Fig. 1: Data transformation with an example. | Communications Medicine

Fig. 1: Data transformation with an example.

From: Simulated patient systems powered by large language model-based AI agents offer potential for transforming medical education

Fig. 1: Data transformation with an example.The alternative text for this image may have been generated using AI.

Data transformation of EHRs from (a) raw discharge notes (with extracted entities) to (b) constructed knowledge graph (through NER). Here, legend indicates medical entity and color mapping and only a subset of symptoms and medical history is presented in the figure for clarity. The patient’s family medical history is described as “Both parents died from CVA’s.” NER extracts “both parents” as Family Member, and “cerebrovascular accident (CVA)” as Medical History. A relationship of “HAS_FAMILY_MEMBER” is established between the Patient node and Family Member node (both parents), and “HAS_MEDICAL_HISTORY” is established between the Family Member node and the Medical History node (CVA).

Back to article page