Loading...

Using Bump Modeling in Brain Wave Analysis

Ghanbari Garakani, Zahra | 2011

884 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 41595 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Shamsollahi, Mohammad Bagher
  7. Abstract:
  8. In this thesis, the efficiency of bump modeling has been investigated on brain signals, in a variety of aspects including analysis, detection, classification and prediction. The aim of bump modeling is to provide an optimized representation of the signal in time-frequency domain. This would be done by discriminating oscillatory bursts from background signal and then showing them by half-ellipsoid functions called bump. Consequently, the problem of dealing with large numbers of parameters and hence complicated calculations, which are serious concerns in similar methods, can be overcome. This is in addition to the benefits of using time-frequency representation of the signal.The aim of bump modeling is to approximate a time-frequency map in terms of sum of parametric functions (bumps) in order to achieve a simple and optimized model. The adaptation of bump parameters is done so that the modeling error is minimized. There are 5 years passed from introducing bump modeling. Using bump modeling has considerable benefits rather than similar methods. However before this study, it had been used particularly by the group who introduced it and except two cases, it had been used just on Alzheimer's data bases.In the present thesis, bump modeling is used for a 4-class classification of MEG signals. Additionally it is used in order to analysis and detection of sleep spindle and K-complex patterns in sleep EEG and terminated in satisfactory results. In the next step, the similarity measure SES, which is based on bump modeling, is applied to epileptic EEG and its robustness in separating ictal and interictal intervals has been compared with four other similarity measures including linear correlation, dynamic similarity, fuzzy similarity and dissimilarity measure. Obtaining satisfactory results in discriminating ictal and inteictal intervals encouraged us to use it for seizure prediction. The primary results make many hopes of the on-line usage of this measure and to warn the patient based on early prediction.

  9. Keywords:
  10. Epilepsy ; Classification ; Prediction ; Magnetoencephalography (MEG) ; Bump Modeling ; Sleep Spindles Pattern ; K-Complex Pattern

 Digital Object List

 Bookmark