• DocumentCode
    3044585
  • Title

    An integrated low-power asynchronous epileptic seizure detector

  • Author

    Mirzaei, Mohammad ; Salam, M. Tariqus ; Dang Khoa Nguyen ; Sawan, Mohamad

  • Author_Institution
    Polystim Neurotechnologies Lab., Polytech. Montreal, Montreal, QC, Canada
  • fYear
    2012
  • fDate
    28-30 Nov. 2012
  • Firstpage
    152
  • Lastpage
    155
  • Abstract
    In this paper, we present a new asynchronous seizure detector that is part of an implantable integrated device intended to identify onset seizure and trigger focal treatment to block seizure progression. The proposed system eliminates the unnecessary clock gating during normal neural activity monitoring mode and reduces the total power consumption. The proposed detector includes analog and digital building blocks with a variable time frame and four concurrent variable voltage window detectors to extract seizure onset information. The algorithm is validated in Matlab and the system is implemented in standard 0.13 μm CMOS process. Based on post-layout simulation results, the input referred noise of bioamplifier and the total power consumption of system are 4.4 μVrms and 7.1 μW, respectively. An accurate detection is achieved and no false alarms are recorded during analyzing the iEEG signals of two patients. The average detection delay of 9.5 sec is obtained after seizure onset.
  • Keywords
    CMOS integrated circuits; biomedical electronics; electroencephalography; low-power electronics; mathematics computing; medical signal processing; neurophysiology; patient monitoring; power consumption; signal denoising; Matlab; analog building blocks; average detection delay; bioamplifier; concurrent variable voltage window detectors; digital building blocks; electroencephalography; focal treatment; iEEG signals; implantable integrated device; input referred noise; integrated low-power asynchronous epileptic seizure detector; normal neural activity monitoring mode; post-layout simulation; seizure onset information extraction; seizure progression; size 0.13 mum; standard CMOS process; total power consumption; Algorithm design and analysis; Detectors; Epilepsy; Low pass filters; Noise; Power demand; Variable speed drives;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Biomedical Circuits and Systems Conference (BioCAS), 2012 IEEE
  • Conference_Location
    Hsinchu
  • Print_ISBN
    978-1-4673-2291-1
  • Electronic_ISBN
    978-1-4673-2292-8
  • Type

    conf

  • DOI
    10.1109/BioCAS.2012.6418397
  • Filename
    6418397