Incorporating Topic Assignment Constraint and Topic Correlation Limitation into Clinical Goal Discovering for Clinical Pathway Mining

摘要: Clinical pathways are widely used around the world for providing quality medical treatment and controlling healthcare cost. However, expert-designed clinical can hardly deal with variances among hospitals patients. It calls more dynamic adaptive process, which is derived from various data. Topic-based pathway mining an effective approach to discover a concise process model. Through this approach, latent topics found by Dirichlet allocation (LDA) represent goals. And methods extract temporal relations between these topics. topic usually not desirable due low performance of LDA in In paper, we incorporate assignment constraint correlation limitation into enhance ability discovering high-quality Two real-world datasets evaluate proposed method. The results show that discovered our method higher coherence, informativeness, coverage than original LDA. These suitable Also, illustrate generating comprehensive topic-based