Software Development Of The Non-invasive Cardiac Function Detecting System And Clinical Test

Cardiovascular disease is one of the serious diseases harms the human health,according to statistics, the number of death from cardiovascular disease is about300million people every year, and the morbidity and mortality has continued to rise.Stroke volume, cardiac output and other hemodynamic parameters are the importantparameters in the evaluation of cardiovascular function. Thermodilution method isconsidered the “golden rule” of SV and CO detection in clinical, but it is invasive andusually leads to a variety of complications, thus has a limited clinical application. Onthe contrary, impedance method is non-invasive, and has a widely recognized accuracyand reliability, thus widely used in clinical detection in recent years.A noninvasive detection of cardiac function analyzer named KF_ICG-101wasdeveloped by our laboratory based on the principle of impedance method. In order toaccurately extract and display SV, CO and other hemodynamic parameters from thecardiac function signals, which are collected by the original apparatus, a softwaresystem for cardiac function analysis was developed. In the process of design, theprocessing effects of frequency domain de-noising and wavelet threshold de-noisingalgorithm were firstly compared using MATLAB, and determined wavelet thresholdde-noising algorithm for processing; Then, difference threshold algorithm and wavelettransform were adopted to the different SNR signals of clinical patients for featurepoints detection. At last，a set of a fully functioning, user-friendly cardiac functionanalysis software system was designed on the Visual C++6.0software developmentplatform.In order to verify the safety and effectiveness of the whole system, a clinical testwas conducted. Firstly, Impedance signal were synchronously collected from50patientsby the Ultrasonic Cardiac Output Monitor (USCOM1A) and our impedance signaldetection system in Chongqing Da Ping Hospital. Then, various types of statisticalanalysis were conducted using SPSS17.0and MedCalc medical software, includingerror analysis、paired t-test、correlation analysis and Bland-Altman consistency analysis. The statistical result shows that the two instruments can be used interchangeably. Inaddition, the instrument kept a good accuracy and stability in the clinical test andreaches the preliminary clinically applicable standards.