16 May 2010 in EURASIP Journal on Advances in Signal Processing
Approximately 1% of the world's population has epilepsy, and 25% of epilepsy patients cannot be treated sufficiently by any available therapy. If an automatic seizure-detection system was available, it could reduce the time required by a neurologist to perform an off-line diagnosis by reviewing electroencephalogram (EEG) data. It could produce an on-line warning signal to alert healthcare professionals or to drive a treatment device such as an electrical stimulator to enhance the patient's safety and quality of life. This paper describes a systematic evaluation of current approaches to seizure detection in the literature. This evaluation was then used to suggest a reliable, practical epilepsy detection method. The combination of complexity analysis and spectrum analysis on an EEG can perform robust evaluations on the collected data. Principle component analysis (PCA) and genetic algorithms (GAs) were applied to various linear and nonlinear methods. The best linear models resulted from using all of the features without other processing. For the nonlinear models, applying PCA for feature reduction provided better results than applying GAs. The feasibility of executing the proposed methods on a personal computer for on-line processing was also demonstrated.
24 September 2018 in IEEE Journal of Solid-State Circuits
A 16-channel closed-loop neuromodulation system-on-chip (SoC) for human epileptic seizure control is proposed and designed. In the proposed SoC, a 16-channel neural-signal acquisition unit (NSAU), a biosignal processor (BSP), a 16-channel high-voltage-tolerant stimulator (HVTS), and wireless power and bidirectional data telemetry are designed. In the NSAU, the input protection circuit is used to prevent MOSFET from overstressing by the high-voltage stimulations. Hence, NSAUs can share electrodes with stimulators. The auto-reset chopper-stabilized capacitive-coupled instrumentation amplifiers (AR-CSCCIAs) are designed with the chopper-stabilized technique with a new offset reduction loop. The measured input-referred noise is 2.09 $\mu \textrm {V}_{\mathrm {rms}}$ and the noise-efficiency factor (NEF) is 3.78. The entropy-and-spectrum seizure detection algorithm is implemented in the BSP with 0.76-s seizure detection latency and 97.8% detection accuracy. When the seizure onset is detected by the BSP, the HVTS with adaptive supply control delivers 0.5–3-mA biphasic current stimulation to suppress the seizure onset. The proposed SoC is powered wirelessly, and the bidirectional data telemetry is realized through the same pair of coils in 13.56 MHz. The downlink data rate is 211 Kb/s with the binary phase-shift keying (BPSK) modulation and a new BPSK demodulator. The uplink data rate is 106 Kb/s with the load-shift keying (LSK) modulation. The proposed SoC is fabricated in a 0.18- $\mu \text{m}$ CMOS technology and occupies 25 mm 2 . Electrical tests have been performed to characterize the SoC performance. In vivo animal experiments using mini-pigs have been performed to successfully verify the closed-loop neuromodulation functions on epileptic seizure suppression.
1 March 2012 in Journal of Neuroscience Methods
21 August 2012 in BioMedical Engineering OnLine
Background: Approximately one-third of the human lifespan is spent sleeping. To diagnose sleep problems, all-night polysomnographic (PSG) recordings including electroencephalograms (EEGs), electrooculograms (EOGs) and electromyograms (EMGs), are usually acquired from the patient and scored by a well-trained expert according to Rechtschaffen & Kales (R&K) rules. Visual sleep scoring is a time-consuming and subjective process. Therefore, the development of an automatic sleep scoring method is desirable. Method: The EEG, EOG and EMG signals from twenty subjects were measured. In addition to selecting sleep characteristics based on the 1968 R&K rules, features utilized in other research were collected. Thirteen features were utilized including temporal and spectrum analyses of the EEG, EOG and EMG signals, and a total of 158 hours of sleep data were recorded. Ten subjects were used to train the Discrete Hidden Markov Model (DHMM), and the remaining ten were tested by the trained DHMM for recognition. Furthermore, the 2-fold cross validation was performed during this experiment. Results: Overall agreement between the expert and the results presented is 85.29%. With the exception of S1, the sensitivities of each stage were more than 81%. The most accurate stage was SWS (94.9%), and the least-accurately classified stage was S1 (<34%). In the majority of cases, S1 was classified as Wake (21%), S2 (33%) or REM sleep (12%), consistent with previous studies. However, the total time of S1 in the 20 all-night sleep recordings was less than 4%. Conclusion: The results of the experiments demonstrate that the proposed method significantly enhances the recognition rate when compared with prior studies.
13 November 2017 in BioMedical Engineering OnLine
Effect of neurofeedback training (NFT) on enhancement of cognitive function or amelioration of clinical symptoms is inconclusive. The trainability of brain rhythm using a neurofeedback system is uncertainty because various experimental designs are used in previous studies. The current study aimed to develop a portable wireless NFT system for alpha rhythm and to validate effect of the NFT system on memory with a sham-controlled group. The proposed system contained an EEG signal analysis device and a smartphone with wireless Bluetooth low-energy technology. Instantaneous 1-s EEG power and contiguous 5-min EEG power throughout the training were developed as feedback information. The training performance and its progression were kept to boost usability of our device. Participants were blinded and randomly assigned into either the control group receiving random 4-Hz power or Alpha group receiving 8–12-Hz power. Working memory and episodic memory were assessed by the backward digital span task and word-pair task, respectively. The portable neurofeedback system had advantages of a tiny size and long-term recording and demonstrated trainability of alpha rhythm in terms of significant increase of power and duration of 8–12 Hz. Moreover, accuracies of the backward digital span task and word-pair task showed significant enhancement in the Alpha group after training compared to the control group. Our tiny portable device demonstrated success trainability of alpha rhythm and enhanced two kinds of memories. The present study suggest that the portable neurofeedback system provides an alternative intervention for memory enhancement.
20 July 2011 in Journal of Neural Engineering
In this paper numerous alternative treatments in addition to pharmacological therapy are proposed for their use in epileptic patients. Epileptic animal models can play a crucial role in the performance evaluation of new therapeutic techniques. The objective of this research is to first develop various epileptic rat models; second, develop a portable wireless closed-loop seizure controller including on-line seizure detection and real-time electrical stimulation for seizure elimination; and third, apply the developed seizure controller to the animal models to perform on-line seizure elimination. The closed-loop seizure controller was applied to three Long-Evans rats with spontaneous spike-wave discharges (non-convulsive) and three Long-Evans rats with epileptiform activities induced by pentylenetetrazol (PTZ) injection (convulsive) for evaluation. The seizure detection accuracy is greater than 92% (up to 99%), and averaged seizure detection latency is less than 0.6 s for both spontaneous non-convulsive and PTZ-induced convulsive seizures. The average false stimulation rate is 3.1%. Near 30% of PTZ-induced convulsive seizures need more than two times of 0.5 s electrical stimulation for suppression and 90% of the non-convulsive seizures can be suppressed by only one 0.5 s electrical stimulation.