Browsing by Author "Kolev M."

Now showing 1 - 11 of 11
  • No Thumbnail Available
    Item
    A mathematical model for single cell cancer - Immune system dynamics
    (2005-05-01) Kolev M.; Kozłowska E.; Lachowicz M.
    In this paper, we propose and analyse the model of competition between a single cell cancer and the immune system. The model is a system of integro-differential bilinear equations and it describes both very early stage of a solid tumor and all stages of leukemias. © 2005 Elsevier Ltd. All rights reserved.
  • No Thumbnail Available
    Item
    A mathematical model of cellular immune response to leukemia
    (2005-05-01) Kolev M.
    The cell-mediated immune response is a very important part of the defence mechanism against cancer. In this paper, we present a model of the cellular immune response to leukemia. The model is developed with statistical methods analogous to those of kinetic theory. The cells of the interacting populations are characterized by a microscopic functional state variable. The development of the concept of inner functional state is considered. A new possibility for definition of the activation state, suitable for experimental evaluation, for three particular cell populations is proposed. The presented simulations are related to the modelling of three types of vaccinations. © 2005 Elsevier Ltd. All rights reserved.
  • No Thumbnail Available
    Item
    Applications of Machine Learning in Human Microbiome Studies: A Review on Feature Selection, Biomarker Identification, Disease Prediction and Treatment
    (2021-02-19) Marcos-Zambrano L.J.; Karaduzovic-Hadziabdic K.; Loncar Turukalo T.; Przymus P.; Trajkovik V.; Aasmets O.; Berland M.; Gruca A.; Hasic J.; Hron K.; Klammsteiner T.; Kolev M.; Lahti L.; Lopes M.B.; Moreno V.; Naskinova I.; Org E.; Paciência I.; Papoutsoglou G.; Shigdel R.; Stres B.; Vilne B.; Yousef M.; Zdravevski E.; Tsamardinos I.; Carrillo de Santa Pau E.; Claesson M.J.; Moreno-Indias I.; Truu J.
    The number of microbiome-related studies has notably increased the availability of data on human microbiome composition and function. These studies provide the essential material to deeply explore host-microbiome associations and their relation to the development and progression of various complex diseases. Improved data-analytical tools are needed to exploit all information from these biological datasets, taking into account the peculiarities of microbiome data, i.e., compositional, heterogeneous and sparse nature of these datasets. The possibility of predicting host-phenotypes based on taxonomy-informed feature selection to establish an association between microbiome and predict disease states is beneficial for personalized medicine. In this regard, machine learning (ML) provides new insights into the development of models that can be used to predict outputs, such as classification and prediction in microbiology, infer host phenotypes to predict diseases and use microbial communities to stratify patients by their characterization of state-specific microbial signatures. Here we review the state-of-the-art ML methods and respective software applied in human microbiome studies, performed as part of the COST Action ML4Microbiome activities. This scoping review focuses on the application of ML in microbiome studies related to association and clinical use for diagnostics, prognostics, and therapeutics. Although the data presented here is more related to the bacterial community, many algorithms could be applied in general, regardless of the feature type. This literature and software review covering this broad topic is aligned with the scoping review methodology. The manual identification of data sources has been complemented with: (1) automated publication search through digital libraries of the three major publishers using natural language processing (NLP) Toolkit, and (2) an automated identification of relevant software repositories on GitHub and ranking of the related research papers relying on learning to rank approach.
  • No Thumbnail Available
    Item
    Correlation between animal and mathematical models for prostate cancer progression
    (2009-01-01) Jackiewicz Z.; Jorcyk C.L.; Kolev M.; Zubik-Kowal B.
    This work demonstrates that prostate tumour progression in vivo can be analysed by using solutions of a mathematical model supplemented by initial conditions chosen according to growth rates of cell lines in vitro. The mathematical model is investigated and solved numerically. Its numerical solutions are compared with experimental data from animal models. The numerical results confirm the experimental results with the growth rates in vivo. © 2009 Taylor & Francis.
  • No Thumbnail Available
    Item
    Deep Learning-Driven Insights into Hardness and Electrical Conductivity of Low-Alloyed Copper Alloys
    (2025-12-01) Kolev M.; Javorova J.; Simeonova T.; Hadjitodorov Y.; Krastev B.
