Biomedical Knowledge Graph Using NLP for Accurate Data Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for obtaining biomedical information are inefficient, time-consuming, and often yield inconsistent or outdated results due to reliance on generalized search tools and manual updates.
Innovation Solution
A biomedical knowledge graph is generated by extracting and structuring data from disparate sources using natural language processing and machine learning, enabling targeted and accurate retrieval of information through a query engine.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If generalized search tools with keywords are used to obtain biomedical information, then information can be retrieved from various sources, but the results are inconsistent and include non-relevant material
Solution Approach 1:
The patent introduces a knowledge graph as an intermediary structure between search queries and information sources. The knowledge graph pre-structures biomedical entities and relationships from multiple sources, allowing queries to be answered through structured traversal rather than unstructured searching. This mediator filters out non-relevant material and provides consistent, accurate results by relying on the pre-established semantic relationships in the knowledge graph.
Solution Approach 2:
The system performs preliminary action by pre-processing and structuring biomedical information into a knowledge graph before queries are submitted. Entities, attributes, and relationships are extracted and organized in advance from diverse sources including clinical trials, publications, and databases. This preliminary structuring enables rapid, accurate query resolution without requiring real-time information gathering and filtering.
2Reliability
If manual review or verification is used to update biomedical compilations, then information can be updated, but the process is time-consuming and potentially unreliable
Solution Approach 1:
The patent replaces the mechanical manual review process with automated computational methods. Natural language processing algorithms, machine learning models, and entity recognition systems automatically extract, verify, and integrate biomedical information from new sources. This substitution eliminates human time constraints and reduces variability in verification quality, enabling continuous, reliable updates without manual intervention.
Solution Approach 2:
The knowledge graph system performs self-service by automatically detecting new information sources, extracting relevant entities and relationships, and integrating them into the existing structure. The system monitors incoming data from clinical trials, publications, and databases, autonomously updates the knowledge graph, and maintains information reliability through automated consistency checks and validation rules without requiring external manual verification.
3Quantity of substance
If comprehensive biomedical information from multiple sources is gathered, then information completeness is improved, but searching and extracting relevant information becomes cumbersome and time-consuming
Solution Approach 1:
The patent segments comprehensive biomedical information into discrete, structured entities (e.g., diseases, treatments, genes, clinical trials) with defined attributes and relationships. Each entity is independently extracted and tagged with metadata from its source. This segmentation transforms the cumbersome task of searching through unstructured comprehensive information into efficient queries across structured entity types and relationships, dramatically improving extraction productivity while maintaining completeness.
Data Source
AI summary
A biomedical knowledge graph system includes a computer database of records comprising nodes of biomedical entities and connections between the entities representing biomedical relationships. One or more processors are configured to extract data from a plurality of data sources and determine biomedical entities and relationships between the entities based on analyzing the data, including searching for predetermined identifiers or patterns in the data. Based on the determined biomedical entities, each biomedical entity is assigned to a cluster of biomedical entity types and a context is identified for each of the entities. Based on the identified context and type of the biomedical entity, records of nodes and connections between nodes are incorporated into the knowledge graph, the nodes representing biomedical entities and the connections representing biomedical relationships between the entities structured according to the predefined schema.