A Concept of Applying Social Network Analysis in Medical Diagnosis

This paper aims to apply a social network analysis (SNA) to a medical diagnosis issue, which is the Caesarean sections due to Cephalopelvic disproportion (CPD). Firstly, the pregnant women, here is called "Patients", are connected with each other by a medical examination consideration. The patients are also grouped based on medical pattern similarity. Secondly, the centrality measures in SNA, such as Degree, Hub or Authority, Closeness centrality, and Betweenness Centrality are applied to identify the cluster representative in each group of patients. To evaluate the proposed idea, the sample test of patients is conducted. The results show that hub or authority is taken to identify the cluster representative with minimal average distance between the test patient and existing groups, which is 0.684 by Euclidean distance. Finally, the medical diagnosis of test patient, here, is also acceptable by comparing to its real diagnosis by physician.