Radiology Report Structuring

RadExtract: Radiology Report Structuring Demo

warning This demonstration is for illustrative purposes only to show the baseline capabilities of LangExtract, the library that powers this demo. It does not represent a finished or approved product, is not intended to diagnose or suggest treatment for any disease or condition, and should not be used for medical advice.

LangExtract (LX) is a multi-purpose NLP extraction library that uses large language models such as Gemini to convert free-text into schema-controlled data, with every datum linked back to its exact words in the source.

Demo Overview

This demo uses LangExtract with Gemini 2.5 Flash or Gemini 2.5 Pro to structure radiology report findings into organized data. Using a comprehensive prompt with task descriptions and select examples across different imaging modalities, LangExtract processes the text and produces the structured output displayed below. Hover over any item to highlight its source text. Pre-generated Gemini 2.5 Pro cache responses are available for faster evaluation and reduced server load. Clinical significance is color-coded with yellow underlines for general findings and red underlines for significant findings.

Clinical Background:
Structured reporting helps ensure completeness, reduces ambiguity, and facilitates data sharing in radiology. For background on the value of structured radiology reports, see this European Society of Radiology paper.

Interface Controls

LX Generated Prompt • LX Structured Output • Use Cache expand_more

RadExtract: Radiology Report Structuring Demo

Demo Overview

Try Sample Reports

Interface Controls

Input

Output