Hyperlinked Radiology Reports via NLP Image Reference Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Radiologists face a time-consuming and laborious process when navigating to specific images referenced in prior radiology reports, leading to potential compromises in diagnostic quality due to the manual navigation required in current workflows.
Innovation Solution
A system and method that automatically extracts image references from narrative radiology reports using Natural Language Processing (NLP) and classifier modules to generate hyperlinked diagnostic reports, enabling easy navigation to corresponding images within the radiology information system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual navigation to referenced images is performed, then diagnostic quality can be maintained, but time consumption and labor increase significantly
Solution Approach 1:
The patent introduces an intermediary system (image reference extraction module, hyperlink generation module) that mediates between the radiology report text and the PACS image database. This intermediary automatically extracts image references from narrative reports, generates hyperlinks, and enables direct navigation to referenced images, eliminating the need for manual navigation while maintaining diagnostic quality.
Solution Approach 2:
The patent replaces the mechanical manual navigation process with an automated information processing system. Instead of manually opening PACS, navigating series, and selecting images, the system uses NLP to extract image references from text and automatically generates hyperlinks that directly link to the referenced images, substituting manual mechanical actions with automated digital processing.
2Measurement precision
If manual navigation to referenced images is performed, then image accuracy can be ensured, but operational complexity and time consumption increase
Solution Approach 1:
The system enables self-service navigation where the radiology report itself contains embedded hyperlinks to the referenced images. The report automatically serves as both the diagnostic documentation and the navigation guide, eliminating the need for separate manual navigation operations while ensuring accurate image reference through the automated extraction and hyperlink generation processes.
3Productivity
If automated image reference extraction is implemented, then navigation efficiency improves, but system complexity increases
Solution Approach 1:
The patent implements a multi-functional integrated system where the same infrastructure (processing modules, database connections, hyperlink generation) serves multiple purposes: extracting image references from reports, generating hyperlinks, enabling direct PACS navigation, and maintaining diagnostic documentation. This universal approach consolidates functionality to manage complexity while improving productivity.
Data Source
AI summary
A system, method and computer readable storage medium for retrieving a narrative report for at least one study including a plurality of images of a patient from a memory, determining text structure boundaries to identify and classify each text structure in the narrative report, determining image references in each text structure of the narrative report, extracting image references from text structures classified as including an image reference and determining a study to which an extracted image reference corresponds.