Medical equipment maintenance in modern hospital management is an emerging marginal discipline which is one of the important branches of hospital management with the dual function of management and technology. The good or bad management of medical equipment directly affects the quality of hospital work. With the development of science and technology, the scale of hospitals continues to expand and develop; hospitals are equipped with more and more medical equipment and its technical content is getting higher and higher, which makes the medical equipment management and maintenance practitioners more demanding. However, the situation of existing medical equipment management and maintenance personnel is not optimistic. For this reason, this paper designs a good database structure, which can effectively reduce data redundancy and better avoid the problems that may occur during the database operation. As a reference guideline for database design, the three database paradigms can be studied and understood in depth to help software designers design more “robust” databases, enabling effective maintenance and management of medical imaging equipment.

Epilepsy is a disease with chronic disorder of nervous system, which owns the characteristic of repeat attacking and difficult to cure. So, it's significant to study the forecasting method and give preventive treatment. In this paper, the approximate entropy and the maximum Lyapunov exponent based on the improved Wolf algorithm were used to extract and analyze the characteristic of nonlinear dynamics of the human epileptic electroencephalograph under different states, the results indicated that two methods can effectively predict preictal state and approximate entropy show better performance. Results calculated in the wavelet domain were compared between the two methods and better effect can be obtained under first order discrete wavelet transform. The early waning and automatic drug release system was simulated by LabVIEW Virtual Instrument, with 80% sensitivity and 90.9% specificity. The automatic process and drug-release rate controlling by characteristic value is simulated, which show important application value.Keywords-the warning system of epilepsy; approximate entropy; Lyapunov exponent; discrete wavelet transform; automatic drug release system

An EEG integrated analysis system was developed by LabVIEW, which expanded bandwidth above 3 (KHz) to collect more useful information. The novel bandwidth preamplifier was designed and the system of data acquisition, integrated analysis was implemented, which could analyze EEG with an optional combination of time domain, frequency domain, time-frequency, nonlinearity, chaotics, several spectral analysis, wavelets and other special algorithms under multi-parameter. The system syncretized a variety of dedicated processing methods to meet the need of feature extraction for weak EEG signals. In the system the dual-channel analysis system with multiparameter was developed to strengthen comparing analysis of multi-signal feature. Epileptic EEG as example signal was analyzed and system showed good performances. As novel system it could comparatively analyze the human physiological signals at different stages by combining different algorithms, parameters and filters. The system will be an important tool in researches of human disease and brain function, which is better way to explore the relationships between EEG and health, brain function, brain cognitive.