Background. Mucopolysaccharidosis Type 1 (MPS 1) is a rare inherited lysosomal metabolism disorder that leads to the accumulation of glycosaminoglycans in body tissues. Due to neurodegenerative changes, patients with MPS 1 experience cognitive and motor impairments that require continuous clinical monitoring. Magnetic resonance imaging (MRI) is the primary tool for visualizing structural changes in the brain; however, traditional MRI data analysis methods have several limitations related to subjective image interpretation and the high complexity of data processing. Materials and methods. This study proposes an algorithm for automated MRI data analysis aimed at assessing the dynamics of brain structure changes in patients with MPS 1. The algorithm includes several key stages: image preprocessing, segmentation of brain structures, quantitative analysis of gray and white matter volumes, as well as ventricular structures. The calculation of brain structure volumes is performed based on segmented data, followed by comparison with age– and sex-specific normative values. A particular focus of the algorithm is the calculation of deviations in the volumes of gray and white matter, as well as brain ventricles, from reference values, which allows for a quantitative assessment of the degree of neurodegenerative changes. Based on this data, disease progression is evaluated by comparing parameters at successive time points. The introduced criteria for changes in brain structure volumes help determine the degree of pathological process progression and forecast further disease development through extrapolation of current trends. Results and conclusions. The algorithm has been implemented in the MATLAB programming environment, enabling automation of image processing, increased calculation accuracy, and ease of result interpretation. The developed approach was tested on clinical data from patients diagnosed with MPS 1, which confirmed its effectiveness in assessing dynamic changes in brain structures.

Statement of the problem. Diagnosis and monitoring of mucopolysaccharidoses (MPS), a group of hereditary lysosomal storage diseases, remain challenging due to the variability of clinical manifestations and the need for an integrated approach. Advances in molecular genetic and neuroimaging methods require systematization and analysis of modern approaches, including the integration of advanced technologies such as artificial intelligence (AI) and machine learning (ML), to improve the accuracy of diagnosis, especially when assessing brain changes. There is a need to standardize methods and develop new algorithms for quantitative data analysis and assessment of disease dynamics. Objective. To conduct a comprehensive analysis of modern methods for diagnosing and monitoring MPS, covering laboratory, instrumental and neuroimaging approaches; to consider the use of advanced technologies (ML, AI) to improve the accuracy of detection of pathological changes in the brain; to pay attention to the role of quantitative image analysis, segmentation methods and their application in automated systems; to describe the main biomarkers and problems of data interpretation; to assess the need for an interdisciplinary approach and standardization of methods for the formation of promising areas for the development of diagnostics and monitoring of MPS. Results. A comprehensive analysis of the methods for diagnostics and monitoring of MPS is presented. Laboratory (urine analysis for GAG, enzyme analysis, molecular genetic analysis) and instrumental methods (X-ray, CT, ultrasound, MRI) are described, with an emphasis on MRI as the "gold standard" for CNS visualization. Universal programs for processing MRI images (SPM, FreeSurfer, 3D Slicer, BrainSuite) and image correction methods (Gaussian, median, Fourier, anisotropic diffusion, wavelet filtering), their advantages and limitations in MPS are considered. The problems of assessing disease progression are discussed, including variability of the course, limited availability of quantitative analysis methods and the inaccuracy of existing segmentation algorithms in the presence of pronounced structural changes. The role of AI and ML in solving these problems is emphasized. The article covers the use of biomarkers and the need to standardize diagnostic approaches. The importance of interdisciplinary interaction (genetics, neurology, radiology, bioinformatics) is substantiated. Practical significance. The results of the analysis systematize information on modern diagnostic and monitoring capabilities for MPS. A review of advanced technologies, such as AI and image processing methods, informs specialists about the potential for their implementation to improve the accuracy of diagnosis and assess the dynamics of the disease. The identified problems and proposed directions (standardization, development of new algorithms, integration of biomarkers, interdisciplinary approach) can serve as a basis for further research and improvement of clinical practice in managing patients with MPS, contributing to improving their quality of life and the effectiveness of treatment.

