Software Department

Head of the Department – Prof. Marek Krętowski, D.Sc.

m.kretowski [at] pb.edu.pl

Research and scientific profile

1. SCIENTIFIC WORK
   Scientific research conducted at the Software Department is focused on several main areas:
   
   • knowledge discovery in databases (KDD) and Data Mining: techniques based on artificial neural networks, evolutionary algorithms, Bayesian networks and fuzzy sets;
   • parallel and distributed computing: use of GPU acceleration, computer cluster and grid computing technologies, especially applied in data mining and biomedical informatics;
   • applications of computer science in biomedicine (bioinformatics, medical informatics): methods of image analysis and modelling, medical decision support systems;
   • software engineering;
   • digital image processing and analysis (including biometrics).

The topics and scope of research of individual staff members include:

1. • data mining, in particular: classification, regression, feature selection based on CPL-type criterion function minimization techniques;
   • data mining (including genomics) and Big Data using, using various tools e.g., evolutionary algorithms and decision trees;
   • statistical, biostatistical and bioinformatic data analysis with a focus on genomic data mining resulting from Next Generation Sequencing experiments;
   • methods of analysis (both classical, statistical and exploratory) and data visualization;
   • applications of artificial intelligence in medicine, with particular emphasis on probabilistic graphical models;
   • construction of probabilistic models in medicine based on data and expert knowledge;
   • probabilistic graphical models; data analysis; discovering causality from data;
   • Big data analysis; flat pattern detection; knowledge extraction algorithms;
   • knowledge engineering;
   • decision support systems based on decision theory methods;
   • medical imaging;
   • classification of medical images based on textural feature analysis;
   • biomedical informatics; nuclear magnetic resonance; computer modeling; computer simulation; computer fluid mechanics;
   • biometric systems (automatic verification of human identity, techniques for automatic recognition of human identity), in particular methods of identification and verification of persons on the basis of behavioral characteristics;
   • parallel and distributed computing;
   • clustering algorithms; evolutionary algorithms;
   • software reuse, software production lines; identification of reusable components, similarity analysis in software families;
   • software clustering algorithms;
   • mobile applications supporting patient monitoring and psychotherapy;
   • hardware and software virtualization; technical and security aspects of cloud computing service;
   • geospatial data mining, including airborne laser scanning data (LiDAR) for environmental information;
2. SOME OF ONGOING RESEARCH PROJECTS
   1. international projects:
      • European Cooperation in Science & Technology (COST) project; Number: BM1304 (project title: Applications of MR imaging and spectroscopy techniques in neuromuscular disease: collaboration on outcome measures and pattern recognition for diagnostics and therapy development). Participation in management committee and WG4 working group (work topic: Explore strategies for muscle imaging texture analysis) (years: 2013 – 2017). Link: http://myo-mri.eu/working-group/explore-strategies.
      • ImPRESS project – International Interdisciplinary PhD studies in Biomedical Research and Biostatistics. Supporting the career and training in omic-based research and biostatistics by inter-national and-sectoral mobility. Project of the Medical University of Bialystok, in which the Bialystok University of Technology is a partner university. The project includes training in large-scale technology-based research and biostatistics and career support for young scientists through international and inter-sectoral mobility. The project is funded under the call of COFUNF 2016 Maria Skłodowska-Curie of the Horizon 2020 Programme. (years: 2018 – 2023). Link: http://cordis.europa.eu/project/rcn/210603_en.html
   2. national projects:
      • STRATEGMED MOBIT project funded by The National Centre for Research and Development (NCRD): “Creation of a reference model for Personalized Diagnostics of Cancer Tumors based on the analysis of tumor heterogeneity using genomic biomarkers, transcriptome and metabolome and PET/MRI imaging studies as a tool for implementing and monitoring individualized therapy”. Number: STRATEGMED2/266484/2/NCBR/2015  (years: 2015 – 2018). The main objective of the MOBIT project is to create a novel model of personalized diagnosis of cancerous tumors based on an innovative system of biobanking of biological material and large-scale omic analyses of patients with the most common malignancies.
      • project funded by NCRD under the Intelligent Development Operational Programme 2014 – 2020: “Sensor system in vehicles for post-accident condition recognition with information transmission to eCall reception point” (years: 2017 – 2020). The aim of the project is to develop methods to pre-determine the condition and number of injured people using smart sensors and transmitting them according to eCall standards to an emergency response center.
      • project funded by National Science Centre (NSC) (call: PRELUDIUM 5): “Evolutionary algorithms in global model tree induction”. Number: 2013/09/N/ST6/04083 (years: 2014 – 2017).
      • project co-financed by European Funds, under the Operational Programme Intelligent Development (measure 2.3 “Pro-innovative services for enterprises”, sub-measure 2.3.2 “Innovation vouchers for SMEs”): “Implementation of a system for automatic recognition of features of medical products from textual sources describing their properties” (years: 2017 – 2018).
      • project funded by NSC (call: OPUS 17): “Relative relation analysis in integrated omics data mining”, Number: 2019/33/B/ST6/02386. The aim of this study is to verify the hypothesis that analysis of integrated omics data using relationships and dependencies between characteristics of an individual patient can be successfully used for prognosis and diagnostic support (years: 2020 – 2023).
      • project funded by NSC (call: PRELUDIUM 13): “Image segmentation methods integrating texture analysis and deformable models” Number: 2017/25/N/ST6/01849. The aim of the work is to develop new segmentation methods, their validation and efficient implementation, allowing their practical application in the analysis of medical imaging (years: 2018 – 2023).  
      • commissioned work (Powszechna Kasa Oszczędności Bank Polski Spółka Akcyjna): “Audit model based on continuous risk assessment methodology and control mechanisms using statistical methods and artificial intelligence” (years: 2021 – 2023)
      • internal grants at Bialystok University of Technology
3. INTERNATIONAL COOPERATION
   • University of Rennes 1 (Université de Rennes 1), Signal and Image Processing Laboratory (Laboratoire Traitement du Signal et de l’Image – LTSI), National Institute for Health and Medical Research (Institut National de la Santé et de la Recherche Médicale – INSERM), Rennes, France (Prof. Johanne Bézy – Wendling). Collaboration within the Polonium program and working program “Biomedical image modeling and processing” (years: 2014 – 2019).
   • Nuclear Magnetic Resonance Laboratory (Laboratoire RMN – Résonance Magnétique Nucléaire), Institute of Myology, Hospital University Pitié-Salpêtrière (Institut de Myologie, Hôpital Universitaire Pitié-Salpêtrière), Paris, France (Prof. Jacques D. de Certaines; Dr. Noura Azzabou). Collaboration in the framework of the COST project (number: BM1304). Topics: Use of texture analysis in the diagnosis of stages of muscular dystrophy in Golden Retriever dogs, based on magnetic resonance images (years: 2013 – 2017).
   • Kharkiv State University of Radioelectronics (Харківський національний університет радіоелектроніки), Kharkiv, Ukraine. Working Program: “Data exploration based on methods of computational intelligence and their applications”. (years 2013 – 2018).
   • Collaboration with Prof. Francisco Javier Díez and Dr. Manuel Luque from the Department of Artificial Intelligence, Universidad Nacional de Educación a Distancia (UNED), Madrid, Spain. The collaboration is on modeling probabilistic graphical models with applications in medicine, (years: 2000 – present).
   • Collaboration with pathologists at Magee-Womens Hospital, Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, USA. Collaboration involves analysis of gynecologic pathology and breast cancer pathology data (years: 2007 – present).
   • Collaboration with Innovation Center for Computer Assisted Surgery, University of Leipzig, Leipzig, Germany. The collaboration concerns the construction of diagnostic Bayesian network models in the medical field, (years: 2015 – present).
4. TEACHING
   Members of the Department of Software teach classes mainly related to the process of software development, with particular emphasis on programming languages and system design techniques: both database, application and Internet systems. In addition, courses taught include parallel computing, operating systems, and statistical algorithms and models.
   Obligatory courses:
   
