizing a tissue. On the other hand, it is usually harder to express domain specific knowledge about tissues, such as the spatial organization of tissue components, using only the pixel-level descriptors. This may become even harder for tissue images, which typically consist of a considerable amount of variation and noise at their pixel-level, such as similar color distribution of different tissue components, distortion in cell alignments, and color contrast caused by too much stain in a particular region. The previous segmentation algorithms are more susceptible to these problems as they work on pixel-level descriptors. In order to successfully address these issues, in this thesis, we introduce three new texture descriptors, namely ObjSEG, GraphRLM, and ObjCooc textures, and implement algorithms that use these descriptors for segmenting histopathological tissue images. We extract these texture descriptors on tissue components that are approximately represented by circular objects. Since these objectoriented texture descriptors are defined on the tissue components, and hence domain specific knowledge, they represent the spatial organization of the components better than their previous counterparts. Thus, our algorithms based on these descriptors give more effective and robust segmentation results. Furthermore, since the descriptors are not directly defined on image pixels, they are effective to alleviate the pixel-level problems. In our experiments, we tested our algorithms that use the proposed objectoriented descriptors on a dataset of 200 colon tissue images. Our experiments demonstrated that our new object-oriented feature descriptors led to high segmentation accuracies, also providing a reasonable number of segmented regions. Compared with its previous counterparts, the experimental results also showed that our proposed algorithms are more effective in segmenting histopathological images.
Open Access
Resampling-based Markovian modeling for automated cancer diagnosis
(2011) Özdemir, Erdem
Correct diagnosis and grading of cancer is very crucial for planning an effective treatment. However, cancer diagnosis on biopsy images involves visual interpretation of a pathologist, which is highly subjective. This subjectivity may, however, lead to selecting suboptimal treatment plans. In order to circumvent this problem, it has been proposed to use automatic diagnosis and grading systems that help decrease the subjectivity levels by providing quantitative measures. However, one major challenge for designing these systems is the existence of high variance observed in the biopsy images due to the nature of biopsies. Thus, for successful classifications of unseen images, these systems should be trained with a large number of labeled images. However, most of the training sets in this domain have limited size of labeled data since it is quite difficult to collect and label histopathological images. In this thesis, we successfully address this issue by presenting a new resampling framework. This framework relies on increasing the generalization capacity of a classifier by augmenting the size and variation in the training set. To this end, we generate multiple sequences from an image, each of which corresponds to a perturbed sample of the image. Each perturbed sample characterizes different parts of the image, and hence, they are slightly different from each other. The use of these perturbed samples for representing the image increases the size and variability of the training set. These samples are modeled with Markov processes which are used to classify unseen image. Working with histopathological tissue images, our experiments demonstrate that the proposed framework is more effective for both larger and smaller training sets compared against other approaches. Additionally, they show that the use of perturbed samples is effective in a voting scheme which boosts the performance of the classifier.