Electronic Clinical Pathways (eCPs) is important and necessary in the digitalized hospital. However, most of the data in eCPs is still isolated. And general methods to implement the data exchange between the eCPs and other medical systems are flawed in different respects. This paper presents a modeling method for the semantic integration of Clinical Pathways with related medical systems, particularly the Electronic Medical Record (EMR). We adopt the ontological approach to construct the middle tier model which is proposed against the problems in traditional heterogeneous data integration. Ontologies provide an opportunity to capture and manage common data and map concepts from one application to another. Different kinds of data exchange methods are discussed and compared in this paper and some real-world use cases are discussed to illustrate the procedures of the semantic integration.