DEVELOPMENT OF A METHOD FOR ANALYZING TOMOGRAPHIC IMAGES OF BONE STRUCTURES

Abstract

The subject matter of research in the article is the morphological structure of bone tissue in the lumbar spine, visualized by tomography in the sagittal and axial planes. The goal of the research is to create the most informative investigation method for analyzing the structure of bone tissue, taking into account pathology in the form of metastatic bone lesions. This is justified by the fact that the detection of pathological processes is one of the most important tasks of image processing and analysis; at the same time, early diagnosis of various pathologies, including cancer, significantly increases the chances of patient recovery. Tasks: to consider the existing modern methods for analyzing the structure of bone tissue, to develop and propose a method for detecting bone tissue pathology in multiple myeloma. For the development of methods for the analysis of tomographic images with lesion of the bone structure, one of the fundamental issues is visualization of tomographic data. In this case, it is advisable to provide modules for both two-dimensional and three-dimensional visualization with methods of processing and segmentation of vertebral bodies, as well as correcting the results obtained in an interactive mode. The research uses methods of improving the quality of images, filtering using adaptive local filters, segmentation and stereology methods, cluster analysis method. The result of the work is to obtain a method suitable for use in the analysis of bone tissue with its accompanying pathology in the form of bone metastases. This method will be the basis for the further development of a method aimed at analyzing the microstructure of bone tissue, which will significantly increase the accuracy of calculations. Conclusions. The relevance of the topic under study is of vital importance for a huge number of patients suffering from cancer. In the course of the research, algorithms for processing and analyzing input images were developed, taking into account the modality of the input data. This makes it possible to develop the next stage of analysis aimed at the microstructure of the bone tissue.