Comparison of Fuzzy Output Optimization with Expectation Maximization Algorithm and Its Modification for Epilepsy Classification

  • Sunil Kumar PrabhakarEmail author
  • Harikumar Rajaguru
Conference paper
Part of the Lecture Notes in Networks and Systems book series (LNNS, volume 14)


Due to the sudden and hyper excessive electrical discharges occurring in a specific group of cells in brain, a seizure is caused. The occurrence of the seizure can be in different regions of the brain. The person experiences different symptoms depending on the location of the seizure in the brain. So this disorder is usually witnessed and understood by seizures which are recurrent in nature. Due to epilepsy, abnormal behaviour rises such as the muscle movements becomes involuntary, the consciousness level of the patient becomes severely disturbed and it is accompanied by unusual perceptions. It is such a relatively common neurological condition that affects the person irrespective of age and sex. The diagnosis of the epilepsy can be confirmed with the help of Electroencephalography (EEG) signals. The EEG helps to get an accurate observation and description to know what has happened, the current circumstances and its respective severity levels. In this work, the main objective is to optimize the fuzzy output with the help of Expectation Maximization (EM) and Modified Expectation Maximization (MEM) algorithm for the classification and detection of epilepsy risk levels from EEG signals. At the beginning to classify the risk levels of epilepsy from the obtained signals which is based on the extracted parameters like variance, covariance, events, energy, sharp waves, spike waves and peaks, fuzzy techniques are incorporated. The EM and MEM algorithms are then effectively utilized on the initially classified data to get an exact risk level optimization value so that the risk of epilepsy in the patient is characterized easily.


Epilepsy EEG EM MEM Fuzzy 


  1. 1.
    Prabhakar SK, Rajaguru H (2015) An efficient classification of epilepsy risk levels from EEG signals using hard thresholding computation applied to code converters. In: Proceedings of IEEE student symposium in biomedical engineering and sciences (ISSBES), Universiti Teknologi Mara, MalaysiaGoogle Scholar
  2. 2.
    Prabhakar SK, Rajaguru H (2015) A different approach to epilepsy risk level classification utilizing various distance measures as post classifiers. In: Proceedings of the 8th biomedical engineering international conference (BMEiCON), Pattaya, ThailandGoogle Scholar
  3. 3.
    Shoeb A, Edwards H, Connolly J, Bourgeois B, Treves ST, John G (2004) Patientspecific seizure onset detection. Epilepsy Behav 5:483–498CrossRefGoogle Scholar
  4. 4.
    LeVan P, Urrestarazu E, Gotman J (2006) A system for automatic artifact removal in ictal scalp EEG based on independent component analysis and bayesian classification. Clin Neurophysiol 117(4):912–927CrossRefGoogle Scholar
  5. 5.
    Antonio Q, Batatia H, Pereyra M, Risk M (2014) Detection of onset in epilepsy signals using generalized gaussian distribution. In: Fifth international conference on advances in new technologies, interactive interfaces and communicability, ISBN: 978.88.96.471.37.1, doi:10.978.8896471/371
  6. 6.
    Valenti P, Cazamajou E, Scarpettini M, Aizemberg A, Silva W, Kochen S (2006) Automatic detection of interictal spikes using data mining models. J Neurosci Methods 150(1):105–110CrossRefGoogle Scholar
  7. 7.
    Siuly S, Li Y (2015) Designing a robust feature extraction method based on optimum allocation and principal component analysis for epileptic EEG signal classification. Comput Methods Programs Biomed 119:29–42CrossRefGoogle Scholar
  8. 8.
    Yusaf M, Nawaz R, Iqbal J (2016) Robust seizure detection in EEG using 2D DWT of time–frequency distributions. Electron Lett 52(11):902–903CrossRefGoogle Scholar
  9. 9.
    Wang F, Meng Q, Chen Y, Zhao Y (2013) Feature extraction method for epileptic seizure detection based on cluster coefficient distribution of complex network. WSEAA Trans Comput 13(2224–2872):351–360Google Scholar
  10. 10.
    Zhu G, Li Y, Wen P (2014) Epileptic seizure detection in EEGs signals using a fast weighted horizontal visibility algorithm. Comput. Methods Programs Biomed 115(2):64–75Google Scholar
  11. 11.
    Song Y (2011) A review of developments of EEG-based automatic medical support systems for epilepsy diagnosis and seizure detection. J Biomed Sci Eng 4:788–796CrossRefGoogle Scholar
  12. 12.
    Adlassnig KP (1986) Fuzzy set theory in medical diagnosis. IEEE Trans Syst Man Cybern 16(3):260–265CrossRefGoogle Scholar
  13. 13.
    Prabhakar SK, Rajaguru H (2015) ICA, LGE and FMI as dimensionality reduction techniques followed by GMM as post classifier for the classification of epilepsy risk levels from EEG Signals. In: 9th IEEE European modelling symposium, Madrid, Spain, 6–8 Oct, 978-1-5090-0206-1/15, doi: 10.1109/EMS.2015.20
  14. 14.
    Prabhakar SK, Rajaguru H (2015) GMM better than SRC for classifying epilepsy risk levels from EEG signals. In: Proceedings of the international conference on intelligent informatics and biomedical sciences (ICIIBMS), Okinawa, Japan, 28–30 NovGoogle Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2018

Authors and Affiliations

  1. 1.Department of ECEBannari Amman Institute of TechnologyCoimbatoreIndia

Personalised recommendations