Departments
Home Departments
Contacts
220012, Republic of Belarus, Minsk, Surganov st., 6
Directions of activity
Cooperation and services
- Development of methods of classification and recognition of images and their adaptation to modern tasks of analysis of medical images
- Development of new methods and algorithms for detecting and localizing structural differences and individual changes in the spatial-bright structure of 2D and 3D images of organs and systems of the human body and their application in medical diagnostics
- Development of methods and software tools for joint analysis of images of various types (computer-tomographic, nuclear-magnetic-resonance, emission, ultrasound, cytological, hystological) for use in early diagnosis of cancer and other socially significant diseases
- Development of statistical models and procedures for the automatic detection of links between morphological, space-bright and textural features of the analyzed images, as well as clinical and genetic data on patients and parameters of biological samples used
- Development of data mining methods, automatic search of patterns in complex databases combining biomedical images of various modalities and factographic information
- Research and development of new technologies, algorithms and software for the effective organization of computational processes and the use of advanced hardware to solve problems of processing and analysis of biomedical images
- Development of methods and algorithms for identifying the current position of the autonomous moving device based on incoming sensory information and on-board digital maps
Results of scientific research and development
- Micromodule of operative recognition, selection and compression of video information on board small spacecraft The neural network software for decision-making in the diagnosis of lung diseases based on x-ray and tomographic images (lung expert)
- Methods, algorithms and software tools for detecting relationships between the structure of lung images and drug resistance of tuberculosis patients
- Models, methods and basic software for the search for objects in medical images
- Development and research of binarized space images descriptors based on the use of local specimisms of pixels
- Software complex of radiological image analysis to support early diagnosis of malignant lung neoplasms
- Complex of software tools for the preparation and control of information support of flight tasks Experimental software that allows you to assess the parameters of the development of the capillary circulatory network in ovarian cancer in women based on the analysis of histological images
- Methods of search of linear objects and the places of their intersections based on the rapid transformation of Hafa
- Methods of X-ray images, as well as methods for detecting and visualizing the internal structure of hidden objects (summers) in images
- Methods of processing, analysis and classification of multidimensional data and images
- Medical image search methodology based on the idea of multi-grade matrix joint encounters
- System of prediction and diagnosis of infected pancreatic necrosis
- Methods of exploration analysis of the form of objects on images (image timid mining) in order to detect "interesting" (non-trivial natural patterns)
- Methods of calculation of multidimensional matrixs of joint encountering pixels to describe the structure of three-dimensional medical images with non-parallel layers