A computational technique is proposed in which 2D segmentation is carried out using MRI data to prepare the formation of a 3D model of a brain tumor. The proposed technique is based on: pre-processing of MRI images to reduce noise; increasing brightness and contrast; formation of contours of objects of interest and visualization of the studied areas. The core computational element of the contouring and color coding technique is the Shearlet transform for magnetic resonance imaging (MRI) images.

The study is devoted to the use of algorithms for processing medical tabular data of patients with epilepsy. A data set of patients with epilepsy was analyzed to search for a correlation between the characteristics of the disease (type of epilepsy and drug resistance) and blood microRNA (miRNA) indicators. As a result, it was revealed that miR-134 (1) can be considered as a biomarker for determining the type of epilepsy. A prognostic model was built to classify the type of epilepsy according to the miR-134 indicator (1) – a decision tree. To predict pharmacoresistance a random forest model was built where the features are 9 miRNA indicators.

The study is devoted to improving the computational methodology for assessing the parameters of the tissue regeneration process using mesh nickelidetitanium implants with shape memory in an experiment. Processing and analysis of observational data from electron microscopy and classical histological examination were performed using proprietary algorithms and their modifications, which significantly simplify the data analysis procedure and increase the accuracy of estimates (15-20%).The proposed method as a computational tool for analyzing the dynamics of the processunder study as well as for highlighting the internal geometric features of experimentalimages of objects of interest contains shearlet transformation algorithms as well as algorithms for constructing elastic maps for effective visualization of spatial data. Important aspects of the methodology are computational tools for preprocessing visual data to increase the contrast and brightness of analyzed images based on Retinextechnology which significantly affects the quality of the use of computer evaluation tools.

The study of the wavelet characteristics of the signal helps to identify characteristic patterns in the patterns of physiological rhythms of a healthy person and to determine changes in the dynamic complexity of the patterns in the event of various pathological conditions. The paper shows the possibility of using the wavelet characteristics of EEG (electroencephalogram) patterns for automatic detection of epileptiform discharges. An algorithm for compressing MRI medical images associated with brain pathologies based on the use of Haar wavelets is proposed.

One of the most dangerous complications during brain surgery is bleeding. Hemostasis can be difficult due to the lack of visibility caused by blood filling the surgical wound. To restore the visibility of the surgical area it is proposed to use NIR-camera data and Shearlet transform with color-coding algorithms. At the same time it is also assumed to conduct brightness characteristics enhancement of images (frames) and segmentation of biological tissues of interest. NIR beams are able to penetrate deeper into tissues than visible light and NIR is also absorbed to a greater extent by hemoglobin than by surrounding tissues. The blood has a significantly higher absorption coefficient of NIR rays in the range from 800 nm to 1050 nm in comparison with the absorption coefficient of the same spectrum by the tissues involved during the operation. Due to this effect there is the potential to detect structures of interest despite bleeding in the wound cavity. During experimental study it was found that it becomes possible to visualize all tissues that are at a depth of up to 3 mm. Use of BCET algorithm with the mask for processing made it possible to improve the image contrast from 34.20% to 198.73% depending on the depth of biological structures. In case of model images processing the best average accuracy of determining ROI contours taking into account depth was 0.961 ± 0.021 according Dice similarity coefficient.

The work is devoted to the development of a system for visualizing vessels in a surgical wound with massive bleeding to improve the observation of structures in the surgical field. A modification of the video endoscope and the method of illumination of the surgical wound is proposed. At the same time during the development of a video endoscopic system the ability to operate in near infrared and color modes is taken into account. As part of the experiments an analysis was made of the quality of visualization of biological structures depending on the depth of their location in the surgical wound. As a result a system and methodology have been created that improve the quality of visualization of blood vessels under bleeding conditions. The results obtained can be applied in endoscopic surgery.

