The body is at a hypo-coagulation status after the heart mechanic valve prosthesis replacement operation, and the incidence of anticoagulation complications is rather high because of that administration of warfarin may result in “anticoagulation vacuum” at an early stage. Moreover, the necessary application of other anticoagulation methods assisting the employment of warfarin have still not been scientifically normalized. Blood coagulation factor Ⅱ,Ⅶ, prothrombin fragment1+2 (F1+2 ), urine fibrimopeptide A (UFPA) , and International Normalized Ratio(INR), could exactly reflect the anticoagulation intensity 48-72 hours after the replacement operation,reasonable use of anticoagulant therapy as well as accurate and in-time monitoring methods is significant to reduce complications,elevate survival rate, and improve quality of life.
Abstract： Diseases prognosis is often influenced by multiple factors, and some intricate non-linear relationships exist among those factors. Artificial neural network （ANN）, an artificial intelligence model, simulates the work mode of biological neurons and has a b capability to analyze multi-factor non-linear relationships. In recent years, ANN is increasingly applied in clinical medical fields, especially for the prediction of disease prognosis. This article focuses on the basic principles of ANN and its application in disease prognosis research.