Drug Knowledge Graph Construction via Entity Normalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for constructing drug knowledge graphs face challenges in accurately extracting and representing usage and dosage information from drug prescriptions, leading to a high workload for pharmacists in reviewing large numbers of prescriptions.
Innovation Solution
A method and apparatus that identify entities in drug texts, replace medical key entities with character strings, form linear entity relationships, and generate a drug knowledge graph through syntactic parsing, incorporating preprocessing steps like normalization and mapping relationships to ensure accurate representation of drug usage and dosage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing methods are used to construct drug knowledge graphs, then the construction process can be completed, but the accuracy of extracting usage and dosage information is poor
Solution Approach 1:
The patent segments the drug text processing into distinct stages: entity identification, linear entity relationship formation, and syntactic parsing. This segmentation allows each stage to focus on specific extraction tasks, improving the accuracy of usage and dosage information extraction while automating the process to reduce pharmacist workload.
Solution Approach 2:
The patent introduces a linear entity relationship as an intermediary structure between the original drug text and the final knowledge graph. This intermediary representation captures the sequential relationships among entities (drug, disease, usage, dosage) and facilitates more accurate information extraction through structured syntactic parsing.
2Productivity
If manual review of prescriptions is performed, then accuracy can be maintained, but the workload and time consumption increase significantly
Solution Approach 1:
The patent implements a self-service system where the drug knowledge graph automatically extracts and structures usage and dosage information from prescriptions through entity identification and syntactic parsing. This automation enables the system to perform review functions independently, significantly improving review efficiency and reducing the time pharmacists need to spend on manual verification.
3Loss of information
If complex syntactic structures are used to represent drug information, then information completeness is improved, but processing complexity increases
Solution Approach 1:
The patent performs preliminary action by first identifying entities and forming linear entity relationships before conducting syntactic parsing. This preliminary structuring of information simplifies the subsequent parsing process while ensuring that all relevant usage and dosage information is captured, thus maintaining information completeness without excessive processing complexity.
Data Source
AI summary
A method and apparatus for constructing a drug knowledge graph are provided. The method may include: identifying entities in a drug text; replacing medical key entities among the entities with character strings that conform to a preset rule, to obtain a replaced text; restoring the character strings in a word segmentation result, determined based on the replaced text, to the medical key entities replaced by the character strings; forming a linear entity relationship between the entities based on the entities; and generating a drug knowledge graph according to a parsing result obtained by syntactically parsing the linear entity relationship.