Baikun Wan

Infrared gait recognition based on wavelet transform and support vector machine

To detect human body and remove noises from complex background, illumination variations and objects, the infrared thermal imaging was applied to collect gait video and an infrared thermal gait database was established in this paper. Multi-variables gait feature was extracted according to a novel method combining integral model and simplified model. Also the wavelet transform, invariant moments and skeleton theory were used to extract gait features. The support vector machine was employed to classify gaits. This proposed method was applied to the infrared gait database and achieved 78%-91% for the probability of correct recognition. The recognition rates were insensitive for the items of holding ball and loading package. However, there was significant influence for the item of wearing heavy coat. The infrared thermal imaging was potential for better description of human body moving within image sequences.

Study on EEG-based mouse system by using brain-computer interface

This paper aimed to design an EEG-based mouse system by using brain-computer interface (BCI) to move a cursor on a computer display. This system to provide an alternative communication or control channel for patients with severe motor disabilities. Such patients might become able to select target on a computer monitor by moving a cursor through mental activity. The user could move the cursor just through imaging his/her hand operation on mouse without any actual action while the movement direction that he/she wanted to choose was lighted in the cue line of four-direction choice circulation. This system used an adaptive algorithm to recognize cursor control patterns in multichannel EEG frequency spectra. The algorithm included preprocessing, feature extraction, and classification. A Fisher ratio was defined to determine the characteristic frequency band. The spectral powering this band was calculated as feature parameter to distinguish the task state of imagination of hand movements (IHM) from free state of non-IHM. Mahalanobis distance classifier was employed to recognize the effective task pattern and produce the trigger signal as cursor controller. Relevant experiment results showed that this system achieved 80% accuracy for IHM task/free pattern classification. This EEG-based mouse system is feasible to drive the cursors four-direction movement and may provide a new communication and control option for patients with severe motor disabilities.

Study on a quantitative electroencephalography power spectrum typical of Chinese Han Alzheimer's disease patients by using wavelet transforms

Our objective was to investigate the quantitative electroencephalogram (EEG) power spectrum typical of Chinese Han ethnic Alzheimers disease (AD) patients. A study on the resting EEG was carried out on 103 local AD (NINCDS-ADRDA criteria) patients, and 124 age-matched normal elderly subjects served as controls. A novel multi-resolution decomposition algorithm based on Daubechies wavelet transform was employed for EEG spectral analysis. This algorithm decomposed recorded EEG signals into components with five frequency subbands, which especially provided more electroneural activity details in comparison with the conventional four subbands. A significant prevalence of an EEG spectrum characterized by increased slow activity with decreased fast activity was found in these patients. Moreover, the spectral power increase/decrease was mainly centralized in the below-2 Hz/over-8 Hz band, whereas the 2-8 Hz band did not show any widespread change. In conclusion, this study may provide some evidence of specific spectral changes of EEG affected by AD in China.

A new dynamometer walker system for the measurement of handle reaction vector (HRV)

The handle reaction vector (HRV) is commonly prescribed as an effective index for evaluating paraplegic walking efficiency assisted by functional electrical stimulation (FES). But up to now relevant studies on HRV have been limited because its direct measurement on walker handle is rather inconvenient. This paper reports the design of a new dynamometer walker to indirectly measure the HRV, which was based on 12 strain-gauge bridges instrumented on a walker frame. To adapt the data acquisition?analysis system of the dynamometer walker to quantitative and reliable measurement, a redundant-optimized technique was developed in instrumentation and calibration. The effect of its application was validated by the comparison of force measurement errors during the centrostigma shift of HRV. The measurement nonlinearity and crosstalk of the designed system were also investigated and were found to be better than 2.2% and 3.2%, respectively. One preliminary clinical trail was done to demonstrate the potential usefulness of this system in walking assessment. The results of the experiment and clinical trail showed that a practical dynamometer walker system was accomplished in this paper to measure the HRV without any impact on normal paraplegic walking training and might be employed further to guide the effective use of FES.

