[4C2-OS-27b] 人工知能の医療応用(2)

[4C2-OS-27b-03] クラスタリングによる看護クリニカルパス生成

〇津本 周作¹、木村 知広¹、岩田 春子¹、平野 章二¹ (1. 島根大学)

キーワード：クラスタリング、クリニカルパス、病院情報システム

This paper proposes the following two-fold clinical care generation method. First, the system proposes how clinical cases with the same DPC code are characterized by mixture model clustering, and construct classification model by the analysis of discharge summaries. Then, cases are classified by using the classification model and a clinical pathway is generated for each new class. The proposed method was evaluated on the dataets extracted hospital information system, whose results show that plausible clinical pathways were obtained.