Peer-Reviewed Publications

• Nature
• Oct 26, 2024
• DOI: 10.1038/s41598-024-77496-2

Post-COVID-19 syndrome (PCS) is an emerging health problem in people recovering from COVID-19 infection within the past 3–6 months. The current study aimed to define the predictive factors of PCS development by assessing the mitochondrial DNA (mtDNA) levels in blood leukocytes, inflammatory…

• Machine learning
• PCS

• Elsevier
• Jun 1, 2023
• DOI: 10.1016/j.vaccine.2023.05.010

This study compares the long-term antibody persistence in participants vaccinated with PreHevbrio™/PreHevbri®, a 3-antigen hepatitis B vaccine, against those receiving a single-antigen vaccine, revealing significantly higher seroprotection rates and mean antibody levels…

• Life science
• Vaccinology
• 3-antigen

• Wiley
• Oct 20, 2022
• DOI: 10.1002/jcc.27016

This study suggests a novel graph neural network architecture for chemical yield prediction. The network combines structural information about participants of the transformation as well as molecular and reaction-level descriptors.…

• Drug Discovery
• Deep Learning
• Graph Neural Networks
• Chemical Reaction Yield

• CSIT
• Sep 29, 2022
• DOI: 10.31891/csit-2022-3-5

This work introduces a method that detects and removes redundant weights of fully connected neural networks, simultaneously increasing the generalized efficiency of simplified models.…

• Deep Learning
• Pruning

• MDPI
• Jun 6, 2022
• DOI: 10.3390/math10111942

This paper focuses on the problem of effectively data preprocessing to improve the accuracy of intellectual analysis in the case of performing medical diagnostic tasks. The authors developed a new two-step data normalization method for numerical medical datasets. It is based on the possibility…

• Medical diagnostics
• Machine learning
• Multiclass classification

• MDPI
• May 22, 2022
• DOI: 10.3390/app12105238

This paper deals with the task of assessing the quality of the biocompatible medical implants production via additive technologies.The authors proposed PNN-SVM system that is based on the combining use of PNN, Ito Decomposition, and SVM. The PNN’s summation layer outputs were…

• Machine learning
• Smart technologies

• Wiley
• Feb 24, 2022
• DOI: 10.1002/jcc.26831

We present a graph-based deep neural network for prediction of protein-drug binding affinity and assess its predictive power under thorough testing conditions.…

• Drug Discovery
• Binding Affinity
• Deep Learning

• Elsevier
• Nov 4, 2021
• DOI: 10.1016/j.procs.2021.12.313

The authors in this paper proposed a new cascade structure based on linear SGTM neural-like structures with non-iterative machine learning, that: i) provides a significant increase in the accuracy of IoMT data approximation both due to the combined use of non-linear expansion of inputs…

• Non-linear ANN
• SGTM neural-like structure

• Elsevier
• Jun 10, 2021
• DOI: 10.1016/j.compbiolchem.2021.107529

Ensembling machine learning models to enhance the prediction of a molecular binding affinity between large protein molecules and small organic molecules.…

• Drug Discovery
• Binding Affinity
• Biotech
• Pharma
• Deep Learning

• MDPI
• Apr 6, 2021
• DOI: 10.3390/sym13040612

This paper considers  the  problem of effective intellectual analysis in the case of handling small data. The authors proposed a new additive input-doubling method for processing short and extremely short datasets. The main steps of the method should include the procedure of…

• Small data approach
• Machine learning
• SVR

• MBE
• Mar 17, 2021
• DOI: 10.3934/mbe.2021132

This paper considers the problem of handling short sets of medical data. The authors improved the regression method of data analysis based on artificial neural networks by introducing additional elements into the formula for calculating the output signal of the existing RBFbased input-doubling…

• Small data approach
• Neural networks
• RBF

• Wiley
• Feb 14, 2021
• DOI: 10.1002/jcc.26494

This paper focuses on a multi-stage pipeline, in which several deep neural networks are combined to generate novel drug candidates with desired properties.…