A gait stability investigation into FES-assisted paraplegic walking based on the walker tipping index

The gait outcome measures used in clinical trials of paraplegic locomotor training determine the effectiveness of improved walking function assisted by the functional electrical stimulation (FES) system. Focused on kinematic, kinetic or physiological changes of paraplegic patients, traditional methods cannot quantify the walking stability or identify the unstable factors of gait in real time. Up until now, the published studies on dynamic gait stability for the effective use of FES have been limited. In this paper, the walker tipping index (WTI) was used to analyze and process gait stability in FES-assisted paraplegic walking. The main instrument was a specialized walker dynamometer system based on a multi-channel strain-gauge bridge network fixed on the frame of the walker. This system collected force information for the handle reaction vector between the patients upper extremities and the walker during the walking process; the information was then converted into walker tipping index data, which is an evaluation indicator of the patients walking stability. To demonstrate the potential usefulness of WTI in gait analysis, a preliminary clinical trial was conducted with seven paraplegic patients who were undergoing FES-assisted walking training and seven normal control subjects. The gait stability levels were quantified for these patients under different stimulation patterns and controls under normal walking with knee-immobilization through WTI analysis. The results showed that the walking stability in the FES-assisted paraplegic group was worse than that in the control subject group, with the primary concern being in the anterior-posterior plane. This new technique is practical for distinguishing useful gait information from the viewpoint of stability, and may be further applied in FES-assisted paraplegic walking rehabilitation.

Use of a steady-state baseline to address evoked vs. oscillation models of visual evoked potential origin

There has been a long debate about the neural mechanism of event-related potentials (ERPs). Previously, no evidence or method was apparent to validate the two competing models, the evoked model and the oscillation model. One argument is whether the pre-stimulus brain oscillation could influence the following ERP. This study carried out an innovative visual oddball task experiment to investigate the dynamic process of visual evoked potentials. A period of stable oscillations of specified dominant frequencies and initial phases, i.e. the steady-state baseline, would be induced before responses to transient stimuli of different contrasts, which could overcome the artifact problem caused by the sorting method. The result first revealed a three-period-transition for the generation of visual evoked potentials by an objective decomposition. The ERP almost retained the preceding oscillation during the first period, provided an unstable negative potential in the second period, and generated the N1 component in the third period. The cross term analysis showed that the evoked model couldnt be the whole explanation for the ERP generation. Furthermore, the component analysis revealed that the N1 latency was sensitive to the initial phase under the low stimulus contrast (supporting the oscillation model) but not under the high stimulus contrast (supporting the evoked model). It demonstrated that the external stimulus contrast is a significant factor deciding the explicit model for ERPs. Our method and preliminary results may help reconcile the previous, seemly contradictory findings on the ERP mechanism.

Identification of humans using infrared gait recognition

Human body is a natural emitter of infrared ray. Usually the body temperature is different from that of surrounding. This leads to gray-scale difference between human body and background in infrared thermal image. So the method of infrared thermal imaging was presented to classify gaits. Firstly, infrared videos of 23 subjects were collected by using infrared thermal camera. Body silhouettes were extracted and normalized to obtain the gait feature image. Then the wavelet transform was combined with invariant moments to compute the moment parameters based on integral model. The gait feature image was simplified to extract the parameters based on the body skeleton. Finally, the parameters were applied to support vector machine for classification. This method achieved 71%∼92% for the probability of correct recognition. The results showed that it was insensitive for loading objects (backpack and volleyball) to recognize gaits in infrared video. Also it is easy to detect the moving human body during the whole day.

Feature recognition of multi-class imaginary movements in brain-computer interface

Feature recognition of multi-class imaginary movements is an important subject of brain-computer interface based on imaginary movement. In this paper, using the method of two-dimensional time-frequency analysis combined with Fisher separability analysis to study multi-channel synchronization, multi-class imaginary movements potential information of typical subjects. Also we have extracted the feature data of event related resynchronization/synchronization that could be used to identify different classes, and then use the support vector machine to establish classifiers, and have completed a higher accuracy rate of classification for multi-motor patterns. The result shows that the identification accuracy could basically satisfy the requirements of BCI systems under the circumstances that the subjects are better trained.