Исследование посвящено разработке системы визуализации операционного поля во время массивного кровотечения. Для решения проблемы отсутствия визуализации биологических структур предлагается система из эндоскопа ближнего инфракрасного диапазона (NIR) и алгоритмического обеспечения, особенностью которого является возможность сегментации кровеносных сосудов и иных биологических тканей, задействованных во время операции. В рамках экспериментального исследования выполнено тестирование системы и комбинации алгоритмов. The study is devoted to the development of a visualization system for the surgical field during massive bleeding. In order to solve the problem of the lack of visualization of biological structures a system of a near infrared range (NIR) endoscope and algorithmic support is proposed. It's feature is the possibility of segmenting blood vessels and other biological tissues involved during the operation. As part of the experimental study, testing of the system and a combination of algorithms was performed.

The study is devoted to the processing and analysis of medical experiment data as part of the task of segmenting biological structures in images obtained from the endoscope camera. A review of methods for improving the brightness characteristics of images, segmentation and edge detection is presented. As a result of the research, a computational technique has been developed for segmentation and edge detection in medical images based on shearlet transform and color coding. Experimental studies have been carried out to solve the problems of segmentation and contour detection in images obtained as a result of medical experiments. Testing of the technique showed high accuracy in identifying the boundaries of objects of interest taking into account the depth of the studied tissues.

The aim of the work is to develop a method for extracting a breast tumor and an uninfected area of the breast within the framework of medical imaging using a combination of fuzzy clustering with discrete wavelet transform (FCMDWT) tools. Two sets of mammographic breast image data were used in the study: a subset of breast images of a digital base mammography screening data (CBIS-DDSM) and private data set. Contrast enhancement techniques were used to improve image quality before using FCMDWT for tumor segmentation and analysis. To assess the accuracy of tumor isolation metrics such as the Dice coefficient (F1 score) and intersection by union (IoU) were used. The mean IoU was 98.41 and F1 score was 96.47 using the proposed FCMDWT.

Camera traps generating a huge number of images help to study and monitor the wildlife. However, camera traps work at any time of the day and under any weather conditions. Therefore, many images have low or high illumination, blurring, and other defects. This complicates image analysis by both humans and computer systems. In this study, we develop an adaptive illumination correction algorithm based on a modified Multi-Scale Retinex (MSR). First, we accelerate computation by using recursive implementation of the Gaussian filter and utilizing look-up tables to find logarithms and new brightness values. Second, response of the MSR function is transformed by a modified threshold normalization to improve image quality. The upper and lower thresholds are calculated based on statistical information. Finally, we offer automatic adjustment of parameters depending on the area of the image in order to increase usability. Proposed algorithm was tested with various settings on a set of images obtained from camera traps. Experimental results show a high potential for its application.

ФЕДЕРАЛЬНЫЙ ИССЛЕДОВАТЕЛЬСКИЙ ЦЕНТР ФУНДАМЕНТАЛЬНОЙ И ТРАНСЛЯЦИОННОЙ МЕДИЦИНЫ ФЕДЕРАЛЬНЫЙ ИССЛЕДОВАТЕЛЬСКИЙ ЦЕНТР «КРАСНОЯРСКИЙ НАУЧНЫЙ ЦЕНТР СИБИРСКОГО ОТДЕЛЕНИЯ РОССИЙСКОЙ АКАДЕМИИ НАУК» ИНСТИТУТ ВЫЧИСЛИТЕЛЬНОГО МОДЕЛИРОВАНИЯ СО

Currently wavelets are widely used in various fields related to the processing and analysis of signals and images. One application is the task of image compression. The advantage of compression using wavelet algorithms is that they work with the entire image and not with its individual blocks. At high compression ratios this avoids a block structure. This paper analyzes the effectiveness of the compression algorithm for medical MRI images with brain pathology which is based on the Haar wavelet transform. Implementation effective compression methods will allow storing a larger number of patient tomograms on a central server without increasing storage volumes. Experiments were performed and the dependences of indicators for assessing the quality of compressed MRI images on the compression ratio were constructed. The algorithm with zeroing wavelet coefficients by a threshold showed a better result compared to zeroing by levels based on assessing the quality of a compressed MRI image using a set of corresponding metrics.

