|
|
|
||
The course includes selected methods of the biomedical modeling and data analysis using relevant information systems. Custom theme includes a description of the acquisition and analysis of multichannel biomedical data and images with their subsequent modeling. The focus of the course is on mathematical data processing and well-researched assessment results. It allows students to unify view on the biological data processing by using computer technology and database systems.
Last update: TAJ445 (14.12.2013)
|
|
||
Students completing the course will be able to model basic biological processes at the single-cell level, interactions of cells group, organs and the whole organism. For the modeling of biological processes in space and time they will be able to process 1D, 2D and 3D signals. To test the response of the organism to external stimuli they will be able to prepare experiments on the Vernier equipment (for sensing temperature, pressure, ECG, ventilation parameters and muscle activity) and Walter (to capture visual and cognitive evoked potentials and EEG). Last update: VYSATAO (23.08.2013)
|
|
||
R: Reddy D.C.: Biomedical Signal Processing � Principles and Techniques, McGraw Hill, 2005,ISBN: 0070583889 A: Weitkunat R.: Digital Biosignal Processing, Elsevier, 1991, ISBN-10: 0444891447, ISBN-13: 978-0444891440 Z: Drongelen W., Signal Processing for Neuroscientists: An Introduction to the Analysis of Physiological Signals, Elsevier, 2007, ISBN-10: 0123708672 ISBN-13: 978-0123708670 A: Izhikevich E. M., Dynamical Systems in Neuroscience: The Geometry of Excitability and Bursting (Computational Neuroscience), The MIT Press, 2007, ISBN 0262090430, 9780262090438 Last update: TAJ445 (30.09.2013)
|
|
||
1. Overview of methods for modeling biological signals, modeling of biological systems control, homeostasis Exercise: Modeling regulation of skin temperature in different parts of the body (experiment No. 02 Vernier Labquest), methods of time series prediction 2. Displaying signals in time and frequency domain, phase portrait, Poincaré sections, recurrent views, types of signals - deterministic, stochastic, fractal and chaos, the calculation of basic signal characteristics, methods of noise removing Exercise: Modeling responses to respiratory breath arrest, rapid breathing and exercise (experiment No. 20 Vernier Labquest) 3. Chaos and dynamic analysis of biological signals. One-dimensional maps and flows, two-dimensional equilibrium, dissipative chaotic flow, Lyapunov exponents, Kaplan-York's dimension, reconstruction state space Exercise: Calculation of local Lyapunov exponents, Lyapunov numbers and global Lyapunov exponent for the logistic and sine map, estimation of Lyapunov exponents from experimental data 4. Formats of the biomedical data, "Universal Data Format for biosignals" (GDF, EDF), DICOM, proprietary formats, the biological signal and "Data mining" methods, object and relational databases Exercise: Convert EEG signals in EDF format into a matrix in MATLAB, create object-relational database in Access 5. Analytical and piecewise linear model ECG parameter estimation of normal and pathologic ECG. Compression and transmission of ECG Exercise: Capturing and analyzing ECG (experiment No. 12 Vernier Labquest), PL calculation model from the measured data using the Haar discrete wavelet transform, the calculation of the RMSE of the model display in the phase space and recurrent display of measured data 6. Modeling the electrical activity of neurons. Modeling alpha attenuation reaction and the rebound phenomenon, modeling rhythm following during photic stimulation through a network of chaotic neural oscillators. Modeling self-organization of coupled chaotic neural maps, modeling changes in EEG in dementia Exercise: Recording EEG signal on Walter device, rebound phenomenon, photic stimulation. Estimation based on the ratio of signal energy characteristic frequency of alpha activity and frequency photostimulation. 7. Modeling synchronization in EEG, estimates of global synchronization, anticipated synchronization and synchronization delay, phase synchronization. Discrete Hilbert transform estimate instantaneous phase estimate of the characteristic frequency. Exercise: Calculation of coherence, wavelet coherence, correlation of wavelet coefficients of wavelet coefficients of mutual information and global synchronization of occipital EEG leads with open and closed eyes 8. Detection, separation, localization, classification and modeling of evoked potentials and summation of muscle action potentials. Prony's method Exercise: Comparison of PCA, ICA, wavelet transform and modeling Prony's method for estimating of the development habituation, amplitude and latency of visual evoked potentials 9. Encoding information in visual and auditory analyzer, modeling communication in biomedical objects, Granger causality, spectral Granger causality, directional partial coherence , directional transfer function and corticomuscular coherence Exercise: Comparison of estimated synchronization delay, partial directional coherence and directional transfer function between EEG channels 10. Biostatistics, the most common errors in hypothesis testing in biomedical studies, statistical parametric mapping and Bonferroni correction methods used in epidemiological studies, hypothesis testing of the type "person at a time", Kaplan-Meier estimator, Weibull model, nonlinear statistics, Exercise: Testing the signal variance differences between EEG channels 19 and 19 segments in one channel, relationship to stationarity and correlation of signals, SPM with functional magnetic resonance 11. Analysis of texture in ultrasound diagnostics, segmentation, registration, visualization and simulation, Procrust registration method, cookurence histogram, Haralick's textural features Exercise: Textural segmentation of ultrasound images of various parts of the body 12. Three-dimensional segmentation, classification and modeling of two-dimensional images of magnetic resonance imaging Tutorial: 3-D view of the skeleton from spine MR images, 3-D view of the extent of ischemic zone - images of the brain computed tomography 13. Feature selection, biomedical data classification methods, decision making and expert systems in medicine Exercise: Automatic detection of lung tumors from lung computed tomography images 14. Advanced modeling in biology and physiology, the advantages and disadvantages Simulink, Modelica language, simulator QCP, QHP / Hummod, Golem. Exercise: Testing training simulators - ECGsim to simulate pathology, ECG heartsim to simulate the pressure profile in the heart, NEURON simulation of biological neurons and biological neural networks, AIDA for simulating the response of the organism diabetic insulin administration Last update: VYSATAO (23.08.2013)
|
|
||
none Last update: VYSATAO (23.08.2013)
|
Teaching methods | ||||
Activity | Credits | Hours | ||
Obhajoba individuálního projektu | 0.5 | 15 | ||
Účast na přednáškách | 1 | 28 | ||
Práce na individuálním projektu | 1 | 28 | ||
Účast na seminářích | 0.5 | 14 | ||
3 / 3 | 85 / 84 |
Coursework assessment | |
Form | Significance |
Regular attendance | 30 |
Defense of an individual project | 20 |
Examination test | 20 |
Continuous assessment of study performance and course -credit tests | 30 |