Sheng-Fu Liang

Support-vector-based fuzzy neural network for pattern classification

Fuzzy neural networks (FNNs) for pattern classification usually use the backpropagation or C-cluster type learning algorithms to learn the parameters of the fuzzy rules and membership functions from the training data. However, such kinds of learning algorithms usually cannot minimize the empirical risk (training error) and expected risk (testing error) simultaneously, and thus cannot reach a good classification performance in the testing phase. To tackle this drawback, a support-vector-based fuzzy neural network (SVFNN) is proposed for pattern classification in this paper. The SVFNN combines the superior classification power of support vector machine (SVM) in high dimensional data spaces and the efficient human-like reasoning of FNN in handling uncertainty information. A learning algorithm consisting of three learning phases is developed to construct the SVFNN and train its parameters. In the first phase, the fuzzy rules and membership functions are automatically determined by the clustering principle. In the second phase, the parameters of FNN are calculated by the SVM with the proposed adaptive fuzzy kernel function. In the third phase, the relevant fuzzy rules are selected by the proposed reducing fuzzy rule method. To investigate the effectiveness of the proposed SVFNN classification, it is applied to the Iris, Vehicle, Dna, Satimage, Ijcnn1 datasets from the UCI Repository, Statlog collection and IJCNN challenge 2001, respectively. Experimental results show that the proposed SVFNN for pattern classification can achieve good classification performance with drastically reduced number of fuzzy kernel functions.

Noninvasive Neural Prostheses Using Mobile and Wireless EEG

Neural prosthetic technologies have helped many patients by restoring vision, hearing, or movement and relieving chronic pain or neurological disorders. While most neural prosthetic systems to date have used invasive or implantable devices for patients with inoperative or malfunctioning external body parts or internal organs, a much larger population of ldquohealthyrdquo people who suffer episodic or progressive cognitive impairments in daily life can benefit from noninvasive neural prostheses. For example, reduced alertness, lack of attention, or poor decision-making during monotonous, routine tasks can have catastrophic consequences. This study proposes a noninvasive mobile prosthetic platform for continuously monitoring high-temporal resolution brain dynamics without requiring application of conductive gels on the scalp. The proposed system features dry microelectromechanical system electroencephalography sensors, low-power signal acquisition, amplification and digitization, wireless telemetry, online artifact cancellation, and signal processing. Its implications for neural prostheses are examined in two sample studies: 1) cognitive-state monitoring of participants performing realistic driving tasks in the virtual-reality-based dynamic driving simulator and 2) the neural correlates of motion sickness in driving. The experimental results of these studies provide new insights into the understanding of complex brain functions of participants actively performing ordinary tasks in natural body positions and situations within real operational environments.

Combination of EEG complexity and spectral analysis for epilepsy diagnosis and seizure detection

Approximately 1% of the worlds 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 patients 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.

Estimating driving performance based on EEG spectrum analysis

The growing number of traffic accidents in recent years has become a serious concern to society. Accidents caused by drivers drowsiness behind the steering wheel have a high fatality rate because of the marked decline in the drivers abilities of perception, recognition, and vehicle control abilities while sleepy. Preventing such accidents caused by drowsiness is highly desirable but requires techniques for continuously detecting, estimating, and predicting the level of alertness of drivers and delivering effective feedbacks to maintain their maximum performance. This paper proposes an EEG-based drowsiness estimation system that combines electroencephalogram (EEG) log subband power spectrum, correlation analysis, principal component analysis, and linear regression models to indirectly estimate drivers drowsiness level in a virtual-reality-based driving simulator. Our results demonstrated that it is feasible to accurately estimate quantitatively driving performance, expressed as deviation between the center of the vehicle and the center of the cruising lane, in a realistic driving simulator.

A Fully Integrated 8-Channel Closed-Loop Neural-Prosthetic CMOS SoC for Real-Time Epileptic Seizure Control

An 8-channel closed-loop neural-prosthetic SoC is presented for real-time intracranial EEG (iEEG) acquisition, seizure detection, and electrical stimulation in order to suppress epileptic seizures. The SoC is composed of eight energy-efficient analog front-end amplifiers (AFEAs), a 10-b delta-modulated SAR ADC (DMSAR ADC), a configurable bio-signal processor (BSP), and an adaptive high-voltage-tolerant stimulator. A wireless power-and-data transmission system is also embedded. By leveraging T-connected pseudo-resistors, the high-pass (low-pass) cutoff frequency of the AFEAs can be adjusted from 0.1 to 10 Hz (0.8 to 7 kHz). The noise-efficiency factor (NEF) of the AFEA is 1.77, and the DMSAR ADC achieves an ENOB of 9.57 bits. The BSP extracts the epileptic features from time-domain entropy and frequency spectrum for seizure detection. A constant 30- μA stimulus current is delivered by closed-loop control. The acquired signals are transmitted with on-off keying (OOK) modulation at 4 Mbps over the MedRadio band for monitoring. A multi-LDO topology is adopted to mitigate the interferences across different power domains. The proposed SoC is fabricated in 0.18- μm CMOS and occupies 13.47 mm2. Verified on Long Evans rats, the proposed SoC dissipates 2.8 mW and achieves high detection accuracy (> 92%) within 0.8 s.

