Graphs have soon become the backbone of multiple applications - from search engines and recommender systems to intelligent chatbots. Hence, it is no surprise that they are effective in the healthcare domain as well. In this blog post, Rhicheek Patra, a Senior Researcher at Oracle Labs, introduces some of the data-related challenges faced by healthcare providers and covers the benefits of machine learning. Subsequently, he discusses an example knowledge graph (KG) to illustrate how healthcare data is captured in a graph model.
In healthcare services, knowledge graphs have proven to be effective at mapping the relationships between the enormous variety and structure of healthcare data. For instance, graphs can capture what other data models are unable to do. It can model latent relationships between information sources and capture linked information like entity relationships. This innate capability enables doctors and service providers to discover the information they need from the humongous data.
Let us take an example knowledge graph from the healthcare domain and an instance of a patient being admitted to a hospital. When a patient is admitted to a hospital, it is considered an event—“admission”— which can vary in duration depending on the criticality of the associated diagnoses. In addition, an “admission” has various properties, such as the medications during the on-going admission or the "pre-diagnosis" note from the medical practitioner. Subsequently, one or more properties could be linked to the diseases or symptoms in a knowledge base having information on various diseases and the symptoms associated with them.
Hence, when “we illustrate an instance of the healthcare knowledge graph which demonstrates a heterogeneous graph with entities as either "Disease", "Symptom" or "Admission." In this KG the relationships are either "Admission has similar pre-diagnosis as either a disease or a symptom" or "Disease has the following symptoms". To compute the similarity, we employ a combination of n-grams with TF-IDF (Term Frequency-Inverse Document Frequency) to link Admissions with diseases or symptoms. Furthermore, the "Internal information" is for information collected from the hospital database (with patient admissions) whereas the "External information" is the information aggregated from external sources which in this case is from a Disease-Symptom KG built externally and available publicly on Github through this link.”
To conclude, there is a sustained focus on developing various systems to make proper diagnoses, predictions, and treatments. Among the various innovative systems, knowledge graphs populated with healthcare data can help doctors and healthcare givers find information easily from a wide array of variables and data sources.
Click here to read the article.
Please give your feedback on this article or share a similar story for publishing by clicking here.
