Use the medical knowledge graph in smart applications such as clinical diagnostics

A medical knowledge graph is a specific way that researchers can organize and present information for use in medical research and clinical applications. The Knowledge Graph concept was developed by Google in 2012 as part of his search algorithm.

Since then, medical knowledge graphs have shown wide utility in the fields of medicine and medical research. For example, a knowledge graph can be based on data extracted from electronic medical records, clinical trials, or other medical literature. When that information is plugged into computer systems, the medical knowledge graph becomes the backbone of intelligent applications such as chatbots, diagnosis and treatment plan recommendations, and clinical education.

In a paper published in Big data mining and analysis On January 26, researchers set out to provide a comprehensive catalog of medical knowledge graphs, how they were created and when they should be used.

“Medical knowledge graphs are the cornerstone of intelligent healthcare and have been used in a variety of medical applications.

“Understanding the research and application development of the medical knowledge graph is important for future related research in the biomedical field. It shows progress on the graph.”

One of the most important factors in determining the credibility and reliability of any medical knowledge graph is its data source. With the spread of electronic medical records and online databases of clinical trial data, it has become possible to create a wide range of medical knowledge graphs. The researchers outline four potential data sources, including real-world data obtained from services such as electronic medical records and other available clinical data. Scientific publications such as journals, textbooks, guidelines, standards libraries, and open and shared databases of medical knowledge. These are free, open and accessible databases.

Researchers also analyze when to use knowledge graphs and different ways to organize data. They identify disease diagnosis as a hotspot for future research, where researchers use knowledge graphs to identify diseases and provide predictive outcomes. They also noted that medical knowledge graphs could be useful in drug development, and research is underway to determine how these tools could be used to develop his COVID-19 drug. doing.

“We hope that our research will help readers understand the research value of medical knowledge graphs, their main research objectives, progress and challenges, and help researchers in related fields to explore these tools.” ,” Li said.

In the future, researchers will focus on various challenges related to medical knowledge graphs. One of the challenges is getting data from different sources and making sure the data is cohesive and integrated for applications to read. Second, the medical field is incredibly complex. Future researchers should continue to explore how the medical knowledge graph can explain this complexity, especially with respect to clinical diagnosis and treatment.

Another challenge is that as different medical knowledge graphs are created, they are all created using different frameworks, terminology, and technologies. Finally, they argue, this kind of knowledge graph should incorporate common sense knowledge in addition to specialized medical knowledge.

“The next step will focus on detailed research on how to construct and reason about medical knowledge graphs,” said Li. “The ultimate goal is to build a large-scale, high-quality medical knowledge graph and propose newer reasoning methods and applications in precision medicine.”

Provided by Tsinghua University Press