Hangzhou, China
Hangzhou, China

Time filter

Source Type

Dong L.,Hangzhou Diana University | Li S.,Hangzhou Diana University | Chen J.,Hangzhou Diana University | Yan H.,Toshiba Corporation | And 3 more authors.
Chinese Journal of Sensors and Actuators | Year: 2010

Optimizing the structure of a sensing capacitance can improve the performances of MEMS accelerometer. The characteristics of MEMS capacitive accelerometers with comb, bar and comb-bar capacitances, respectively, have been analyzed. The proof mass, air damping and damping ratio of the three different sensor structures are compared on the condition of same dimension, elastic constant of the spring, thickness of the proof mass, distance between the proof mass and the substrate. The results show that the proof mass of bar capacitance sensor is largest, and its air damping is least, which is suitable for high-precise sensor. The sensitivity of comb capacitance sensor in air is increased while the air damping is increased too, so it is suitable for high-sensitivity and low-precise sensor. The comb-bar capacitance is suitable for the sensors which need both high sensitivity and precision. An example is taken to prove the results.


Sun L.,Hangzhou Diana University | Xu W.-D.,Hangzhou Diana University | Li L.-H.,Hangzhou Diana University | Liu W.,Hangzhou Diana University | And 2 more authors.
Chinese Journal of Biomedical Engineering | Year: 2011

Classification of masses in mammography is an important part of Computer-Aided Diagnosis (CAD). How to improve the accuracy and stability of the classification is the focus of the current studies. Based on information fusions of multi-view and multi-classifier, four classification models on masses in mammograms are proposed. The first classification model uses multi-classifier fusion in single-view; In the second model, the outputs of each classifier in two views are fused, and then these results of multi-view are used for multiclassifier fusion; In the third model, fusion of multi-classifier is applied in each view, then the two fusion results are used for multi-view fusion; In the fourth model, feature vectors of two views are fused, then classification of singe-classifier and multi-classifier fusion are used. In the experiments, we randomly selected 148 benign masses and 148 malignant masses from the DDSM database. The experiment results revealed that the second and third models are superior to the first and fourth models in terms of accuracy, sensitivity, specificity and stability.


Yang H.-L.,Hangzhou Diana University | Zhu L.,Hangzhou Diana University | Han B.,Hangzhou Diana University | Li L.-H.,Hangzhou Diana University | And 2 more authors.
Chinese Journal of Biomedical Engineering | Year: 2013

To analysis high throughput and high resolution mass spectrometry data effectively and capture the cancer related protein feature from the mass spectrometry data, diagnosis called a feature selection based on affinity propagation clustering of mass spectrometry was proposed in this paper. Firstly, the t-test was used on mass spectrometry data, followed by feature selection based on affinity propagation clustering. Next, affinity propagtion clustering and NS-LDA was used for reducing dimensions and correlation. Thirdly, SVM-RFE was used to select the features. Finally, we used four classifiers to estimate the performance of the algorithm. The proposed method was tested and evaluated on the ovarian cancer database OC-WCX2a, OC-WCX2b, and breast cancer database BC-WCX2a. Classification was achieved 96.43%, 99.66% and 90.88%, sensitivity was achieved 97.00%, 100% and 96.17%, specificity was achieved 95.85%, 99.08% and 81.92%, respectively. And 10 m/z features were selected for each dataset. The experimental results showed good performance of the method, and the method is expected to be used in cancer diagnosis.


Xu W.,Hangzhou Diana University | Meng M.,Hangzhou Diana University | Ma Y.,Hangzhou Diana University
Chinese Journal of Sensors and Actuators | Year: 2010

Surface electromyography(SEMG)was a nonlinear and nonstationary signal. In order to get effective features of the SEMG and analyze features, HHT method was applied. The signal could be decomposed by using the empirical mode decomposition( EMD) method into a series of intrinsic mode function( IMF) components, at the same time, the frequency of each IMF was analyzed, these IMFs were transformed into 3-D Hubert spectra which exhibited the time-frequency-amplitude, and then the marginal spectra were obtained by integrating the Hubert spectra with respect to time. A wavelet threshold filtering method based on EMD was proposed, this method has good effect on suppressing noise, comparing with wavelet threshold filter and has advantages of retaining edge and detail information. Experimental results show that HHT method provides reliable basis for the feature extraction and pattern recognition of SEMG of the lower limb and has a good prospect of application.


Wu Y.,Hangzhou Diana University | Liu Q.,Hangzhou Diana University
Chinese Journal of Sensors and Actuators | Year: 2010

A device capturing six-dimension information of a helical spring deformation is designed. The helical spring is fixed on the base of the device, by applying a force on the top of the spring, and then the six-dimension information of the spring ' s deformation is obtained by an ad hoc sensing technology. The principle of this sensing technology is described, and the circuit detecting the strain of the spring wire surface is designed. Finally, an experimental system of the device capturing six-dimension information is developed, and the experimental result verifies the feasibility of this method.


Zheng B.,Hangzhou Diana University | Li L.-H.,Hangzhou Diana University
Chinese Journal of Biomedical Engineering | Year: 2013

Protein sequence feature and machine learning algorithm are two important aspects to determine the results of protein structural class prediction. In this study, we established 17-D and 57-D feature information sets through fusing the sequence information, physical and chemical information with the secondary structure information based on the k-word statistical frequency and the k-fragment distribution feature extraction method. By introducing Multi-Agent's idea into Adaboost. M1 algorithm, a novel method for protein structural class prediction, called Ma-Ada multi-classifier fusion algorithm, was proposed, which fully utilized the information of the single classifier metric layer and the fusion of information among individual classifiers. Four protein datasets including Z277, Z498, 1189, D640 were used to validate the performance of the Ma-Ada algorithm. Classification accuracies are 91.3%, 96.8%, 85.3% and 87.2% with 57-D features, and 90.6%, 95.8%, 84.8% and 88.3% with 17 D features on datasets Z277, Z498, 1189 and D640, respectively. The experimental results show better.


Zhang Q.,Hangzhou Diana University | Xi X.,Hangzhou Diana University | Luo Z.,Hangzhou Diana University
Chinese Journal of Sensors and Actuators | Year: 2012

This paper is aimed at raising the pattern recognition rate of physical movement based on electromyography (EMG) signal generation mechanism by presenting a new method of pattern recognition in accordance with EMG morphological characteristics. The complexity and self-similarity of the EMG are represented by the concepts of correlation dimension and fractal dimension in the fractal theory respectively. The calculation of the correlation dimension adopts an improved G - P algorithm, named G - P correlation dimension approximation method. In hand gestures pattern recognition, the combination of correlation dimension and fractal dimension is used as an input eigenvector of multi-pattern recognition classifier and the binary-tree architecture classifier is constructed with twin support vector machines(TSVM). The experiment, designed to classify four hand gestures including hand open, hand grasp, wrist extension and wrist flexion, shows that by using this method, the recognition rate has reached 91.0%, which demonstrates the practicality of this approach.

Loading Hangzhou Diana University collaborators
Loading Hangzhou Diana University collaborators