README.md

metadata

language: en
license: mit
library_name: spacy
pipeline_tag: token-classification
tags:
  - spacy
  - token-classification
  - clinical-nlp
  - medication-ner
  - med7
  - word-vectors
model-index:
  - name: en_core_med7_lg
    results:
      - task:
          type: token-classification
          name: Medication named entity recognition
        dataset:
          name: Med7 held-out evaluation split
          type: private
        metrics:
          - type: f1
            value: 0.8886872353
            name: NER F1
          - type: precision
            value: 0.8796407186
            name: NER precision
          - type: recall
            value: 0.8979217604
            name: NER recall

Installation

pip install -U pip setuptools wheel
pip install "en-core-med7-lg @ https://huggingface.co/kormilitzin/en_core_med7_lg/resolve/main/en_core_med7_lg-1.1.0-py3-none-any.whl"

en_core_med7_lg

en_core_med7_lg is a non-transformer, large-vector spaCy pipeline for medication-related named entity recognition in English clinical text.

The model extracts the following entity types:

• DRUG
• STRENGTH
• DOSAGE
• DURATION
• FREQUENCY
• FORM
• ROUTE

Version

Current release: 1.1.0

This release was retrained and repackaged for the modern spaCy/transformers stack:

• Python 3.12.13
• spaCy 3.8.14

Unlike en_core_med7_trf, this model does not use a transformer backend and does not require spacy-transformers, transformers, tokenizers, or torch at runtime.

Citation

If you use this model, please cite the original Med7 paper:

Kormilitzin, A., Vaci, N., Liu, Q., & Nevado-Holgado, A. (2021). Med7: A transferable clinical natural language processing model for electronic health records. Artificial Intelligence in Medicine, 118, 102086. https://doi.org/10.1016/j.artmed.2021.102086

BibTeX:

@article{kormilitzin2021med7,
  title = {Med7: A transferable clinical natural language processing model for electronic health records},
  author = {Kormilitzin, Andrey and Vaci, Nemanja and Liu, Qiang and Nevado-Holgado, Alejo},
  journal = {Artificial Intelligence in Medicine},
  volume = {118},
  pages = {102086},
  year = {2021},
  doi = {10.1016/j.artmed.2021.102086},
  publisher = {Elsevier}
}