    Understanding the intricate relationship between composition, processing conditions, and material properties is essential for optimizing Cu-based alloys. Machine learning offers a powerful tool for decoding these complex interactions, enabling more efficient alloy design. This work introduces a comprehensive machine learning framework aimed at accurately predicting key properties such as hardness and electrical conductivity of low-alloyed Cu-based alloys. By integrating various input parameters, including chemical composition and thermo-mechanical processing parameters, the study develops and validates multiple machine learning models, including Multi-Layer Perceptron with Production-Aware Deep Architecture (MLP-PADA), Deep Feedforward Network with Multi-Regularization Framework (DFF-MRF), Feedforward Network with Self-Adaptive Optimization (FFN-SAO), and Feedforward Network with Materials Mapping (FFN-TMM). On a held-out test set, DFF-MRF achieved the best generalization (R2_test = 0.9066; RMSE_test = 5.3644), followed by MLP-PADA (R2_test = 0.8953; RMSE_test = 5.7080) and FFN-TMM (R2_test = 0.8914; RMSE_test = 5.8126), with FFN-SAO slightly lower (R2_test = 0.8709). Additionally, a computational performance analysis was conducted to evaluate inference time, memory usage, energy consumption, and batch scalability across all models. Feature importance analysis was conducted, revealing that aging temperature, Cr, and aging duration were the most influential factors for hardness. In contrast, aging duration, aging temperature, solution treatment temperature, and Cu played key roles in electrical conductivity. The results demonstrate the effectiveness of these advanced machine learning models in predicting critical material properties, offering insightful advancements for materials science research. This study introduces the first controlled, statistically validated, multi-model benchmark that integrates composition and thermo-mechanical processing with deployment-grade profiling for property prediction of low-alloyed Cu alloys.
  • No Thumbnail Available
    Item
    Hybrid Stochastic–Machine Learning Framework for Postprandial Glucose Prediction in Type 1 Diabetes
    (2025-10-01) Naskinova I.; Kolev M.; Karova D.; Milev M.
    This research introduces a hybrid framework that integrates stochastic modeling and machine learning for predicting postprandial glucose levels in individuals with Type 1 Diabetes (T1D). The primary aim is to enhance the accuracy of glucose predictions by merging a biophysical Glucose–Insulin–Meal (GIM) model with advanced machine learning techniques. This framework is tailored to utilize the Kaggle BRIST1D dataset, which comprises real-world data from continuous glucose monitoring (CGM), insulin administration, and meal intake records. The methodology employs the GIM model as a physiological prior to generate simulated glucose and insulin trajectories, which are then utilized as input features for the machine learning (ML) component. For this component, the study leverages the Light Gradient Boosting Machine (LightGBM) due to its efficiency and strong performance with tabular data, while Long Short-Term Memory (LSTM) networks are applied to capture temporal dependencies. Additionally, Bayesian regression is integrated to assess prediction uncertainty. A key advancement of this research is the transition from a deterministic GIM formulation to a stochastic differential equation (SDE) framework, which allows the model to represent the probabilistic range of physiological responses and improves uncertainty management when working with real-world data. The findings reveal that this hybrid methodology enhances both the precision and applicability of glucose predictions by integrating the physiological insights of Glucose Interaction Models (GIM) with the flexibility of data-driven machine learning techniques to accommodate real-world variability. This innovative framework facilitates the creation of robust, transparent, and personalized decision-support systems aimed at improving diabetes management.
  • No Thumbnail Available
    Item
    Mathematical analysis of an autoimmune diseases model: Kinetic approach
    (2019-11-01) Kolev M.
    A new mathematical model of a general autoimmune disease is presented. Basic information about autoimmune diseases is given and illustrated with examples. The model is developed by using ideas from the kinetic theory describing individuals expressing certain functions. The modeled problem is formulated by ordinary and partial equations involving a variable for a functional state. Numerical results are presented and discussed from a medical view point.
  • No Thumbnail Available
    Item
    Mathematical modeling of autoimmune diseases
    (2020-09-01) Kolev M.
    The human organism is a very complex system. To be in good health, its components must function properly. One of the most important systems of an organism is the immune system. It protects the body from the harmful effects of various external and internal agents. Sometimes, however, the immune system starts attacking its own healthy cells, tissues and organs. Then autoimmune diseases arise. They are widespread in recent decades. There is evidence that often autoimmune responses occur due to viral infections. In this paper, a new mathematical model of a general autoimmune disease is proposed. It describes the interactions between viral particles and host cells. The model is formulated by using integro-differential equations of Boltzmann type. This approach is typical for the nonequilibrium statistical mechanics. A preliminary qualitative and quantitative analysis of the model is presented.