The diagnosis and monitoring of neurodegenerative changes in children with mucopolysaccharidoses (MPS) represent one of the most complex challenges in modern pediatric neuroradiology. Despite the widespread use of magnetic resonance imaging (MRI), the interpretation of images often remains subjective and depends on the specialist’s level of expertise. Existing software tools, such as FreeSurfer and SPM, do not provide sufficient accuracy in the analysis of pediatric MRI data, particularly in rare hereditary disorders accompanied by specific MRI signal abnormalities. The absence of specialized tools for quantitative assessment of morphometric changes and disease progression prediction limits the physician’s ability to objectively evaluate therapy effectiveness and plan further patient management strategies. Objective – to develop and validate a software-algorithmic complex (SAC) for brain MRI analysis in children with mucopolysaccharidosis, enabling automated filtering, segmentation, quantitative assessment of brain structures, and disease progression forecasting, with the goal of improving diagnostic accuracy and reproducibility, as well as creating a clinical decision support tool for practical use. A software-algorithmic complex for brain MRI analysis in children with mucopolysaccharidosis has been developed and validated. The use of a combined wavelet transform and anisotropic diffusion method significantly improved image filtering quality while preserving clear structural boundaries. The diagnostic model based on logistic regression demonstrated high classification accuracy for MPS patients, achieving an AUC of 0.95. The prognostic module enabled disease dynamics modeling using linear and exponential models, showing less than 6% deviation from actual clinical data. The exponential model most accurately described the accelerated neurodegenerative changes typical of MPS type I. Validation was performed on data from 19 patients, including 15 with genetically confirmed MPS and 4 conditionally healthy children. The results showed a 10.1% increase in segmentation accuracy compared to FreeSurfer. The complex demonstrated high reproducibility and diagnostic reliability. The obtained data confirm its effectiveness as a tool for quantitative diagnosis and disease dynamics monitoring in clinical practice. The developed complex provides physicians with objective quantitative indicators for diagnosing and monitoring neurodegenerative changes in MPS, improving analysis reproducibility and reducing subjectivity. The software can be integrated into clinical workstations for automated MRI processing, applied in the evaluation of enzyme replacement therapy effectiveness, and used in clinical trials of new treatment methods and drug delivery approaches across the blood-brain barrier. Moreover, the proposed approach is universal and can be adapted for other neurodegenerative and metabolic diseases in children.

The effectiveness of the function of any application also depends on the structure itself, compiled by the developers. Each function should work according to the compiled algorithm, the health of the application, but it should be noted that some applications may lack any functions. Application development time may depend on the procedure for its development, as well as on the quality of application development by structures. There are several categories of applications: engineering applications; medical applications; information systems. This program was developed for laboratory examinations and diagnosis of brain diseases. The software is designed to work with brain research, data from conducted studies, as well as violations of the integrity and performance of the brain. MRI is one of the modern and reliable research methods. An analysis of the results of the developed program in comparison with existing brain MRI methods shows that they can be detected by similarities, but also have differences. It depends on how the methods themselves are conducted and how well the program is used. The application was written in C++, a higher-level programming language, and at a basic level, to display the scales of images, as well as to display the exact size of brain tumors, as well as its cortex, in order to ensure the practice of its treatment.

Background. Research into the formation of algorithms for obtaining MRI images is highly relevant. The purpose of the work is to analyze advances in the development of a software algorithm for obtaining MRI images of the brain. Materials and methods. Found 17 articles in specialized databases Cyberleninka, eLibrary, PubMed, ScienceDirect. Methods of analysis, synthesis and induction were used. Results. The best way to suppress noise in brain MRI images is the Gaussian Filter, the improvement of which has been achieved through the evolution of neural networks. Automatic segmentation achieved performance comparable to manual segmentation by using a combined system with integrated modules to eliminate the influence of noise and background, to identify image features and edge information. The Sobel operator allows the bright edges of an MRI image to be more clearly identified for removal. For volumetric visualization of brain images, due to its labor-intensive nature, the use of layer-by-layer representation of MRI data is proposed. Watershed segmentation and the K-nearest neighbor classification algorithm resulted in an MRI image accuracy of only 89 %; the wavelet transform was performed without calculating the accuracy. Support Vector Machine (SVM) using the GLCM algorithm showed an accuracy of up to 93 %, but only 36 images were used for training. Based on 150 MRI images of the brain, their classification was performed using the MATLAB 2018a software package (Matrix Laboratory) and a testing accuracy of 96,7 % was achieved. Conclusions. Improved algorithms for removing noise and bright edges from MRI brain images, segmenting them, and creating them with volumetric visualization are being created, including effective software modules for automatic segmentation based on convolutional neural networks.