METHODOLOGY FOR FORMING A DATABASE OF REFERENCE IMAGES OF BONE MARROW CELLS FOR THE AUTOMATED DIAGNOSIS OF ACUTE LEUKEMIA IN ONCOHEMATOLOGY

В статье представлено описание факторов, влияющих на качество формирования базы эталонных изображений клеток костного мозга для диагностики острых лейкозов с применением методов компьютерной микроскопии. Отмечена важность контроля качества подготовки препаратов и микроскопа для применения в автоматизированных системах анализа изображений. Рассмотрены особенности регистрации цифровых микроскопических изображений клеток костного мозга в системах компьютерной микроскопии. Исследовано влияние фокусировки оптической системы микроскопа и уровня освещения препарата на формирование цифровых изображений клеток костного мозга. Установлены требования к условиям регистрации цифровых изображений, используемых в автоматизированных системах микроскопического анализа препаратов костного мозга. Предложена концептуальная модель базы эталонных изображений костного мозга, являющаяся основой для разработки инструментов эффективного распознавания клеток костного мозга в системах компьютерной микроскопии. Следование указанным требованиям к регистрации изображений призвано обеспечить надлежащее качество эталонной базы, что имеет непосредственной влияние на повышение точности и достоверности медицинской диагностики с применением методов компьютерной микроскопии. Результаты работы могут быть использованы в системах поддержки принятия врачебных решений при диагностике острых лейкозов The article describes the factors affecting the quality of the formation of a database of reference images of bone marrow cells for the diagnosis of acute leukemia using computer microscopy methods. The importance of quality control of specimen and microscope preparation for use in automated image analysis systems is noted. The features of registration of digital microscopic images of bone marrow cells in computer microscopy systems are considered. The effect of focusing of the optical system of the microscope and the level of illumination of the specimen on the formation of digital images of bone marrow cells is investigated. The requirements for the conditions of registration of digital images used in automated systems of microscopic analysis of bone marrow preparations have been established. A conceptual model of the base of reference images of bone marrow is proposed. It is the basis for the development of tools for effective recognition of bone marrow cells in computer microscopy systems. Following the specified requirements for image registration is designed to ensure the proper quality of the reference base of images, which has a direct impact on improving the accuracy and reliability of medical diagnostics using computer microscopy. The results of the work can be used in medical decision support systems for the diagnosis of acute leukemia

Integrated information and measurement system for the diagnosis of acute leukemia and minimal residual disease based on computer microscopy, flow laser cytofluorimetry and artificial intelligence

The article considers a new integrated information and measurement system for the diagnosis of acute leukemia and minimal residual disease based on computer microscopy and flow laser cytometry. The system is based on combining the results of computer microscopy in the analysis of bone marrow preparations and the results of flow laser cytofluorimetry. A special feature of the system is the use of artificial intelligence technologies in the recognition of images of bone marrow cells in the computer microscopy subsystem. The work was the result of joint work of the Department of Computer Medical Systems of the National Research Nuclear University "MEPhI" and the Laboratory of Hematopoietic Immunology of the National Medical Research Center of Oncology named after N. N. Blokhin.

Using Computer Microscopy Methods to Control the Microstructure of Malleable Cast Iron Product with Spherical Graphite

The method of identification of objects on images of the microstructure of cast iron with spherical graphite of the correct shape with uniform distribution is presented. Morphological analysis techniques were used to identify shrinkage pores and graphite inclusions in microstructure images. Geometric features of the shape of graphite inclusions were used as methods for identifying graphite, in particular, particle size analysis, which is widely used to identify various objects in computer microscopy. The computer analysis of the image was performed with the program ImageJ. To determine the pores against the background of graphite inclusions, two characteristics were used - the shape and size of the objects themselves. The pores, presented on the image, differ from graphite inclusions by a complex, fractal border and comparatively large areas. For the visualization of the research results, the combination of the graphite part with the calculation and analytical part was used. Such presentation of the results is the most significant and allows to perform the most correct evaluation of the graphitized cast iron microstructure in accordance with GOST 3443-87.

A Study of Noise Characteristics on Images in Computer Microscopy Systems

The characteristics of noise on images in computer microscopy systems are investigated, which is important when solving problems of automated analysis of low-contrast objects. The parameters of noise in a computer microscopy system were determined using a Watec 231 camera and an Olympus CX31 motorized microscope. Recommendations for suppressing the detected noise are given.

Computer Microscopy of Biological Fluid Dry Patterns for Medical Diagnostics

.