Search
Biodat
Department of Cybernetics, Karlovo náměstí 13, 121 35 Prague 2
Tel: +420 224 357 325
http://bio.felk.cvut.cz/
Who we are
Lenka Lhotská
Head
Members:
Vladana Djordjevič, Eva Krajčovičová, Monika Martinková, Miroslav Burša, Václav Gerla, Michal Huptych, Václav Chudáček, Václav Křemen, Jakub Kužílek, Martin Macaš, David Macků, Jiří Spilka, David Steiner, Michal Vavrečka
PhD students:
Martina Šrutová, Karla Štěpánová, Radim Bělobrádek, Honza Hlúbik, Martin Holub, Filip Ježek
Research description
The main strength of the BioDat Group is in combination of classical signal processing methods, advanced artificial intelligence methods and general comprehension of the needs of the every-day clinical needs. We aim at both advancing the theoretical foundations of the field and design and development of biomedical applications. The key research topics include:
- Filtration and segmentation of non-stationary biomedical signals, e.g. EEG
- Signal analysis in time and frequency domains
- Design and development of methods for ECG processing
- Visualization tools for biomedical data processing
- Methods for feature extraction from the signals and large data volumes
- Data mining and machine learning methods applicable to biomedical domain
- Advanced nature-inspired optimization algorithms
- Decision support systems for medicine
- Cognitive modeling
What is it good for
Computer assisted processing of biological signals is gaining a growing importance. Some of the aims of computer assisted processing are to simplify tedious and time consuming work of doctors, make the evaluation more objective, and visualize results and represent them in a convenient form. The automatic systems cannot fully replace a physician but they are to make his/her work more efficient. For example, they identify segments of the signal where there are deviations from standard activity and in this way they shorten the time required for visual inspection of the whole recording.
Projects
ECG + CTG
Cardiotocography evaluation by means of artificial intelligence
The main goal of this project is to research the possible ways of bringing automatic preprocessing to the CTG evaluation for possible future full-fledged obstetrician's decision support system. The subgoals of the project are:
- Collection of the database with proper size and proper structure
- Development of preprocessing tools for CTG signal preprocessing
- Development of new GUI for semi-automatic CTG evaluation
- Expert-system like approach to the CTG evaluation using clinical data from the HIS.
Mobile Cardiotocography within the MAS project
We are responsible for software development in several parts of the project:
- Automatic detection of the best site for phonocardiography microphone placement on the abdomen.
- Development of the Android application for the "mobile" processing of the CTG signal. Transfer of the signal is also big part of this task.
- Development of the clinical application for the obstetrician. The application should serve as light decision support system for the clinicans. It should also allow responses of the clinician to the patient at home.
Myocardial Infarction
The aim of this project is to automatically detect the morphological changes of ECG caused by myocardial ischemia/infarction and diagnose where and when these changes occurred, i.e. time and location. The automatic classification will serve as decision support for doctors and help them with diagnosis and assesment of ECG.
EEG
Spatial navigation and orientation
The research is focused on the navigation in the virtual tunnel task and its EEG correlates. We searched for the features in the EEG signal to discriminate the employment of the allocentric and the egocentric reference frames. These two reference frames differ in the center of deixis (the origin of the coordinating system).
Sleep and neonatal EEG
Sleep problems belong to the most common serious neurological disorders. Reliable and robust detection of these disorders would improve the quality of life of many people. The aims of automated processing of sleep data are on one side to ease the work of medical doctors and on the other side to make the evaluation more objective.
We are also developing methods for differentiation between three important neonatal behavioral states: quiet sleep, active sleep and wakefulness, both in pre-term and full-term newborns. The developed algorithms are tested on real neonatal data. Obtained results can be used as a reference for developing and enhancing neonatal sleep EEG/PSG classification algorithms.
PSGLab Matlab Toolbox
PSGLab is a Matlab toolbox for processing of polysomnographic (PSG) data. PSG recording encompasses a set of heterogeneous biological signals recorded simultaneously. Electroencephalographic (EEG) signals, electrooculogram (EOG) and electromyogram (EMG) are important parts of this kind of recording. PSG recording may also include electrocardiogram (ECG), respiratory effort and respiratory airflow, blood oxygen saturation and temperature, as well as movement or body position.