Breast tumor segmentation and boundary detection are crucial stages in breast cancer therapy and follow-up. Radiologists can minimize the high workload associated with screening for breast cancer by automating this complex process. This article proposes a system for segmenting breast and unaffected areas (breast) tumors on medical images using a combination of fuzzy clustering and thresholding (FCMT) tools. This computer diagnostic method works with each section of the mammary gland without learning segmentation and definition of boundaries. As part of the approbation, two databases of breast mammogram images were used. To increase the image quality, we used pre-processing techniques such as contrast augmentation before applying the FCMT for segmentation. To assess the effectiveness of the devised approach, the Mean Square Error, dice coefficient, Structured Similarity Index, Peak Signal-to-Noise Ratio, accuracy, and sensitivity were computed. Using the proposed FCMT segmentation technique, a mean intersection over union (IoU) of 93.85 was attained. The presented approach is more resilient and accurate in segmenting tumor progression on medical images, according to the findings of the experiments.

The study of principles of brain structure formation and development is necessary for enriching fundamental knowledge both in the field of neurophysiology and in medicine. A detailed description of all the brain features will allow you to choose the most effective therapy method or check the effectiveness of drugs being developed. The basis for creating a model of a biological neural network is a map of nerve cells and their connections. To obtain it, it is necessary to carry out microscopy of the cell culture, which will allow obtaining a low-contrast image. The study of these images is a complex task, so we have developed a contrasting algorithm using color-coding for images processing, designed to improve the process of creating a neural network model.

One of the most dangerous complications during brain surgery is bleeding. Hemostasis can be difficult due to the lack visibility caused by blood filling the surgical wound. To restore the visibility of the surgical area, it is proposed to use NIR-camera data and shearlet-transform with color-coding algorithms. At the same time, it is also assumed to conduct brightness characteristics enhancement of images (frames) and segmentation of biological tissues of interest

In recent years computed tomography of the lungs has been the most common diagnostic procedure aimed at detection of the pathological changes associated with COVID-19. The study is aimed at the use of the developed algorithmic support in combination with texture (geometric) analysis to highlight a number of indicators characterizing the clinical state of the object of interest. Processing is aimed at the solution of a number of diagnostic tasks such as highlighting and contrasting the objects of interest, taking into account the color coding. Further, an assessment is performed according to the appropriate criteria in order to find out the nature of the changes and increase both the visualization of pathological changes and the accuracy of the X-ray diagnostic report. For these purposes, it is proposed to use preprocessing algorithms for a series of images in dynamics. Segmentation of the lungs and areas of possible pathology are performed using wavelet transform and Otsu threshold value. Delta-maps and maps obtained using Shearlet transform with contrasting color coding are used as a means of visualization and selection of features (markers). The analysis of the experimental and clinical material carried out in the work shows the effectiveness of the proposed combination of methods for studying of the variability of the internal geometric features (markers) of the object of interest in the images.

The study is devoted to the development of information and methodological support in relation to expert assessment of the interaction and mutual influence of objects of the business ecosystem taking into account their spatial location. As a computational toolkit the work uses adapted methods of clustering and visualization in the form of constructing a set of Voronoi and elastic maps. The purpose of the study is to identify hidden patterns in multidimensional data in relation to the studied complex processes. The article was prepared within the framework of the Grant of the RFBR and the Government of the Krasnoyarsk Territory No. 20-410-242916 / 20 r_mk Krasnoyarsk.

The paper discusses modern approaches and digital transformations in business models and interactions. In this regard for a quantitative description of interactions in ecosystems a variant of methodological support based on neural networks is proposed for fast nonlinear multiparametric regression of large data sets within the projected expert system. The possibility of effective solution of the problem of filling gaps in the observational data arrays and processing of not precisely specified information is shown. This approach is proposed for solving predictive problems in the problem of interaction of objects of interest in business ecosystems. The article was prepared within the framework of the Grant of the RFBR and the Government of the Krasnoyarsk Territory No. 20-410-242916 / 20 r_mk Krasnoyarsk.