  • No Thumbnail Available
    Item
    Mathematical modelling of the competition between tumors and immune system considering the role of the antibodies
    (2003-06-01) Kolev M.
    A variety of immune system cells can recognize and destroy the autologous tumor cells. For the realization of the cytotoxic function of the effector cells, their close contact with the surface of the target cells is necessary. Such close contact can be realized not only through direct cell-cell recognition mechanism but also through some adaptor molecules (antibodies) binding with the effector cells and with the target cells. In the presented paper, this mechanism of the so-called humoral immunity is considered and described in a model of cellular tumor dynamics in competition with the immune system. The model is developed with statistical methods analogous to those of the kinetic theory. The model is expressed in terms of a system of integrodifferential equations. The role of the antibodies is taken into account. A numerical scheme for treating the system is proposed. The results of the computational simulations are presented and compared with some theoretical and experimental data. © 2003 Elsevier Science Ltd. All rights reserved.
  • No Thumbnail Available
    Item
    Numerical Solutions for a Model of Tissue Invasion and Migration of Tumour Cells
    (2011-01-01) Zubik-Kowal B.; Kolev M.
    The goal of this paper is to construct a new algorithm for the numerical simulations of the evolution of tumour invasion and metastasis. By means of mathematical model equations and their numerical solutions we investigate how cancer cells can produce and secrete matrix degradative enzymes, degrade extracellular matrix, and invade due to diffusion and haptotactic migration. For the numerical simulations of the interactions between the tumour cells and the surrounding tissue, we apply numerical approximations, which are spectrally accurate and based on small amounts of grid-points. Our numerical experiments illustrate the metastatic ability of tumour cells. Copyright © 2011 M. Kolev and B. Zubik-Kowal.
  • No Thumbnail Available
    Item
    The “GEnomics of Musculo Skeletal Traits TranslatiOnal NEtwork”: Origins, Rationale, Organization, and Prospects
    (2021-08-16) Koromani F.; Alonso N.; Alves I.; Brandi M.L.; Foessl I.; Formosa M.M.; Morgenstern M.F.; Karasik D.; Kolev M.; Makitie O.; Ntzani E.; Pietsch B.O.; Ohlsson C.; Rauner M.; Soe K.; Soldatovic I.; Teti A.; Valjevac A.; Rivadeneira F.
    Musculoskeletal research has been enriched in the past ten years with a great wealth of new discoveries arising from genome wide association studies (GWAS). In addition to the novel factors identified by GWAS, the advent of whole-genome and whole-exome sequencing efforts in family based studies has also identified new genes and pathways. However, the function and the mechanisms by which such genes influence clinical traits remain largely unknown. There is imperative need to bring multidisciplinary expertise together that will allow translating these genomic discoveries into useful clinical applications with the potential of improving patient care. Therefore “GEnomics of MusculoSkeletal traits TranslatiOnal NEtwork” (GEMSTONE) aims to set the ground for the: 1) functional characterization of discovered genes and pathways; 2) understanding of the correspondence between molecular and clinical assessments; and 3) implementation of novel methodological approaches. This research network is funded by The European Cooperation in Science and Technology (COST). GEMSTONE includes six working groups (WG), each with specific objectives: WG1-Study populations and expertise groups: creating, maintaining and updating an inventory of experts and resources (studies and datasets) participating in the network, helping to assemble focus groups defined by phenotype, functional and methodological expertise. WG2-Phenotyping: describe ways to decompose the phenotypes of the different functional studies into meaningful components that will aid the interpretation of identified biological pathways. WG3 Monogenic conditions - human KO models: makes an inventory of genes underlying musculoskeletal monogenic conditions that aids the assignment of genes to GWAS signals and prioritizing GWAS genes as candidates responsible for monogenic presentations, through biological plausibility. WG4 Functional investigations: creating a roadmap of genes and pathways to be prioritized for functional assessment in cell and organism models of the musculoskeletal system. WG5 Bioinformatics seeks the integration of the knowledge derived from the distinct efforts, with particular emphasis on systems biology and artificial intelligence applications. Finally, WG6 Translational outreach: makes a synopsis of the knowledge derived from the distinct efforts, allowing to prioritize factors within biological pathways, use refined disease trait definitions and/or improve study design of future investigations in a potential therapeutic context (e.g. clinical trials) for musculoskeletal diseases.