Diabetes - EDS-NLP

[giscus[bot]]

Diabetes - EDS-NLP

https://aphp.github.io/edsnlp/latest/pipes/ner/disorders/diabetes/

[0dey]

[Insulin and Type]

If your looking for all insulin and type status, in addition to extracting assigned values, you should look at the entity. For example, "dt2", "did" or "dnid" will not be included into the assigned attribute.

Example:

import edsnlp, edsnlp.pipes as eds

nlp = edsnlp.blank("eds")
nlp.add_pipe(eds.sentences())
nlp.add_pipe(
    eds.normalizer(
        accents=True,
        lowercase=True,
        quotes=True,
        spaces=True,
        pollution=dict(
            information=True,
            bars=True,
            biology=True,
            doctors=True,
            web=True,
            coding=True,
            footer=True,
        ),
    ),
)
nlp.add_pipe(eds.diabetes())

text = "Le patient a un did"
doc = nlp(text)
span_diab = doc.spans["diabetes"]

print(span_diab[0]._.assigned) 
# {}

text = "Le patient a un dt2"
doc = nlp(text)
span_diab = doc.spans["diabetes"]

print(span_diab[0]._.assigned) 
# {}

text = "Le patient a un diabète insulino-dépendant"
doc = nlp(text)
span_diab = doc.spans["diabetes"]

print(span_diab[0]._.assigned) 
# {'insulin': [dépendant]}

A solution is to look directly in the span for the corresponding regex patterns that can be found in the code (regex=[r"\bds?n?id\b", r"\bdt.?(i|ii|1|2)\b"])

[mtech419]

good