Hybrid knowledge graph for healthcare applications

A method comprising:

• transmitting, by at least one processor, a medical knowledge graph to a user, the medical knowledge graph
• containing a plurality of medical terms, the plurality of medical terms comprising a plurality of diseases, a plurality of conditions associated with the plurality of diseases, a plurality of symptoms associated with the plurality of diseases and the plurality of associated conditions, a plurality of complications, a plurality of laboratory tests for testing for the plurality of diseases and the plurality of associated conditions, and a conditions, plurality of treatments for treating the plurality of diseases and the plurality of associated conditions,
• depicting a plurality of nodes and a plurality of connectors connecting the plurality of nodes, the plurality of nodes comprising:
• a first set of nodes representing the plurality of diseases,
• a second set of nodes representing the plurality of associated conditions,
• a third set of nodes representing the plurality of symptoms,
• a fourth set of nodes representing the plurality of complications,
• a fifth set of nodes representing the plurality of laboratory tests,
• a sixth set of nodes representing the plurality of treatments,
• the plurality of connectors representing linkages between the plurality of nodes that are derived from (i) a plurality of ontological relationships between the plurality of medical terms, and (ii) a plurality of clinical relationships between the plurality of medical terms,
• the plurality of connectors comprising:
• a plurality of one-sided arrows, each of the plurality of one-sided arrows representing a causal relationship,
• a plurality of two-sided arrows, each of the plurality of two-sided arrows representing a clinical association;
• receiving, by the at least one processor, health data associated with a patient;
• selecting, by the at least one processor, one or more pathways between each pair of nodes by selecting one or more connectors that connect the pair of nodes, the one or more pathways representing one or more aspects of the patient's health data;
• determining, by the at least one processor, a connection strength for each pathway between each pair of nodes;
• aggregating, by the at least one processor, each connection strength to obtain a total connection strength;
• generating, by the at least one processor, a statistical correlation between each disease in the one or more nodes and the patient's health data;
• ranking, by the at least one processor, the statistical correlations;
• generating, by the at least one processor, a list of differential diagnoses for the patient, the list of differential diagnoses comprising each disease in the one or more nodes and the statistical correlations;
• transmitting, by the at least one processor, the list of differential diagnoses to the user, and
• diagnosing, by the at least one processor, the patient with a disease from the list of differential diagnoses that is consistent with the patient's health data.

A method comprising:

• transmitting, by at least one processor, a graphical user interface (GUI) configured to permit a healthcare practitioner to enter health data associated with a patient;
• transmitting, by the at least one processor, a medical knowledge graph depicting a plurality of nodes and a plurality of connectors connecting the plurality of nodes, wherein:
• a first set of nodes in the plurality of nodes represents a plurality of diseases,
• a second set of nodes in the plurality of nodes represents a plurality of conditions associated with the plurality of diseases,
• a third set of nodes in the plurality of nodes represents a plurality of symptoms associated with the plurality of diseases and the plurality of associated conditions,
• a fourth set of nodes in the plurality of nodes represents a plurality of complications associated with the plurality of diseases and the plurality of associated conditions,
• a fifth set of nodes in the plurality of nodes represents a plurality of laboratory tests for testing for the plurality of diseases and the plurality of associated conditions,
• a sixth set of nodes in the plurality of nodes represents a plurality of treatments for treating the plurality of diseases and the plurality of conditions,
• the plurality of connectors comprises arrows denoting probabilistic linkages between nodes in the plurality of nodes;
• receiving, by the at least one processor, the patient's health data;
• selecting, by the at least one processor, one or more pathways between each pair of nodes by selecting one or more connectors that connect the pair of nodes, the one or more pathways representing one or more aspects of the patient's health data;
• transmitting, by the at least one processor, the one or more pathways;
• determining, by the at least one processor, an aggregated connection strength for each of the one or more pathways, wherein the aggregated connection strength is determined based on the probabilistic linkages;
• generating, by the at least one processor, a statistical correlation between each disease in the one or more pathways and the patient's health data;
• ranking, by the at least one processor, the statistical correlations;
• generating, by the at least one processor, a list of differential diagnoses for the patient, the list of differential diagnoses comprising each disease in the one or more pathways and the statistical correlations;
• diagnosing, by the at least one processor, the patient with a disease from the list of differential diagnoses;
• and
• transmitting, by the at least one processor, to the user both the list of differential diagnoses and the disease.