Igor B. Gurevich

Comparative Analysis and Classification of Features for Image Models

This study has been conducted in the framework of developing one of the directions of descriptive approach to image analysis and recognition, and it is devoted to one of the main tools of this approach, namely, the use of formal image models in solving recognition problems. We systematized the image features widely used in solving applied problems of image analysis and recognition. It is well known that the mathematical nature and functional meaning of these features, as well as computational and measurement methods employed, are extremely various. The main results are the following: different approaches to the classification of image features are introduced, comparative analysis of them is performed, and the aspect of descriptivity is realized by numerous examples of the considered classifications being filled in by features (descriptors). On the basis of these results, certain recommendations and criteria for choosing the features in applied problems of image analysis and recognition are derived. The considered classifications of image features enable the construction of multiple-aspect image representations that preserve information essential to an applied problem. As a tool for choosing the features that depend on specific characteristics of a given problem and the initial data, we propose using parametrical generating descriptive trees, which support the creation and use of multiple-aspect image representation on the basis of different classifications of image features.

Information technology for the morphological analysis of the lymphoid cell nuclei

The new results of the research in the field of automation of hematopoietic tumor diagnostics by analysis of the images of cytological specimens are presented. The main result is a new information technology for the morphological analysis of the lymphoid cell nuclei of patients with hematopoietic tumors based on the combined use of pattern recognition and image analysis techniques. The principal characteristic of the proposed technology is that the features used for description of lymphocyte nuclei are chosen and calculated from the images of specimens by image processing and analysis methods, and also by methods of mathematical morphology and Fourier analysis. The proposed technology provides transition from the diagnostic analysis of lymphocyte nuclei to diagnosing the patients with hematopoietic tumors by means of pattern recognition techniques. Experimental check of the technology shows that it can be successfully used in program system for automated diagnostics of the hematopoietic tumors.

Automating extraction and analysis of dopaminergic axon terminals in images of frontal slices of the striatum

We develop a method to automate the obtaining of experimental data necessary to supplement the model of the preclinical stage of Parkinson’s disease. As a source of experimental data for constructing the model, images of stained slices of the brains of experimental animals are used. The suggested method is based on mathematical morphology operations and intended for automatic extraction of dopaminergic neuron terminals in images of frontal slices of the striatum. Experimental studies have confirmed the possibility and expediency of automating the processing and analysis of images of slices using this method.

Fundamental concepts and elements of image analysis ontology

The problem of development of the domain ontology “Analysis and evaluation of information represented by images” is considered. Brief description of methodology of ontology use in solution of image analysis problems is provided. The structure of an experimental version of image analysis ontology is described.

Descriptive approach to image analysis: Image models

The brief review of main methods and features of the descriptive approach to image analysis (DAIA), viz. forming the system of concepts that characterize the initial information-images-in recognition problems, and descriptive image models designed for recognition problems, is given.At present, in terms of development of image analysis and recognition, it is critical to understand the nature of the initial information, viz. images, find methods of image representation and description to be used to construct image models designed for recognition problems, establish the mathematical language for the unified description of image models and their transformations that allow constructing image models and solving recognition problems, construct models to solve recognition problems in the form of standard algorithmic schemes that allow, in the general case, moving from the initial image to its model and from the model to the sought solution. The DAIA gives a single conceptual structure that helps develop and implement these models and the mathematical language. The main DAIA purpose is to structure and standardize different methods, operations and representations used in image recognition and analysis. The DAIA provides the conceptual and mathematical basis for image mining, with its axiomatic and formal configurations giving the ways and tools to represent and describe images to be analyzed and evaluated.

A two-step approach for automatic microscopic image segmentation using fuzzy clustering and neural discrimination

The early diagnosis of lymphatic system tumors heavily relies on the computerized morphological analysis of blood cells in microscopic specimen images. Automating this analysis necessarily requires an accurate segmentation of the cells themselves. In this paper, we propose a robust method for the automatic segmentation of microscopic images. Cell segmentation is achieved following a coarse-to-fine approach, which primarily consists in the rough identification of the blood cell and, then, in the refinement of the nucleus contours by means of a neural model. The method proposed has been applied to different case studies, revealing its actual feasibility.

Descriptive approach to image analysis: Image formalization space

The paper presents and discusses the main results obtained using the descriptive approach to analyzing and understanding images when solving fundamental problems of the formalization and systematization of the methods and forms of representing information in the problems of the analysis, recognition, and understanding of images, in particular that arise in connection with the automation of information extraction from images in order to make intelligent decisions (diagnosis, prediction, detection, evaluation, and identification of patterns). In this direction, so far, the following results have been obtained: (1) the conceptualization of a system of concepts that describe the initial information (images) in recognition problems has been carried out; (2) descriptive models of images focused on the recognition problem have been defined; (3) the image-formalization space has been introduced, the elements of which include different forms (states, phases) of representing the image transformed from the original form into the recognizable one, i.e., into the image model; (4) the basic axioms of the descriptive approach were introduced. Axiomatics and its formal structures provide the methods and tools of representation and the description of images for their subsequent analysis and evaluation.

Descriptive approach to medical image mining. An algorithmic scheme for analysis of cytological specimens

The present paper is devoted the development and formal representation of a descriptive model for an information technology to automate the morphological analysis of cytologic preparations (a tumor of the lymphatic system). The theoretical basis of the model is a descriptive approach to image analysis and understanding and its main mathematical tools. Practical application of the algebraic tools of the descriptive approach is demonstrated, and the algorithmic scheme of the technology is described in the language of descriptive image algebras.

The descriptive approach to image analysis current state and prospects

The presentation is devoted to the research of mathematical fundamentals for image analysis and recognition procedures. The final goal of this research is automated image mining: a) automated design, test and adaptation of techniques and algorithms for image recognition, estimation and understanding; b) automated selection of techniques and algorithms for image recognition, estimation and understanding; c) automated testing of the raw data quality and suitability for solving the image recognition problem. The main instrument is the Descriptive Approach to Image Analysis, which provides: 1) standardization of image analysis and recognition problems representation; 2) standardization of a descriptive language for image analysis and recognition procedures; 3) means to apply common mathematical apparatus for operations over image analysis and recognition algorithms, and over image models. It is shown also how and where to link theoretical results in the foundations of image analysis with the techniques used to solve application problems.

Cell image analysis ontology

Cell image analysis in microscopy is the core activity of cytology and cytopathology for assessing cell physiological (cellular structure and function) and pathological properties. Biologists usually make evaluations by visually and qualitatively inspecting microscopic images: this way, they are particularly able to recognize deviations from normality. Nevertheless, automated analysis is strongly preferable for obtaining objective, quantitative, detailed, and reproducible measurements, i.e., features, of cells. Yet, the organization and standardization of the wide domain of features used in cytometry is still a matter of challenging research. In this paper, we present the Cell Image Analysis Ontology (CIAO), which we are developing for structuring the cell image features domain. CIAO is a structured ontology that relates different cell parts or whole cells, microscopic images, and cytometric features. Such an ontology has incalculable value since it could be used for standardizing cell image analysis terminology and features definition. It could also be suitably integrated into the development of tools for supporting biologists and clinicians in their analysis processes and for implementing automated diagnostic systems. Thus, we also present a tool developed for using CIAO in the diagnosis of hematopoietic diseases.