   • Analysis of Big Data
   • Analysis and testing of information systems
   • Databases
   • Data mining
   • Software engineering
   • Probabilistic methods and statistics
   • Data warehouse modeling
   • Business Applications Modeling and Implementation
   • Statistical Modeling
   • Software Development Tools
   • Parallel Computing
   • Basic tools of a scientist’s workshop
   • Fundamentals of programming
   • Object-oriented programming
   • Programming for data analysis
   • Statistical data analysis
   • Operating systems
   • Decision support systems
   • Distributed application development
   • Introduction to machine learning
   • Introduction to scientific research
   • Introduction to photography
   • Selected distributed systems
   • Selected programming techniques
   • Advanced software engineering
   • Advanced databases and data warehousing
   • Advanced operating systems
   • Advanced programming techniques

Elective courses:

• • Evolutionary Algorithms
  • Image Analysis and Processing
  • Technical and fundamental analysis
  • Biometrics – theory and applications
  • Building applications in WPF technology
  • Mathematics in biomedical engineering
  • Optimization methods
  • Optimization methods and linear programming
  • Computing with Graphics Accelerators
  • Fundamentals of biometrics
  • Business application programming based on the Java platform
  • Application programming in JavaScript
  • Object-oriented programming
  • Natural language processing
  • Framework solutions in WWW applications development
  • Bayesian networks
  • Expert systems
  • Decision support systems
  • Artificial neural networks
  • Database applications development
  • Deep learning
  • Introduction to Biomedical Informatics
  • Advanced programming techniques

5. COMPETITIONS organized (co-organized) by members of the Department of Software:
   • Olympiad in Informatics (http://oi.edu.pl/) is a programming competition addressed to high school students conducted annually under the aegis of the Ministry of National Education. The aim of the competition is to enable gifted students to develop their talents and deepen their knowledge of computer science. Participants must demonstrate skills in analyzing algorithmic problems, their specification, creating efficient algorithms and implementing them in high-level programming language, selecting appropriate data structures, testing programs, working in a programming environment.
   • Dżemik (http://dzemik.wi.pb.edu.pl/) – a computer game programming competition organized at the Faculty of Computer Science of the BUT for children and young people (primary schools and high schools).
6. Organization (or co-organization) of cyclic CONFERENCES and seminars:
   • TERW – Knowledge Exploration and Representation Technologies (http://irys.wi.pb.edu.pl/terw);
   • Scientific Workshops of PTSK – Polish Society for Computer Simulation (http://www.ptsk.pl/warsztaty/);
   • Statistics and Clinical Practice seminars organized within the framework of the International Center for Biocybernetics in Warsaw;
   • International conference Computer Information Systems and Industrial Management Applications (CISIM), indexed in Web of Science database.

Individual research grants

Deep learning-based viruses genomes vectorization as a method for multidimensional feature space extraction

Piotr Tynecki MSc, p.tynecki [at] doktoranci.pb.edu.pl 2020 – 2022