Drug Adverse Event Extraction Using Multi-Dimensional Attribute Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for extracting drug adverse events from medical information data face challenges in accurately distinguishing between adverse events and non-adverse events due to similarities in attribute data, leading to difficulties in broadly extracting combinations indicating adverse events with few mistakes.
Innovation Solution
A drug adverse event extraction method that generates attribute data for known positive and negative example combinations, learns a discriminant model, and applies an extraction condition to calculate scores for combinations that are neither positive nor negative examples, incorporating medical events beyond disease occurrence during prescription periods to enhance discrimination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If attribute data is generated based only on disease occurrence during prescription periods, then the extraction process is simple, but the discrimination accuracy between adverse events and non-adverse events deteriorates
Solution Approach 1:
The patent extends the analysis from a single dimension (disease occurrence during prescription periods) to multiple dimensions by incorporating additional medical events such as medical acts performed, events showing medical acts were performed, hospitalization, and medical expenses. This multi-dimensional approach enables better discrimination between adverse events and non-adverse events while maintaining a systematic extraction process
2Measurement precision
If a broader range of medical events is considered, then the detection accuracy of adverse events is improved, but the data processing complexity increases
Solution Approach 1:
The patent segments the broad range of medical events into distinct categories: prescription events, disease occurrence events, medical act events, and related event events. By organizing these diverse events into structured segments with specific attribute data for each, the system achieves high detection accuracy while managing data processing complexity through systematic categorization
3Adaptability or versatility
If more types of medical events are included in attribute data, then the extraction of unknown adverse events is enhanced, but the computational load increases
Solution Approach 1:
The patent creates a universal attribute data structure that can handle multiple types of medical events (prescription, disease occurrence, medical acts, and related events) within a unified framework. This multi-functional attribute system enables the extraction of various types of adverse events using the same processing logic, enhancing versatility while optimizing computational efficiency through standardized data handling
Data Source
AI summary
A method of extracting a combination of a drug and an adverse event related to the drug includes: for each of positive example combinations, negative example combinations and combinations that are neither positive examples nor negative examples, which are combinations of drug and disease, extracting medical events from medical information data about a patient and generating attribute data based on time-series information about the medical events; and learning a discriminant model based on attribute data of the positive and negative examples; and inputting attribute data corresponding to the combinations that are neither positive examples nor negative examples to the discriminant model to determine scores.