Recently, a massive amount of data has been available for research on Alzheimer’s disease. However, the data entities are stored with different names at different levels of granularity and in various formats. Thus, a comprehensive knowledge graph is needed to facilitate the development of analytical models related to Alzheimer’s disease. In our previous work, we created the Alzheimer’s disease Ontology for Diagnosis and Preclinical Classification (AD-DPC), a domain ontology incorporating the knowledge of medical experts in an understandable way for individuals with no medical background. This paper extends our work by employing Neo4j graph database technology and AD-DPC to build a domain-specific knowledge graph. Data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) is used to populate the knowledge graph and to validate its data retrieval and visualisation capabilities. The knowledge graph contains 2996 diagnoses, 154,953 psychometric findings, 24,102 blood findings, 12,471 CSF findings, and 14,703 brain imaging findings from MRI or PET scanning. The nodes were further annotated with 259,260 labels and 673,325 relations based on the AD-DPC ontology. The results prove the efficacy of using ontologies as a base for the semantic modelling of graph databases. They further rely on their straightforward and intuitive data querying and visualisation support.