This graph below is created by using the GenHealth embedding API on each medical concept and creating links between those concepts that are most closely related based on the embedding. There is no notion of text or code value similarity, but rather similarity base on the embedding layer which is a vector of 800 numeric values. Clearly, the embedding layer is a very powerful tool for creating relationships between medical concepts. This can be expanded to include relationships for higher order concepts like episodes of care, patient similarities, providers search, etc.