Dazong Jiang

Using modulus maximum pair of wavelet transform to detect spike wave of epileptic EEG

In this paper, the modulus maximum pair of wavelet transform is used to detect the spike wave of epileptic EEG signals by detecting their singular points for the first time. EEG signals are decomposed with a dyadic spline wavelet by Mallat algorithm. We can detect a spike wave by analyzing the relationship between the signal singularity, spike wave, and the modulus maximum pair of its wavelet transform. The data from 8 patients and 2 normal persons, totalling 754 spikes, is analyzed. Results show that the spike detection rate is 94.2%, and no false detection for normal EEG signals.

Bispectrum analysis of focal ischemic cerebral EEG signal

A model for SD rat focal ischemic cerebral injury is presented. Based on this experimental model, the authors recorded the EEGs of an ischemic region and a normal region for 30 minutes. Then the authors used a third-order spectrum (TOR) method for estimating the bispectrum of these EEG signals, and found that some characteristics of these bispectra which were very sensitive to focal ischemic cerebral injury. The maximum bispectrum magnitude of EEG and the weighted center of bispectrum (WCOB) change sensitively to different injury regions and injury stages. The bispectrum analysis results were verified by the Heat Shock Protein (HSP) test. It is concluded that the ischemic and normal regions can be clearly distinguished by analyzing their EEG bispectra, and the extent can also be detected by this way.

EEG complexity measurement of focal ischemic cerebral injury

Based on a new focal cerebral ischemia experimental model using SD rats, the authors analyzed EEG complexity. They found that EEG complexity is very sensitive to the extent of focal ischemic cerebral injury. In addition ischemic and normal regions can be distinguished by analyzing their EEG complexities.

Detecting nonlinearity in multichannel epileptic EEG

Distinguishing between determinism and noise has attracted many interests in nonlinear dynamical analysis of electroencephalograph (EEG). We apply the phase-randomized Fourier transform algorithm to generate surrogate data of multichannel EEG time series to detect nonlinearity in the epileptic EEG. For this purpose, fifty EEG segments from ten patients were analyzed, and the modified Grassberger and Procaccia algorithm (GPA) with multivariate embedding was used to calculate the discriminating statistic. The results indicate that epileptic EEGs exhibit nonlinearity with high confidence level, and the application of measures from nonlinear dynamics to epileptic EEG analysis appears reasonable.

Adaptive movable window enhancement in digital chest radiography

The global linear min-max windowing is the most common clinical method of contrast enhancement, but it is inadequate for chest images because of wide dynamic range. Here, a method of local contrast enhancement is described, which uses a movable window. This technique automatically calculates the local histogram over the movable window, the maximum and minimum of the gray-level window based on the local cumulative distribution function (CDF) without manual intervention.

Mammalian model study of selective activation of small nerve fibres using biphasic pulses with monopolar electrode

Biphasic pulse can cause lower electrochemical damage to nerve fibres than uniphasic pulse. A simulation system for studying the electrical properties of mammalian myelinated nerve fibres was built. By use of an asymmetric but charge-balanced biphasic stimulation waveform and monopolar point electrode, the sensitivity of excitation and blocking threshold of nerve fibres to fibre diameter was calculated. The results show that it may be possible to selectively activate smaller mammalian myelinated fibres by use of biphasic pulses.

A simulation study of selective stimuliktion of smaller nerve fibers by biphasic pulses

Based on the F-H model, a simulation system for studying the excitation properties of myelinated nerve fibers is built. For the purpose of decreasing electrochemical damage, two selective stimulation methods by use of biphasic pulses are performed on this model. They are single-electrode method and triple-electrode method. The simulation results predict that both of the two methods are effective. Some smaller nerve fibers can be selectively stimulated by biphasic pulses in certain conditions.

A modeling study of selective stimulation of nerve fibers using single cathode electrode

Electrical stimulation of unmyelinated nerve fibers was analyzed, using point sources in a simple volume conductor model and a dynamic model of Hodgkin-Huxley nerve fiber properties. The excitation and block threshold of single cathode stimulation with indifference anode at infinity were calculated by numerical method with axons of different diameters. The relation between the block threshold and the pulse width of single cathode stimulation was also calculated. The results predict a new method of selectively stimulating compound nerve trunks, which is simple and likely to be used clinically.

Nonlinear model of nerve fiber's electrical stimulation

The study of the excitation properties of electrically stimulated nerve fibers has an important significance in many research fields, such as neural science, rehabilitation engineering, etc. It is important to study the electrical field distribution arising from electrical stimulation. In previous work, this study was simplified by ignoring some nonlinear factors and was done using an infinite and homogeneous volume conductor model. In this paper, an analytic solution method of electrical field distribution arising from electrical stimulation is proposed, in which some nonlinear factors, such as connective tissue, anisotropy of nerve fibers, etc., are considered. The solution results of two electrode configurations, intrafascicular point electrode and extraneural ring electrode, are given. Based on the solution, a system for simulating electrical stimulated nerve fiber excitation properties is built combining the membrane dynamic model of mammalian nerve fiber. This work makes it possible to study electrical stimulated nerve fiber excitation properties closer to actual cases.

Selective stimulation of smaller nerve fibers using biphasic rectangular pulses.

Based on the F-H model, a simulation system to study the excitation properties of myelinated nerve fibers was developed. In order to minimize electrochemical damage of nerve tissue, three selective stimulation methods employing biphasic rectangular pulses were studied, which use single, double and triple electrodes, respectively. The simulation results prove that all of the three methods are effective in selective stimulation of smaller nerve fibers in a compound nerve trunk under certain conditions. The feasibility of these methods is also verified by animal experiments on the sciatic nerve trunks of toads. The amplitude and delay of compound action potential and the results of collision experiments proved that the methods are valid. These methods will be very useful in functional neuromuscular stimulation.