Discrimination based on nonparametric cross-validation of diagnostic model of meningitis was established by the community.doc

Discrimination based on nonparametric cross-validation of diagnostic model of meningitis was established by the community Of: Liujuan Cui Xiao-Qin Luo Yanxia erected Lvman Road, GUO Xiu Gu Zhongwei Yan Yuxiang flowers [Abstract] Objective: To establish a community for meningitis and other central nervous system diseases in the differential diagnosis model. Methods: The different bandwidth non-parametric kernel density estimation of the discriminant analysis, Daquan nearly four typical cases of China met the inclusion criteria during the year 161 cases of meningitis and 161 cases of non-meningitis cases in patients with complete data for analysis. Results: The cross-validation method be the discriminant of meningitis group was 83.95% correct, the control group was 71.25%, the total rate of 87.64% correct judgments. The data were also logistic regression and artificial neural network model for analysis and the artificial neural network and logistic regression and compared the model. Conclusion: The non-parametric discriminant analysis model established diagnosis of meningitis is an ideal model. Keywords:] non-parametric identification; cross-validation; diagnostic model; meningitis Meningitis is a serious pathogen causing infections of central nervous system diseases. As complex etiology, diagnosis is difficult, dangerous and can cause epidemic disease, so in some parts of Asia, meningococcal meningitis is the death of a child under the age of 15 one of the main. meningococcal meningitis when the popular will lead to greater intensity of 3,000 to 10,000 deaths [1]. Early symptoms of meningitis are not specific, so doctors at the community level is difficult to be limited attention, and most of the meningitis are acute, especially meningococcal meningitis, if not timely diagnosis and prognosis of the patient’s life, will, after a great impact. Therefore, the establishment of meningitis for the community screening diagnostic model ability to improve communit