Abstract:

Traditional decision-making of medical assistant diagnosis usually relies on experience of experts to make decisions on alternatives, but this method relies on experience of experts and may bring subjective fatigue to experts. Based on the theory of multi-attribute decision-making, a visual multi-attribute decision-making method is proposed in this paper. Firstly, ordered weighted geometric operators are used to sort the data in descending order to generate an ordered sequence. Secondly, the ordered data is transformed into nodes in complex networks through visual graph theory, and the ordered data is visualized through the connection relationship between nodes in complex networks. Finally, law of Coulomb is used for reference, the ordered data is visualized by using nodes in complex networks. The support function is set by the distance between the nodes and the value of the node to describe the support degree of the node. The feasibility and validity of the visual multi-attribute decision-making method proposed in this paper are verified by two cases of etiological diagnosis and disease diagnosis.

Key words: medical assistant diagnosis, multiple attribute decision making, complex network, support function

摘要：

传统的医疗辅助诊断决策通常通过专家依靠经验对备选方案进行决策，但是这种方法依赖于专家已有的经验并且可能会为专家带主观疲劳。基于多属性决策理论，提出了一种可视化多属性决策方法。该方法使用有序加权几何算子将数据进行降序排序，生成一个有序序列；通过可视图理论将有序数据转化为复杂网络中的结点，通过复杂网络中结点之间的连接关系将有序数据可视化；借鉴库伦定律，使用结点之间的距离与结点自身数值的大小设置支持度函数用以描述结点的支持度。实验通过病因诊断与病情诊断两个案例验证提出的医学辅助诊断多属性决策方法的可行性与有效性。

关键词: 医疗辅助诊断, 多属性决策, 复杂网络, 支持度函数