Natural Language Processing of Large Numbers of Radiology Reports in a Public Health System to Extract Structured Data, With a Test Case of CT KUB

Key Points

• The study aims to evaluate the effectiveness of the RATA pipeline in extracting structured data from radiology reports.
• Utilized the RATA pipeline for natural language processing of radiology reports.
• Focused on a large volume of reports within a public health system.
• Conducted a case study involving CT KUB reports to test the data extraction capabilities.
• Demonstrated acceptable performance in extracting structured data.
• Identified variations in the clinical use of CT KUB based on extracted data.

Abstract