Detection of Precarious Situations in Medical Care with Mining Track record of Dosing

摘要

We propose a new approach to detect the precarious situation in medical care analyzing tracking record. Attention is being drawn to the use of incident reports as a means of increasing patient safety. Research teams are being formed across the world by WHO and in Japan by the Health, Labour and Welfare Ministry. In this instance, what is being emphasized as a major direction for future incident analysis is the assimilation of the existing top-down type class grants and bottom-down ontological construction. In this research, targeting incident case studies collected in Japan, we evaluated the degree of similarities between incident documents obtained bottom-up and the links between existing classes granted top-down. In doing so, we made it possible to evaluate overall similarities regarding incident documents through the method of network analysis. In addition, it became clear that the use of the Cos coefficient or the Jaccard coefficient is appropriate in creating networks. As a result of this analysis, existing classes correspond comparatively well with the characteristics of reports regarding the abstract and solution; on the other hand, regarding the background, it demonstrated that existing classes are inadequate in representing the characteristics of documents and that there is a need to improve classes. By the way, we can upgrade patient safety and quality of health care service.