Prediction of epilepsy based on common spatial model algorithm and support vector machine double classification / 生物医学工程学杂志

At present the prediction method of epilepsy patients is very time-consuming and vulnerable to subjective factors, so this paper presented an automatic recognition method of epilepsy electroencephalogram (EEG) based on common spatial model (CSP) and support vector machine (SVM). In this method, the CSP algorithm for extracting spatial characteristics was applied to the detection of epileptic EEG signals. However, the algorithm did not consider the nonlinear dynamic characteristics of the signals and ignored the time-frequency information, so the complementary characteristics of standard deviation, entropy and wavelet packet energy were selected for the combination in the feature extraction stage. The classification process adopted a new double classification model based on SVM. First, the normal, interictal and ictal periods were divided into normal and paroxysmal periods (including interictal and ictal periods), and then the samples belonging to the paroxysmal periods were classified into interictal and ictal periods. Finally, three categories of recognition were realized. The experimental data came from the epilepsy study at the University of Bonn in Germany. The average recognition rate was 98.73% in the first category and 99.90% in the second category. The experimental results show that the introduction of spatial characteristics and double classification model can effectively solve the problem of low recognition rate between interictal and ictal periods in many literatures, and improve the identification efficiency of each period, so it provides an effective detecting means for the prediction of epilepsy.

Spatial clustering and influential factors of hand-foot-mouth disease (HFMD) in Chongqing,China,2008-2012 / 中国人兽共患病学报

ABSTRACT:In order to explore the spatial clustering and influential factors of HFMD in Chongqing ,China from 2008 to 2012 ,spatial autocorrelation and spatial regression analysis (using the spatial lag model in this study ) were carried out using the HFMD data of 38 districts (counties) from 2008-2012 in Chongqing by OpenGeoDa ,and the HFMD case‐based data was collected from the Disease Supervision Information Management System of Chongqing Center for Disease Control and Preven‐tion .We found that the global Moran’s I coefficient of Chongqing from 2009 to 2012 was 0 .458 7 ,0 .567 5 ,0 .398 6 ,and 0 .606 0(P0 .05) .Results of multi‐factor spatial lag regression analysis demonstra‐ted that the incidence of HFMD was positively related with urban rate (β=1 .667 6 , P=0 .001 6) ,and negatively correlated with medical technical personnel per thousand (β= -0 .000 2 ,P=0 .019 8) .In general ,the incidence of HFMD was found ge‐ographically clustered in Chongqing from 2009 to 2012 which was significantly influenced by urban rate and medical technical personnel per thousand population ,and while the urban rate was the main factor .

Comparison on the performance of both temporal and temporal-spatial models for outbreak detection through China Infectious Diseases Automated-alert and Response System(CIDARS) / 中华流行病学杂志

Objective To analyze the pilot results of both temporal and temporal-spatial models in outbreaks detection in China Infectious Diseases Automated-alert and Response System (CIDARS)to further improve the system. Methods The amount of signal, sensitivity, false alarm rate and time to detection regarding these two models of CIDARS, were analyzed from December 6,2009 to December 5,2010 in 221 pilot counties of 20 provinces. Results The sensitivity of these two models was equal(both 98.15%). However, when comparing to the temporal model, the temporal-spatial model had a 59.86% reduction on the signals(15 702)while the false alarm rate of the temporal-spatial model(0.73%)was lower than the temporal model(1.79%), and the time to detection of the temporal-spatial model(0 day)was also 1 day shorter than the temporal model.Conclusion Comparing to the temporal model, the temporal-spatial model of CIDARS seemed to be better performed on outbreak detection.