一类面向中风后吞咽功能障碍康复治疗的群体智慧涌现方法

摘要

传统的中风后吞咽功能障碍康复治疗方案的制订通常以会诊方式,需要群体专家对所有可能的备选治疗方案进行讨论与决策,增加专家主观疲劳,且缺乏针对群体治疗智慧涌现方法的探讨,基层康复医师难以学习群体专家治疗智慧.基于多属性群决策理论,本文提出了群体智慧定义,给出了基于“专家讨论后的备选方案排序结果—子属性特征”的群体智慧涌现方法以及基于群体智慧的多属性决策方法,使计算机逐步学习群体专家经验并代替专家决策,减轻群体专家疲劳感,并具备针对未知备选方案进行自动决策的能力.针对一类数值实例,对传统多属性决策方法与所提决策方法进行了对比,并将所提方法应用于一类实际中风后吞咽功能障碍康复治疗中,验证了本文所提方法的正确性与可行性.%Traditional rehabilitation treatment decision-making process of dysphagia after stroke usually involves group experts to discuss all alternatives in consultation frame,suffering the from the experts' subjective fatigue and difficulties to learn the intelligence for primary doctors.Therein,the definition of group intelligence is proposed based on the multi-attribute group decision-making theories.Moreover,"the ranking results after discussion-the characteristic of the subattributes" group intelligence generating methods and the decision-making approaches based on the group intelligence are provided to help learn the group experts' experience and replace the experts,as well as reduce group's subjective fatigue.The proposed decision-making approaches are compared with the well-known multi-attribute decisionmaking method,meanwhile,real rehabilitation treatment examples of dysphagia after stroke are employed to exemplify applications of the proposed methods.