Automatic clustering method of flow cytometry data based on -distributed stochastic neighbor embedding / 生物医学工程学杂志

The traditional method of multi-parameter flow data clustering in flow cytometry is to mainly use professional software to manually set the door and circle out the target cells for analysis. The analysis process is complex and professional. Based on this, a clustering algorithm, which is based on -distributed stochastic neighbor embedding ( -SNE) algorithm for multi-parameter stream data, is proposed in the paper. In this algorithm, the Euclidean distance of sample data in high dimensional space is transformed into conditional probability to represent similarity, and the data is reduced to low dimensional space. In this paper, the stained human peripheral blood cells were treated by flow cytometry, and the processed data were derived as experimental sample data. The -SNE algorithm is compared with the kernel principal component analysis (KPCA) dimensionality reduction algorithm, and the main component data obtained by the dimensionality reduction are classified using -means algorithm. The results show that the -SNE algorithm has a good clustering effect on the cell population with asymmetric and trailing distribution, and the clustering accuracy can reach 92.55%, which may be helpful for automatic analysis of multi-color multi-parameter flow data.

Research on electroencephalogram emotion recognition based on the feature fusion algorithm of auto regressive model and wavelet packet entropy / 生物医学工程学杂志

Focused on the world-wide issue of improving the accuracy of emotion recognition, this paper proposes an electroencephalogram (EEG) signal feature extraction algorithm based on wavelet packet energy entropy and auto-regressive (AR) model. The auto-regressive process can be approached to EEG signal as much as possible, and provide a wealth of spectral information with few parameters. The wavelet packet entropy reflects the spectral energy distribution of the signal in each frequency band. Combination of them gives a better reflect of the energy characteristics of EEG signals. Feature extraction and fusion are implemented based on kernel principal component analysis. Six emotional states from a public multimodal database for emotion analysis using physiological signals (DEAP) are recognized. The results show that the recognition accuracy of the proposed algorithm is more than 90%, and the highest recognition accuracy is 99.33%. It indicates that this algorithm can extract the feature of EEG emotion well, and it is a kind of effective emotion feature extraction algorithm, providing support to emotion recognition.

Application of chemical pattern recognition to quality control of Radix Paeoniae Alba / 中草药

Objective With generalization and steadiness,a new evaluation model by Integrating Non Linear Features extraction algorithm with artificial neural networks(ANN) used for pattern recognition of quality control of Radix Paeoniae Alba was proposed in this paper.Methods The HPLC data from 29 samples with different quality were proceeded with nonlinear kernel principal component analysis(KPCA) and an improved Back propagation algorithm of ANN.The extract characteristics was fed into BP neural networks as input elements for pattern recognition.In the meantime,the processing data,the optimal numbers of hidden layers,the numbers of hidden nodes,excitation functions,and over-fitting,etc. were discussed wholly so that standardization networks was designed without jamming.Results As recognition ratio was 100%,the pattern recognition of quality control of Radix Paeoniae Alba was established successfully by trained networks and predicted results.Conclusion Integrating KPCA algorithm with ANN for pattern recognition of quality control of Radix Paeoniae Alba has been proved to be available.