Development of Wireless Brain Computer Interface With Embedded Multitask Scheduling and its Application on Real-Time Driver's Drowsiness Detection and Warning

Biomedical signal monitoring systems have been rapidly advanced with electronic and information technologies in recent years. However, most of the existing physiological signal monitoring systems can only record the signals without the capability of automatic analysis. In this paper, we proposed a novel brain-computer interface (BCI) system that can acquire and analyze electroencephalogram (EEG) signals in real-time to monitor human physiological as well as cognitive states, and, in turn, provide warning signals to the users when needed. The BCI system consists of a four-channel biosignal acquisition/amplification module, a wireless transmission module, a dual-core signal processing unit, and a host system for display and storage. The embedded dual-core processing system with multitask scheduling capability was proposed to acquire and process the input EEG signals in real time. In addition, the wireless transmission module, which eliminates the inconvenience of wiring, can be switched between radio frequency (RF) and Bluetooth according to the transmission distance. Finally, the real-time EEG-based drowsiness monitoring and warning algorithms were implemented and integrated into the system to close the loop of the BCI system. The practical online testing demonstrates the feasibility of using the proposed system with the ability of real-time processing, automatic analysis, and online warning feedback in real-world operation and living environments.

Automatic Stage Scoring of Single-Channel Sleep EEG by Using Multiscale Entropy and Autoregressive Models

In this paper, we propose an automatic sleep-scoring method combining multiscale entropy (MSE) and autoregressive (AR) models for single-channel EEG and to assess the performance of the method comparatively with manual scoring based on full polysomnograms. This is the first time that MSE has ever been applied to sleep scoring. All-night polysomnograms from 20 healthy individuals were scored using the Rechtschaffen and Kales rules. The developed method analyzed the EEG signals of C3-A2 for sleep staging. The results of automatic and manual scorings were compared on an epoch-by-epoch basis. A total of 8480 30-s sleep EEG epochs were measured and used for performance evaluation. The epoch-by-epoch comparison was made by classifying the EEG epochs into five states (Wake/REM/S1/S2/SWS) by the proposed method and manual scoring. The overall sensitivity and kappa coefficient of MSE alone are 76.9% and 0.65, respectively. Moreover, the overall sensitivity and kappa coefficient of our proposed method of integrating MSE, AR models, and a smoothing process can reach the sensitivity level of 88.1% and 0.81, respectively. Our results show that MSE is a useful and representative feature for sleep staging. It has high accuracy and good home-care applicability because a single EEG channel is used for sleep staging.

Using novel MEMS EEG sensors in detecting drowsiness application

Electroencephalographic (EEG) analysis has been widely adopted for the monitoring of cognitive state changes and sleep stages because abundant information in EEG recording reflects changes in drowsiness, arousal, sleep, and attention, etc. In this study, micro-electro-mechanical systems (MEMS) based silicon spiked electrode array, namely dry electrodes, are fabricated and characterized to bring EEG monitoring to the operational workplaces without requiring conductive paste or scalp preparation. An isotropic/anisotropic reactive ion etching with inductive coupled plasma (RIE-ICP) micromachining fabrication process was developed to manufacture the needle-like micro probes to pierce the stratum corneum of skin and obtain superior electrically conducting characteristics. This article reports a series of prosperity testing and evaluations of continuous EEG recordings. Our results suggest that the dry electrodes have advantages in electrode-skin interface impedance, signal intensity and size over the conventional (wet) electrodes. In addition, we also developed an EEG-based drowsiness estimation system that consists of the dry-electrode array, power spectrum estimation, principal component analysis (PCA)-based EEG signal analysis, and multivariate linear regression to estimate driverpsilas drowsiness level in a virtual-reality-based dynamic driving simulator to demonstrate the potential applications of the MEMS electrodes in operational environments.

A robust digital image stabilization technique based on inverse triangle method and background detection

In this paper, a novel robust digital image stabilization (DlS) technique is proposed to remove the unwanted shaking phenomena in the image sequences captured by hand-held camcorders without affecting the moving objects in the image sequence and intentional motion of panning condition, etc. It consists of a motion estimation unit and a motion compensation unit. To increase the robustness in the adverse image conditions, an inverse triangle method is proposed to extract reliable motion vectors in plain images which are lack of features or contain large low-contrast area, etc., and a background evaluation model is developed to deal with irregular images which contain large moving objects, etc. In the motion compensation unit, a CMV estimation method with an inner feedback-loop integrator is proposed to remove the unwanted shaking phenomena without losing the effective area of the image in panning condition. We also propose a smoothness index (Sl) to quantitatively evaluate the performances of different image stabilization methods. The experimental results are on-line available to demonstrate the remarkable performance of the proposed DIS technique.

EEG-based subject- and session-independent drowsiness detection: an unsupervised approach

Monitoring and prediction of changes in the human cognitive states, such as alertness and drowsiness, using physiological signals are very important for drivers safety. Typically, physiological studies on real-time detection of drowsiness usually use the same model for all subjects. However, the relatively large individual variability in EEG dynamics relating to loss of alertness implies that for many subjects, group statistics may not be useful to accurately predict changes in cognitive states. Researchers have attempted to build subject-dependent models based on his/her pilot data to account for individual variability. Such approaches cannot account for the cross-session variability in EEG dynamics, which may cause problems due to various reasons including electrode displacements, environmental noises, and skin-electrode impedance. Hence, we propose an unsupervised subject- and session-independent approach for detection departure from alertness in this study. Experimental results showed that the EEG power in the alpha-band (as well as in the theta-band) is highly correlated with changes in the subjects cognitive state with respect to drowsiness as reflected through his driving performance. This approach being an unsupervised and session-independent one could be used to develop a useful system for noninvasive monitoring of the cognitive state of human operators in attention-critical settings.