Automatic detection of interictal epileptiform discharges based on time-series sequence merging method

  • Authors:

  • Jian Zhang;Junzhong Zou;Min Wang;Lanlan Chen;Chunmei Wang;Guisong Wang

  • Affiliations:

  • Department of Automation, School of Information Science and Engineering, East China University of Science and Technology, Shanghai 200237, P.R. China;Department of Automation, School of Information Science and Engineering, East China University of Science and Technology, Shanghai 200237, P.R. China;Department of Automation, School of Information Science and Engineering, East China University of Science and Technology, Shanghai 200237, P.R. China;Department of Automation, School of Information Science and Engineering, East China University of Science and Technology, Shanghai 200237, P.R. China;Department of Electronic Engineering, Shanghai Normal University, Shanghai 200234, P.R. China;Department of Neurosurgery, Renji Hospital Affiliated to Shanghai Jiao Tong University, Shanghai 200233, P.R. China

  • Venue:
  • Neurocomputing
  • Year:
  • 2013

Quantified Score

Hi-index 0.01

Visualization

Abstract

This paper proposes a new automatic detection method of Interictal Epileptiform Discharges (IED) based on the merger of the increasing and decreasing sequences (MIDS) to improve IED detection rate. Firstly, increasing and decreasing sequences as well as complete and incomplete waves are reviewed to highlight the characteristics of clinical visual detection of IED. The sequence merging rules and algorithms are consequently proposed for time-domain electroencephalogram (EEG) signals. Experimental results demonstrate that the performance MIDS detection on rhythm waves and slow waves are very close to clinical visual detection. Secondly, the MIDS detection method is applied to IED fragments according to IED features in the time-domain. The results show that most IED fragments are recognized, although with some false detection of non-IED fragments. To reduce such false detection rate, Support Vector Machine (SVM) was applied with 17 characteristics and a training over 232 fragments from 3 patients' EEG recordings. With the SVM improvement, out-of-sample clinical EEG recordings of 32 suspected epilepsy patients were analyzed and 95.9% of the IED fragments marked by clinicians were successfully detected. The results show that the proposed algorithm performs well in IED detection and is a promising candidate in assisting clinicians' epilepsy diagnosis.