The Neural Network Prediction Method Is Applied Research In The Efficacy Evaluation Of Depression

It has been verified by clinical practice that the therapeutic effects for depression relate to many effect factors and the complex nonlinear interaction between them. While Artificial Neural Network (ANN) is a nonlinear system that simulates the information processing method of the human brain, with high ability to deal with nonlinear problems, and adapts to the modeling for such problems as with complex information, dark background knowledge, or indefinite inference rules. In this study, the back propagation (BP) neural network is used to estimate effectiveness for depression.This study contains 3 continued linked parts : clinical trials, effect factors analysis and forecast modeling for effectiveness.The objective are respectively that:1. To study and analysis comparatively the clinical phenomenon of therapeutic effect of drug and acupuncture for treatment of depression;2. To review and analysis the influence factors of clinical effectiveness for treatment of depression;3. To modeling a forecast model aimed to forecast the quantitative effectiveness which will appear at the end of treatment course.Correspondingly, the methods are shown below:1. Two groups of the patients are chosen: the control group– 30 patients, received Prozac on 20 mg per day, during 6 weeks; and the acupuncture group– 30 patients, received acupuncture 30 minutes per time, 3 times a week, during 6 weeks.2. Review the effect factors of clinical effectiveness for treatment of depression from document material, and analysis what influence the factors of age, severity degree, and disease course have set on the effectiveness.3. Build a forecast model based on neural network method, with the quantitative values of relevant factors of effect as input variables of the network, while the changes in scales scores between pre and post treatment are set as output variables. The collected 110 cases are divided into training samples and testing samples and the neural network is trained and tested.It can be concluded from the results that:1. Notable positive effect and reduction in depressive symptoms are shown in both drug group and acupuncture group; Significant difference is not indicated between the effectiveness of drug and acupuncture. However, the therapeutic effects of these two therapies are difference;No negative effects of acupuncture is found.2. There are many effect factors of clinical effectiveness for treatment of depression between which the complex interactions also influence the effectiveness. Concluded from the clinical data in our trial, the different therapeutic effects are established in different groups compartmentalized by onset age,severity degree,disease course.3. The neural network model in this research has got rather ideal results of estimation in identification for the known samples, however, its predicting ability for unknown samples has not achieved such ideal results as that for known samples. The forecast model based on neural network is expected to be a helpful method for clinical research and practice with the work of expanding the sample size and improving the neural network model.