A5082150058- Advanced Neuroimaging Techniques and Applications
- Functional Brain Connectivity Studies
- Engineering Technology and Methodologies
- Medical Image Segmentation Techniques
- 3D Shape Modeling and Analysis
- Advanced MRI Techniques and Applications
- Neural dynamics and brain function
- Material Science and Thermodynamics
- Economic and Technological Systems Analysis
- Medical Imaging and Analysis
- Engineering and Environmental Studies
- Ergonomics and Human Factors
- Technology and Human Factors in Education and Health
- Distributed and Parallel Computing Systems
- Computer Graphics and Visualization Techniques
- Diverse Industrial Engineering Technologies
- Computational Physics and Python Applications
- Graph Theory and Algorithms
- Control and Dynamics of Mobile Robots
- Advanced Computational Techniques in Science and Engineering
- Energy, Economy, and Technology Trends
- Technology Assessment and Management
- Advanced Graph Neural Networks
- Digital Transformation in Industry
- Engineering Education and Technology
Institute of Precision Mechanics and Control
2014-2024
Saratov State University
2022-2024
Yuri Gagarin State Technical University of Saratov
2019-2022
Russian Academy of Sciences
2012-2020
University of Southern California
2017-2019
Southern California University for Professional Studies
2017-2019
Bashkir State University
2018
Institute for Information Transmission Problems
2016-2018
Institute for Analytical Instrumentation
2018
National Research University Higher School of Economics
2016
Brain imaging researchers regularly work with large, heterogeneous, high-dimensional datasets.Historically, have dealt this complexity idiosyncratically, every lab or individual implementing their own preprocessing and analysis procedures.The resulting lack of field-wide standards has severely limited reproducibility data sharing reuse.
The structural connectome classification is a challenging task due to small sample size and high dimensionality of feature space. In this paper, we propose new data prepossessing method that combines geometric topological normalization significantly improves results. We validate approach by performing between autism spectrum disorder normal development connectomes in children adolescents. demonstrate significant enhancement performance using weighted normalized over the best available model...
This paper aims to tackle the problem of brain network classification with machine learning algorithms using spectra networks' matrices. Two approaches are discussed: first, linear and tree-based models trained on vectors sorted eigenvalues adjacency matrix, Laplacian matrix normalized Laplacian; next, SVM classifier is kernels based information divergence between eigenvalue distributions. The latter approach gives promising results in autism spectrum disorder versus typical development...
The relationship of the Industry 4.0 concept with cyber-physical systems and digital twins is described. analysis capabilities modern information discrete continuous production for creation presented. stages creating a twin are proposed, taking into account requirements standards development automated control technological processes. Using use case diagram, functionality subsystems system determined. architecture managing life cycle process was developed.
The current state and prospects for the use of microfluidic devices in detection nucleic acids at biological medical research are discussed article. main trends development analytical instruments base a platform indicated. most usually used technologies formation micro- nanoscale functional structures noted. Popular materials mass production highlighted. Examples domestic developments given perspective directions their application discussed.
Abstract We present several deep learning models for assessing the morphometric fidelity of grey matter region extracted from brain MRI. test three different convolutional neural net architectures (VGGNet, ResNet and Inception) over 2D maps geometric features. Further, we a novel geometry feature augmentation technique based on parametric spherical mapping. Finally, an approach model decision visualization, allowing human raters to see areas subcortical shapes most likely be deemed failing...
Abstract In recent years, medical imaging methods have been intensively developed to provide comprehensive and extensive data for studying the work of brain, which makes present study relevant. The purpose this is develop a new approach analyzing EEG signals based on nonlinear dynamics methods. To achieve goal, it necessary adapt solving analysis tasks. paper, following are used: principal component method noise clearance, wavelet transform, Wolf, Kanz, Rosenstein, neural networks calculate...
In this paper, we tackle a problem of predicting phenotypes from structural connectomes. We propose that normalized Laplacian spectra can capture properties brain networks, and hence graph spectral distributions are useful for task connectome-based classification. introduce kernel is based on earth mover's distance (EMD) between networks. access performance an SVM classifier with the proposed classification autism spectrum disorder versus typical development publicly available dataset....
Аннотация.В статье проведен анализ формования листового стекла флоат-методом, определены виды нештатных ситуаций технологического процесса.Выполнен кластерный множества ситуаций, каждая из которых описана как теоретико-множественная модель.Для упрощения интерфейса цифрового тренажера разработаны модели бизнес-процессов для обучения и аттестации операторов.На основе анализа существующих предприятия в нотации BPMN 2.0 логико-информационные перспективных Разработка тестов Аттестация...
In this work, we discuss methods for identifying build requisites, and describe their strengths weaknesses. The buildography tool is presented that provides logging of the process by tracking system calls. An estimate time spent on using given.
Abstract There is no consensus on how to construct structural brain networks from diffusion MRI. How variations in pre-processing steps affect network reliability and its ability distinguish subjects remains opaque. In this work, we address issue by comparing 35 connectome-building pipelines. We vary reconstruction models, tractography algorithms parcellations. Next, classify connectome pairs as either belonging the same individual or not. Connectome weights eight topological derivative...