Launching into clinical space with medspaCy: a new clinical text processing toolkit in Python.

Area of Science:

• Natural Language Processing
• Computational Linguistics
• Bioinformatics

Background:

• Machine learning shows success in clinical natural language processing (cNLP).
• Rule-based approaches remain important for clinical text analysis.
• Existing tools may lack flexibility in integrating diverse NLP methods.

Purpose of the Study:

• Introduce medSpaCy, an open-source library for clinical NLP.
• Enable flexible integration of rule-based and machine learning algorithms.
• Facilitate custom pipeline development for clinical text.

Main Methods:

• Developed medSpaCy based on the spaCy framework.
• Incorporated components for context analysis and terminology mapping.
• Leveraged spaCy's conventions for easy integration and customization.

Main Results:

• medSpaCy offers an extensible platform for cNLP.
• The library supports flexible integration of rule-based and ML algorithms.
• Components address common cNLP tasks like context analysis and mapping.

Conclusions:

• medSpaCy provides a versatile toolkit for clinical text processing.
• It enables rapid development of custom NLP pipelines for healthcare data.
• The library supports hybrid approaches combining rule-based and ML methods.