Automated Infiltrate Extraction from Radiology Reports
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Extracting diagnostic-related information from radiology reports, particularly chest x-ray reports, is challenging due to the text-based format and the need for skilled interpretation, which can introduce bias and is resource-intensive, and existing keyword searches are often insufficiently sensitive or specific.
Innovation Solution
A computer-implemented method and system that parses and classifies chest x-ray reports using infiltrate-related keywords, directional words, and negation words to extract and present infiltrate information on a dashboard for clinical decision-making, applying relevant clinical decision support algorithms in real-time for detecting and predicting acute illness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual interpretation of CXR reports by skilled readers is used, then diagnostic accuracy is improved, but resource consumption and time requirements increase
Solution Approach 1:
The patent replaces the mechanical process of manual interpretation by skilled readers with an automated natural language processing system. The system uses computational algorithms to parse, classify, and extract infiltrate information from CXR reports, substituting human cognitive labor with automated text analysis while maintaining diagnostic accuracy.
Solution Approach 2:
The system enables self-service by allowing the CXR report text to be automatically processed and classified without requiring manual intervention. The natural language processing system autonomously identifies infiltrate-related keywords, determines their presence or absence, and structures the information, making the diagnostic data readily available without human resource consumption.
2Productivity
If simple keyword searches are used to extract infiltrate information, then processing speed is improved, but sensitivity and specificity decrease
Solution Approach 1:
The patent segments the keyword search process into multiple hierarchical levels: first identifying infiltrate-related keywords, then checking for negation words, and finally determining the presence or absence of infiltrates based on the combination of these elements. This segmented approach maintains processing speed while significantly improving sensitivity and specificity compared to simple keyword matching.
Solution Approach 2:
The system dynamically adjusts the search strategy by first performing a broad keyword search to identify potential infiltrate mentions, then applying negation word checks to refine results. This dynamic two-stage process optimizes both speed and accuracy by adapting the level of analysis based on initial findings.
3Measurement precision
If comprehensive keyword coverage is increased to improve sensitivity, then detection capability is improved, but false positive rate increases
Solution Approach 1:
The patent applies preliminary anti-action by first searching for infiltrate-related keywords to identify potential positives, then immediately following up with a search for negation words to counteract and eliminate false positives. This preemptive approach to error correction maintains high sensitivity while reducing the false positive rate by systematically canceling out erroneous detections.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system and method for extracting information on infiltrates from an imaging report, and applying the report infiltrate information to clinical guidelines and clinical support applications. The system and method perform the steps of receiving the imaging report; parsing the report for infiltrate-related keywords; classifying the report based on the infiltrate-related keywords and at least one of directional words indicating a location of the infiltrates, negation words, and words canceling the negation words; presenting the classified report on a dashboard for clinical decision-making; and applying the classified report in a real-time clinical application.