Automated Synonym Discovery for Electronic Health Record Interoperability

Solution Overview

Problem

Current health information systems lack interoperability, leading to difficulties in mapping and linking data elements across disparate electronic health record systems, which hinders real-time decision-support, data mining-based research, and the ability to establish reliable nomenclatural linkages, resulting in inefficiencies and inaccuracies in health care data analysis.

Innovation Solution

The development of a method and system for automatic discovery and validation of semantically similar terms across multiple health information systems, utilizing statistical properties and thresholds to establish reliable linkages between nomenclatures, enabling efficient and accurate mapping of synonyms and non-synonyms, even in the absence of unified ontologies.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual mapping methods are used to link data elements across health record systems, then mapping accuracy can be maintained through expert review, but the process becomes labor-intensive, expensive, and difficult to scale

Engineering Contradiction:

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The system enables automatic self-service mapping by computing similarity metrics between data elements from different health record systems using statistical properties and thresholds, eliminating the need for manual expert review while maintaining mapping quality through automated validation

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The invention transforms the mapping process by changing from qualitative expert judgment to quantitative parameter-based comparison, using statistical metrics and configurable thresholds to automatically determine mapping accuracy and reliability

Inventive Principle:

Principle #35Parameter changes

2Productivity

If automated mapping systems are implemented to improve productivity and scalability, then mapping efficiency increases, but mapping accuracy and reliability deteriorate due to lack of expert validation

Engineering Contradiction:

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The system incorporates feedback mechanisms where mapping results are validated against statistical thresholds and quality metrics, with the ability to iteratively refine mappings based on performance feedback while maintaining automated operation

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system performs preliminary statistical analysis and validation before finalizing mappings, pre-computing similarity metrics and establishing quality thresholds in advance to ensure accurate automated mapping without requiring post-hoc expert validation

Inventive Principle:

Principle #10Preliminary action

3Adaptability or versatility

If comprehensive data element mapping is performed across all health record systems, then interoperability and data linkage improve, but system complexity and computational resources increase significantly

Engineering Contradiction:

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The mapping system is segmented into modular components that independently compute similarity metrics for different data element types, allowing selective application to specific data domains and reducing overall system complexity while maintaining comprehensive coverage

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system applies mapping selectively to data elements that meet predefined criteria or thresholds, performing partial mapping on subsets of data rather than exhaustive mapping of all elements, thereby reducing computational complexity while achieving sufficient interoperability

Inventive Principle:

Principle #16Partial or excessive action

4Reliability

If statistical validation is applied to verify nomenclatural linkages, then reliability of mappings improves, but computational time and processing requirements increase

Engineering Contradiction:

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

Statistical validation metrics and thresholds are pre-computed and established before the mapping process, allowing rapid validation during execution without time-consuming real-time statistical analysis, thus maintaining high reliability while minimizing validation time

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11714837B2Synonym discovery

Publication Date: 2023.08.01

AI summary

Methods, systems, and computer-readable media are provided for facilitating mapping of semantically similar terms between and among two or more information systems. In particular, to facilitate automatic discovery, establishment, and/or statistical validation of linkages between a plurality of different nomenclatures employed by a plurality of information systems, such as multiple electronic health record systems. In embodiments, the imputation of latent synonymy in corpora comprised of samples of historical records from each system enables automated terminology mapping between disparate systems' records, thereby establishing reliable linkages that may subsequently be utilized for realtime decision support, data mining-based research, or other valuable purposes.