• Drug Discovery
• Molecular Generation
• Molecular Dynamics
• Deep Learning

• NIH
• Jan 11, 2021
• DOI: 10.3389/fgene.2020.630186

Recent research findings, however, indicate that TL per se can only allow a rough estimate of aging rate and can hardly be regarded as a clinically important risk marker for age-related pathologies and mortality.…

• Age-related telomere shortening
• Biomarker of biological age

• arxiv.org
• Dec 15, 2020
• DOI: 10.48550/arXiv.2012.08275

This study assesses the efficiency of several popular machine learning approaches in the prediction of molecular binding affinity. Presented at NeurIPS 2020.…

• Drug Discovery
• Binding Affinity
• Ensembled Prediction
• Deep Learning

• Cryobiology
• Dec 17, 2019
• DOI: 10.1016/j.cryobiol.2019.10.165

Analysis of the TM-curves showed that temperatures of cooling and heating should be changed taking into account the nature of phase transformations in each temperature range. In particular, at the stage corresponding to the phase transformation of sodium alginate (−5...−37 °С),…

• Life-science
• Cryobiology

• Elsevier
• Nov 4, 2019
• DOI: 10.1016/j.procs.2019.09.461

This paper describes a medical decision-making system based on analyzing patients' health records.…

• Precision Medicine
• Decision Support
• Data Platform

• Elsevier
• Aug 21, 2019
• DOI: 10.1016/j.procs.2019.08.010

Authors developed the method of allocating a pragmatic sign from text and built a mobile system for medical recommendations.…

• NLP
• Recommender Systems
• Missing Data Imputation

• CEUR-WS
• May 9, 2019

This work describes a data platform for tagging scientific articles using deep neural networks. The tagging pipeline includes a complete cycle of tasks - from collecting data to storing tags and managing the data lifecycle.…

• NLP
• Automated Information Extraction
• Documents Processing

• Springer
• May 3, 2019
• DOI: 10.1007/978-3-030-12082-5_31

In this work, the analysis of data verification of personal medical profiles of several major medical information systems is conducted.…

• Personal Medicine
• Data Platform
• Data Consistency Analysis

• Elsevier
• May 1, 2019
• DOI: 10.1016/j.comptc.2019.05.006

A number of computational methods are used to rationalize the experimentally observed formation of heterodimeric complexes…

• Drug Discovery
• Molecular Dynamics
• Chemoinformatics

• Cryobiology
• Dec 1, 2018
• DOI: 10.1016/j.cryobiol.2019.10.165

The aim of our study was to increase the survival of probiotic cells immobilized in a modified sodium alginate gel after low-temperature storage. The subjects were probiotic strains of bifidobacteria, lactobacilli, lactococci and yeast S. cerevisiae (var S. boulardii).…

• Life-science
• Cryobiology

• The American journal of clinical nutrition
• Sep 1, 2018

Even before the onset of age-related diseases, obesity might be a contributing factor to the cumulative burden of oxidative stress and chronic inflammation throughout the life course. Obesity may therefore contribute to accelerated shortening of telomeres. Consequently, obese persons…

• Deep Learning
• Biotech

• ResearchGate
• Aug 8, 2018
• DOI: 10.1109/DSMP.2018.8478556

In this review, the authors compared the efficiency of several computer vision algorithms for face recognition.…

• Face Recognition
• Computer Vision

• IEEE
• Jul 9, 2018
• 10.1109/MECO.2018.8405985

Authors developed a data platform for precise forecasting of clinical outcomes of chronic diseases. The training, evaluation, and interpretation of deep learning models were automated.…

• Personalized Medicine
• Data Platform
• Diagnostic System
• MLOps

• Experimental Biology and Medicine
• Feb 20, 2017

This study investigates the relationship between leukocyte telomere length and glucose levels in 82 non-diabetic, rural Ukrainian residents aged 45+, revealing no correlation with fasting plasma glucose but a negative association with 2-h post-load plasma glucose levels. Despite…

• Deep Learning
• Biotech
• Biochemical

• Front. Genet.
• Mar 16, 2015

Telomeres are specialized structures composed of repetitive DNA sequences and proteins at chromosome ends, protecting them from degradation and fusion while facilitating cell division; however, they shorten with each division due to replication challenges, influencing aging and…

• Deep Learning
• Biotech