Unknown

The Use of Time-Frequency Distributions for Epileptic Seizure Detection in pneumoencephalogram Recordings below the guidance of Prof. Niladri Bihari Puhan Team Members Harish Ravilla(MT2010046) Siva Mulleti(MT2010142) Abstract The detection of recorded epileptic rapture activity in pneumoencephalogram segments is crucial for the hole and classification of epileptic seizures. However, since seizure phylogeny is typically a dynamic and non stationary process and the signals atomic number 18 composed of six-fold frequencies, visual and conventional frequency-based methods crap limited application. In this project, we experimented the suitability of the time-frequency (t-f) analysis to classify EEG signals for epileptic seizures, and we canvas ii methods of t-f analysis for EEG signals. Short-time Fourier transforms (STFT) and Pseudo Wigner-Ville t-f distributions are utilize to aspire the power spectrum density (PSD) of EEG signals. The analysis is performe d in troika stages: 1) t-f analysis and calculation of the PSD of each EEG segment. 2) hold extraction, step the signal segment fractional energy on specialized t-f windows. 3) Classification of the EEG segment (existence of epileptic seizure or not), utilize artificial neural networks.

The methods are evaluated using peerless classification problems obtained from a benchmark EEG dataset, and soft and numeric results are presented. Acknowledgements: We thank all the people who have helped and support us during the entire course of the project. We take the luck to educe special gratitude towards our adviso rs, Prof. Niladri Bihari Puhan for his encou! ragement and valuable guidance that helped us to crucify hurdles and keep ourselves motivated. Table of Contents I. INTRODUCTION------------------------------------------------------------------------------5 II. MATERIALS AND METHODS-------------------------------------------------------------7 A. Dataset B. Time-Frequency...If you compulsion to get a full essay, order it on our website: OrderCustomPaper.com

If you want to get a full essay, visit our page: write my paper