基于模糊理论的高血压药物疗效预测模型研究

ON PREDICTION MODEL OF HYPERTENSION DRUGS EFFICACY BASED ON FUZZY THEORY

针对评估期内药物占有率MPR(Medication Possession Ratio)和血压进行分析研究,建立用于发现二者关系的贝塔分布模型。利用模糊理论和遗传算法通过交叉验证对模型进行优化并与线性分布模型进行对比。实验结果表明,利用贝塔分布模型确定的MPR与血压值的关系能很好地对患者的用药疗效进行预测。对于大多数高血压患者,只有接受长期的药物治疗,才能使血压得到有效控制。

Aiming at the medication possession ratio （MPR） and blood pressure within evaluation period we carried out analysis and study, and built a beta distribution model for finding the relationship between them. Then we employed fuzzy theory, genetic algorithms （GA） and cross validation to optimise the model, and compared it with linear distribution model. Experimental results showed that the MPR and blood pressure value determined by using the beta distribution model can well forecast the meditation efficacy on patients. For most hypertensives, only by long-term treatment with medicine can the blood pressure be effectiveljr controlled.