| Chronic heart failure(CHF) is a kind of clinical syndrome of end-stage heart diseases. For its low diagnosis rate and high mortality, CHF has become a global problem in public health, which seriously endangers the healthy of human being.In recent years, the diagnosis and treatment of heart failure has acquired remarkable progress, and some new diagnosis and inspection methods have come out. But presently the clinical diagnosis mostly relies on single parameter, thus it is difficult to evaluate the risk of heart failure exactly. In clinic an integrated quantitive method for the evaluation of the early stage of heart falure is desired.Under the guide of clinical medicine,the current study applied the technic of biomedical signal processing technic and artificial neural network technic to screen the sensitive parameters for heart failure, and developed a risk evaluation model for chronic heart failure. The model is based on the adaptive resonance theory. Accordingly,the following tasks were completed:1. The clinical parameters were analyzed to screen the heart failure most related parameters.2. Established the risk evaluation model for chronic heart failure, which is based on the adaptive resonance theory.3. Applied the risk evaluation model to evaluate the risk of hypertension and heart failure sufferer.4. Explored new heart failure related predictor.The validity of the risk evaluation model was testified in the cardiovascular sufferers. The results show that the model is able to correctly evaluate the risk of hypertension sufferers,the model was utilized preliminary and its efficiency still needs further investigation. Moreover, new parameters related to heart failure were investingated. The results indicate that heart rate turbulence characteristic and deceleration capacity are usefull parameters in the risk evaluation of heart failure. |
