Getting started

EDS-NLP provides a set of spaCy components that are used to extract information from clinical notes written in French.

If it's your first time with spaCy, we recommend you familiarise yourself with some of their key concepts by looking at the "spaCy 101" page.

Quick start

Installation

You can install EDS-NLP via pip:

fast →pip install edsnlpSuccessfully installed!

restart ↻

We recommend pinning the library version in your projects, or use a strict package manager like Poetry.

pip install edsnlp==0.8.0

A first pipeline

Once you've installed the library, let's begin with a very simple example that extracts mentions of COVID19 in a text, and detects whether they are negated.

import spacy

nlp = spacy.blank("fr")  # We only need spaCy's French tokenizer.


terms = dict(
    covid=["covid", "coronavirus"],  # This example terminology provides a very simple, and by no means exhaustive, list of synonyms for COVID19.

)

# Sentencizer component, needed for negation detection
nlp.add_pipe("eds.sentences")  # In spaCy, pipelines are added via the nlp.add_pipe method. EDS-NLP pipelines are automatically discovered by spaCy.

# Matcher component
nlp.add_pipe("eds.matcher", config=dict(terms=terms))  # See the matching tutorial for mode details.

# Negation detection
nlp.add_pipe("eds.negation")

# Process your text in one call !
doc = nlp("Le patient est atteint de covid")

doc.ents  # spaCy stores extracted entities in the Doc.ents attribute.

# Out: (covid,)

doc.ents[0]._.negation  # The eds.negation pipeline has added a negation custom attribute.

# Out: False

This example is complete, it should run as-is. Check out the spaCy 101 page if you're not familiar with spaCy.

Available pipeline components

CoreQualifiersMiscellaneousNERTrainable

Pipeline	Description
`eds.normalizer`	Non-destructive input text normalisation
`eds.sentences`	Better sentence boundary detection
`eds.matcher`	A simple yet powerful entity extractor
`eds.terminology`	A simple yet powerful terminology matcher
`eds.contextual-matcher`	A conditional entity extractor
`eds.endlines`	An unsupervised model to classify each end line

Pipeline	Description
`eds.negation`	Rule-based negation detection
`eds.family`	Rule-based family context detection
`eds.hypothesis`	Rule-based speculation detection
`eds.reported_speech`	Rule-based reported speech detection
`eds.history`	Rule-based medical history detection

Pipeline	Description
`eds.dates`	Date extraction and normalisation
`eds.consultation_dates`	Identify consultation dates
`eds.measurements`	Measure extraction and normalisation
`eds.sections`	Section detection
`eds.reason`	Rule-based hospitalisation reason detection
`eds.tables`	Tables detection

Pipeline	Description
`eds.covid`	A COVID mentions detector
`eds.charlson`	A Charlson score extractor
`eds.sofa`	A SOFA score extractor
`eds.emergency.priority`	A priority score extractor
`eds.emergency.ccmu`	A CCMU score extractor
`eds.emergency.gemsa`	A GEMSA score extractor
`eds.TNM`	A TNM score extractor
`eds.cim10`	A CIM10 terminology matcher
`eds.drugs`	A Drug mentions extractor
`eds.adicap`	A ADICAP codes extractor
`eds.umls`	A UMLS terminology matcher

Pipeline	Description
`eds.nested-ner`	Nested and overlapping named entity recogntion
`eds.span-qualifier`	A trainable component for multi-class multi-label span qualification

Disclaimer

The performances of an extraction pipeline may depend on the population and documents that are considered.

Contributing to EDS-NLP

We welcome contributions ! Fork the project and propose a pull request. Take a look at the dedicated page for detail.

Citation

If you use EDS-NLP, please cite us as below.

@misc{edsnlp,
  author = {Dura, Basile and Wajsburt, Perceval and Petit-Jean, Thomas and Cohen, Ariel and Jean, Charline and Bey, Romain},
  doi    = {10.5281/zenodo.6424993},
  title  = {EDS-NLP: efficient information extraction from French clinical notes},
  url    = {http://aphp.github.io/edsnlp}
}