失语症第三方残疾功能变量的网络图分析

关键词：失语症  第三方残疾  国际功能、残疾和健康分类  图建模  网络分析

Keywords:aphasia  third-party disability  international classification of functioning, disability and health  graphical modeling  network analysis

摘要：

      摘要 目的：构建失语症第三方残疾功能变量间相关关系的图模型，分析其网络特征，为失语症及其第三方残疾康复提供新的研究思路。 方法：参考国外相关文献及综述，选取失语症第三方残疾功能变量的64个ICF条目，采用访谈法评估30例受试者的功能变量。构建图模型时，以每个条目作为节点，各条目间的风险相关性作为连线。建模采用R软件（3.6.2版），网络分析可视化采用Pajek软件（5.08版），用马赛克图表示各条目在各级限定值上的频数分布。 结果：64个条目中，有14对两两相连的双条目组元，有18个相互联系的条目构成的功能地形图中最大的独立结构，为主组元结构。其中d855（无报酬的就业）占据了重要位置，分别与d920（娱乐与休闲）、b1263（精神稳定性）和e5750（全社会的支持服务）相连。双条目组元及主组元网络结合条目在马赛克图上的频数分布，既可以提供功能变量间的网络关系，又可以提供自身的属性特征，二者相互参照，以提供更多元的康复思路。 结论：采用ICF功能变量构建的图模型，可反映失语第三方残疾功能变量间的复杂关系网络。各功能变量的频数分布，可反映属性特征。从这两个不同的角度进行相关分析，可以为疾病的康复提供多视角的策略。

Abstract:

      Abstract Objective:To construct a graph model of the correlation between the third-party disability functional variables of aphasia， analyze its network characteristics， and provide a new research idea for aphasia and third-party disability rehabilitation. Method: Totally 64 ICF items of the third-party disability functional variables of aphasia were selected by referring to relevant foreign literatures and reviews， and the functional variables of 30 subjects were evaluated by interview.Each item is taken as the node， and the risk correlation coefficient between each item is taken as the line value to construct the graph model.R software (version 3.6.2) was used for modeling， and Pajek software (version 5.08)was used for network analysis. The mosaic diagram was used to represent the frequency distribution of each item at all levels of restricted values. Result:Among the 64 items， there are 14 pairs of double-item components， and 18 interrelated items that constitute the largest independent structure in the functional topographic map， that is， the main component. An important position is d855 (Non-remunerative employment)， which is connected with d920 (Recreation and leisure)， b1263 (Psychic stability) and e5750 (General social support services).The network of the double-item component and the main component and the Mosaic graph can provide the network relationship between the functional variables as well as its own attribute relationship. The two can be cross-referenced to provide more diversified rehabilitation ideas. Conclusion:The ICF functional variables can be used to graph modeling， which can reflect the complex relationship network among the third-party disability functional variables of aphasia.The frequency distribution of each functional variable can reflect the attribute characteristics.Correlation analysis from these two different perspectives can provide a multi-perspective strategy for the rehabilitation of the disease.