Other
Immunogenetics
The goal of this project is to better understand HLA system, improve selection and transplantation process of unrelated stem cell donors and effectiveness of registries by new ICT technologies.
Blind Source Separation
Independent Component Analysis (ICA) is very common signal processing method in many different areas. Our aim is to use ICA in processing of biomedical signals mainly on ECG and EEG.
Nature Inspired Systems
Many advances in the computer sciences have been based on the observation and emulation of processes of the natural world. The origins of bioinspired informatics can be traced to the development of perceptrons and artificial life, which tried to reproduce the mental processes of the brain and biogenesis respectively, in a computer environment.
Financial support and contracts
- Grant Agency of Czech Republic
- Internal Grant Agency Ministry of Health Czech Republic
- ENIAC
- Mindmetic Ltd.
- FRVŠ
International collaboration
- KIT Karlsruhe
- Siemens Austria
- Department of Physiology, 1st Faculty of Medicine, Charles University
- Faculty Hospital Brno
- Faculty Hospital Prague
- Faculty Hospital Na Bulovce
- VFN Karlovo náměstí
- IKEM
- Prague Psychiatric Centre
- Medical Technologies CZ a. s.
- MeiCogSci, Comenius University,Bratislava
Selected publications
- Spilka, J. - Chudáček, V. - Koucký M. - Lhotská L. - Huptych M. - Janků P. - Georgoulas G. - Stylios C. (2011) Using nonlinear features for fetal heart rate classification In: Biomedical Signal Processing and Control. 2011, Article in Press
- Chudáček, V., Spilka, J. - Janků P. - Koucký M. - Lhotská L. - Huptych M. (2011) Automatic evaluation of intrapartum fetal heart rate recordings: a comprehensive analysis of useful features In: Physiological Measurement. 2011, vol. 32, no. 7, p. 1347-1360
- Kos, P. - Varga, F. - Handl, M. - Kautzner, J. - Chudáček, V. - et al. (2011) Correlation of dynamic impact testing, histopathology and visual macroscopic assessment in human osteoarthritic cartilage In: International Orthopaedics 2011, vol. 35, no. 1, p. 1-7. ISSN 0341-2695.
- Vavrečka, M., (2009). The neural correlates of spatial reference frames processing, Cognitive processing, 10:2, Springer Berlin, p. 342-345.
- Huptych, M. - Lhotská, L. (2009) ECG Beat Classification Using Feature Extraction from Wavelet Packets of R Wave Window, In: World Congress on Medical Physics and Biomedical Engineering, September 7 - 12, 2009, Munich, Germany [CD-ROM]. Berlin: Springer Science+Business Media , 2009, ISBN 978-3-642-03881-5.
- Gerla, V. - Lhotská, L. - Krajča, V. (2009): Multivariate Analysis of Full-Term Neonatal Polysomnographic Data. IEEE Transactions on Information Technology in Biomedicine, vol. 13, no. 1, p. 104-110. ISSN 1089-7771.
- Djordjevic, V. - Gerla, V. - Lhotska, L. - Krajca, V. - Paul, K. (2009) Improvements in Processing of Neonatal Electroencephalographic Recordings In: World Congress on Medical Physics and Biomedical Engineering, September 7 - 12, 2009, Munich, Germany [CD-ROM]. Berlin: Springer Science+Business Media , 2009, ISBN 978-3-642-03881-5.
- Macaš, M. - Lhotská, L.: Social Impact and Optimization. International Journal of Computational Intelligence Research. 2008, vol. 4, no. 2, p. 129-134. ISSN 0973-1873.
- Macaš, M. - Lhotská, L. - Gabrys, B. - Ruta, D.: Particle Swarm Optimization of Multiple Classifier Systems. In Computational and Ambient Intelligence. Heidelberg: Springer, 2007, p. 333-340. ISBN 978-3-540-73006-4.
- Kužílek, J. - Lhotská, L. - Hanuliak, M.: Processing Holter ECG signal corrupted with noise: Using ICA for QRS complex detection. In Conference Proceedings of The 3rd International Symposium on Applied Sciences in Biomedical and Communication Technologies. Rome: University of Rome "Tor Vergata", 2010, p. 1-4. ISBN 978-1-4244-8131-6.