Purpose/Significance To construct a knowledge graph based on clinical practice guidelines for pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU (PADIS guidelines), and to provide methodological references for improving the adherence of medical staffs to the guidelines. Method/Process Based on the construction method of the knowledge graph, the guidelines are used as the main data source, and the ontology is constructed based on the guideline framework and expert opinions. The triplet data model is used to represent knowledge. The “BIOES” labeling system, including entities, relationships, and attributes, is adopted for manual labeling. The visualization of the knowledge graph is realized by using the Neo4j graph database. Result/Conclusion The knowledge graph includes 112 entity types and 16 relationships among entities, and 124 triples are formed to describe the relationships among entities, the attributes of entities and their corresponding attribute values. In the future, it is necessary to explore the application of deep learning and other models in knowledge extraction, and explore the self-updating mechanism of guidelines to adapt to the current situation of rapid updating of guidelines in the medical field.