Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Llettuce: An Open Source Natural Language Processing Tool for the Translation of Medical Terms into Uniform Clinical Encoding

James Mitchell-White, Reza Omdivar, Esmond Urwin, Karthikeyan Sivakumar, Ruizhe Li, Andy Rae, Xiaoyan Wang, Theresia Mina, John Chambers, Grazziela Figueredo, Philip R Quinlan

TL;DR

Llettuce leverages advanced natural language processing, including large language models and fuzzy matching, to automate and enhance the mapping process, and can be deployed locally, ensuring data protection while maintaining high performance in converting informal medical terms to standardised concepts.

Abstract

This paper introduces Llettuce, an open-source tool designed to address the complexities of converting medical terms into OMOP standard concepts. Unlike existing solutions such as the Athena database search and Usagi, which struggle with semantic nuances and require substantial manual input, Llettuce leverages advanced natural language processing, including large language models and fuzzy matching, to automate and enhance the mapping process. Developed with a focus on GDPR compliance, Llettuce can be deployed locally, ensuring data protection while maintaining high performance in converting informal medical terms to standardised concepts.

Llettuce: An Open Source Natural Language Processing Tool for the Translation of Medical Terms into Uniform Clinical Encoding

TL;DR

Llettuce leverages advanced natural language processing, including large language models and fuzzy matching, to automate and enhance the mapping process, and can be deployed locally, ensuring data protection while maintaining high performance in converting informal medical terms to standardised concepts.

Abstract

This paper introduces Llettuce, an open-source tool designed to address the complexities of converting medical terms into OMOP standard concepts. Unlike existing solutions such as the Athena database search and Usagi, which struggle with semantic nuances and require substantial manual input, Llettuce leverages advanced natural language processing, including large language models and fuzzy matching, to automate and enhance the mapping process. Developed with a focus on GDPR compliance, Llettuce can be deployed locally, ensuring data protection while maintaining high performance in converting informal medical terms to standardised concepts.

Paper Structure

This paper contains 17 sections, 4 figures, 3 tables.

Table of Contents

  1. Introduction
  2. System Architecture
  3. Access via UI or HTTP
  4. GUI
  5. Input
  6. Natural Language Processing Pipeline --- The Llettuce API
  7. Vector search
  8. LLM
  9. Concept Matches
  10. Output
  11. Case Study: Medication Dataset
  12. Data Description
  13. Experimental Design
  14. Results
  15. Comparison between vector search and Usagi
  16. ...and 2 more sections

Figures (4)

  • Figure 1: Natural language processing architecture pipeline
  • Figure 2: Sankey diagram of outputs from the LLettuce NLP pipeline
  • Figure 3: Comparison of results between GPT-3 and Llettuce
  • Figure 4: Inference times (run on macOS, 2.8GHz quad-core Intel i7, 16 Gb RAM)