Clinical Context Coding Graphs for Automated Document Extraction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The integration of clinical data across different healthcare systems is hindered by interoperability issues and lack of standards, leading to inefficient processing and management of clinical documents.

Innovation Solution

A method for context-based clinical knowledge extraction using a classifier to identify the clinical context of a document, followed by an executable coding graph to generate structured clinical information, which includes a network of branch and coding nodes for semantic analysis and data assignment, enabling automated data extraction and transmission.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If manual processing and management of clinical documents is used, then data accuracy can be maintained through human review, but processing efficiency and productivity are significantly reduced

Engineering Contradiction:
Improveprocessing efficiencyVSAvoidmanual intervention level
Core Design Contradiction:
ProductivityVSExtent of automation

Solution Approach 1:

The system performs self-service by automatically classifying clinical documents, extracting structured data, and routing documents without requiring manual human intervention. The classifier and coding graphs work autonomously to process documents, generating structured clinical information and determining appropriate recipients for transmission.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

Manual mechanical processing of clinical documents is replaced with an automated computational system. The patent substitutes human reviewers and manual data extraction processes with a computer-implemented system that uses machine learning classifiers and executable coding graphs to automatically process, classify, and extract data from clinical documents.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Adaptability or versatility

If different healthcare systems use different protocols and communication methods, then system flexibility and adaptability are maintained, but data integration and interoperability become difficult

Engineering Contradiction:
Improvesystem flexibilityVSAvoiddata integration capability
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The system achieves universality by creating a standardized processing framework that can handle multiple types of clinical documents across different healthcare systems. The classifier is trained on diverse document types, and the executable coding graphs are designed to work with various clinical contexts, enabling the system to process referral letters, test results, discharge summaries, and other document types uniformly.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system adapts to different clinical contexts by dynamically changing processing parameters based on document classification. Once a document is classified into a specific clinical context, the system selects appropriate coding graphs and extraction rules tailored to that context, allowing flexible adaptation to different protocols while maintaining consistent processing standards.

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If comprehensive semantic analysis is performed on clinical documents, then data extraction accuracy is improved, but processing time and computational resources increase

Engineering Contradiction:
Improvedata extraction accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The semantic analysis process is segmented into discrete executable coding graphs that are selected based on document classification. Rather than applying comprehensive analysis to all documents uniformly, the system divides processing into context-specific pathways, applying only the necessary analysis steps for each document type, thereby reducing overall processing time while maintaining accuracy for each specific context.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary classification of documents before executing detailed semantic analysis. By first identifying the clinical context and document type, the system can pre-select the appropriate coding graphs and extraction rules, avoiding unnecessary analysis steps and reducing processing time while ensuring accurate extraction through context-appropriate methods.

Inventive Principle:
Principle #10Preliminary action

4Reliability

If structured clinical information is generated through automated processing, then interoperability and data integration are improved, but system complexity and development requirements increase

Engineering Contradiction:
Improvedata integration capabilityVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The complex system is segmented into distinct modular components: a document classifier, a library of executable coding graphs, and a structured data output generator. Each component has a specific function and can be independently developed, tested, and maintained. The coding graphs themselves are segmented into context-specific modules that can be selectively executed, reducing overall system complexity while enabling comprehensive data integration.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4697197A1Context-based clinical knowledge extraction and document transmission
Publication Date: 2026.02.18 DYAD AI INC
  • EP4697197A1 patent drawingFigure 1A
  • EP4697197A1 patent drawingFigure 1B
  • EP4697197A1 patent drawingFigure 1C

AI summary

A method for context-based clinical knowledge extraction and automatic transmission of clinical documents. A text-based representation of a document having a clinical context is obtained and an identifier which uniquely identifies the clinical context of the document is determined. An executable coding graph is identified from a plurality of executable coding graphs based on the identifier of the clinical context. The executable coding graph is indicative of a procedure for coding the document according to the clinical context and comprises a network of branch nodes interconnected with a plurality of coding nodes thereby forming a directed acyclic graph. The executable coding graph is executed on the text based representation of the document thereby generating a structured set of clinical information linked to the document enabling the automatic transmission of a clinical document based on the data extracted from the clinical document.