Towards Real-Time In-Implant Epileptic Seizure Prediction

Aziz, Joseph; Karakiewicz, R.; Genov, Roman; Bardakjian, Berj L.; Derchansky, M.; Carlen, Peter L.

doi:10.1109/iembs.2006.4398694

Cited by 4 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although there have been very few reported detection or prediction algorithms with hardware implementation on a custom circuit with power numbers, we compare our hardware power numbers to commonly used hardware cores in devices. For the purpose of this study, we compare our work to typical power consumption of a TIC320X DSP processor, TI320 Floating point DSP processor(TI320FP), integrated neural recording system (INI3) [31], VLSI implementation of a wavelet engine (DWT) [40], an analog filtering scheme presented for use with the Flint Hills Scientific (FHS) seizure-detection algorithm [41]. Typical active power consumption numbers were used wherever reported (figure 12) and technical specification datasheets were used to report numbers for the microprocessors and DSPs.…”

Section: Resultsmentioning

confidence: 99%

The design and hardware implementation of a low-power real-time seizure detection algorithm

et al. 2009

View full text Add to dashboard Cite

Epilepsy affects more than 1% of the world's population. Responsive neurostimulation is emerging as an alternative therapy for the 30% of the epileptic patient population that does not benefit from pharmacological treatment. Efficient seizure detection algorithms will enable closed-loop epilepsy prostheses by stimulating the epileptogenic focus within an early onset window. Critically, this is expected to reduce neuronal desensitization over time and lead to longer-term device efficacy. This work presents a novel event-based seizure detection algorithm along with a low-power digital circuit implementation. Hippocampal depth-electrode recordings from six kainate-treated rats are used to validate the algorithm and hardware performance in this preliminary study. The design process illustrates crucial trade-offs in translating mathematical models into hardware implementations and validates statistical optimizations made with empirical data analyses on results obtained using a real-time functioning hardware prototype. Using quantitatively predicted thresholds from the depth-electrode recordings, the auto-updating algorithm performs with an average sensitivity and selectivity of 95.3 +/- 0.02% and 88.9 +/- 0.01% (mean +/- SE(alpha = 0.05)), respectively, on untrained data with a detection delay of 8.5 s [5.97, 11.04] from electrographic onset. The hardware implementation is shown feasible using CMOS circuits consuming under 350 nW of power from a 250 mV supply voltage from simulations on the MIT 180 nm SOI process.

show abstract

Section: Resultsmentioning

confidence: 99%

The design and hardware implementation of a low-power real-time seizure detection algorithm

et al. 2009

View full text Add to dashboard Cite

show abstract

“…The detection performances are compared with published different types of seizure detectors based on wavelet artificial neural network (WANN) [Aziz et al 2006], events (ESD) [Raghunathan et al 2009], nonlinear Energy (NED) [Patel et al 2009], and spectral energy (SED) [Verma et al 2010] in Table IV. Results presented in ESD and NED are based on circuits simulation, but WANN and SED were fully integrated and results were based on animals, human and scalp EEG.…”

Section: Validation Of Seizure Detectionmentioning

confidence: 99%

Implantable Closed-Loop Epilepsy Prosthesis

Salam

Nguyen

2012

J. Emerg. Technol. Comput. Syst.

View full text Add to dashboard Cite

In this article, we present an implantable closed-loop epilepsy prosthesis, which is dedicated to automatically detect seizure onsets based on intracerebral electroencephalographic (icEEG) recordings from intracranial electrode contacts and provide an electrical stimulation feedback to the same contacts in order to disrupt these seizures. A novel epileptic seizure detector and a dedicated electrical stimulator were assembled together with common recording electrodes to complete the proposed prosthesis. The seizure detector was implemented in CMOS 0.18-μm by incorporating a new seizure detection algorithm that models timeamplitude and -frequency relationship in icEEG. The detector was validated offline on ten patients with refractory epilepsy and showed excellent performance for early detection of seizures. The electrical stimulator, used for suppressing the developing seizure, is composed of two biphasic channels and was assembled with embedded FPGA in a miniature PCB. The stimulator efficiency was evaluated on cadaveric animal brain tissue in an in vitro morphologic electrical model. Spatial characteristics of the voltage distribution in cortex were assessed in an attempt to identify optimal stimulation parameters required to affect the suspected epileptic focus. The experimental results suggest that lower frequency stimulation parameters cause significant amount of shunting of current through the cerebrospinal fluid; however higher frequency stimulation parameters produce effective spatial voltage distribution with lower stimulation charge.

show abstract

“…To avoid the dependence of quantization on scale and magnitude a signal is first normalized to the unit variance as in Equation 3. 13. In practice the bins are often allowed to overlap so that edge effects can be reduced.…”

Section: Entropy -How Random Is the System?mentioning

confidence: 99%

“…It was shown in Section 5.4 that this is already (sometimes) feasible for intra-cranial records with detection algorithms available today. Some research groups (e.g., [13]) decided that this direction is the most logical, but although preliminary results are promising whether the seizure can be aborted once it has started has not yet been adequately answered [111] (refer to Section 6.5.1).…”

Section: Parameters κ (Steady State) V Abmentioning

confidence: 99%