|
|
--- |
|
|
language: en |
|
|
tags: |
|
|
- medical |
|
|
- ner |
|
|
- named-entity-recognition |
|
|
- healthcare |
|
|
- i2b2 |
|
|
license: apache-2.0 |
|
|
datasets: |
|
|
- i2b2-2018 |
|
|
metrics: |
|
|
- f1 |
|
|
- precision |
|
|
- recall |
|
|
--- |
|
|
|
|
|
# i2b2 2018 Medical NER Model |
|
|
|
|
|
This model is fine-tuned for medical Named Entity Recognition (NER) using the i2b2 2018 dataset. |
|
|
|
|
|
## Model Details |
|
|
|
|
|
- **Task**: Named Entity Recognition (NER) |
|
|
- **Domain**: Medical/Healthcare |
|
|
- **Dataset**: i2b2 2018 |
|
|
- **Model Type**: Token Classification |
|
|
|
|
|
## Usage |
|
|
|
|
|
```python |
|
|
from transformers import pipeline |
|
|
|
|
|
# Load the model |
|
|
ner_pipeline = pipeline( |
|
|
"ner", |
|
|
model="prakharsinghAI/i2b2-ner-model", |
|
|
aggregation_strategy="simple" |
|
|
) |
|
|
|
|
|
# Example usage |
|
|
text = "Patient was prescribed aspirin 100mg twice daily for headache." |
|
|
results = ner_pipeline(text) |
|
|
print(results) |
|
|
``` |
|
|
|
|
|
## Training Details |
|
|
|
|
|
- **Dataset**: i2b2 2018 Medical NER |
|
|
- **Task**: Token Classification |
|
|
- **Labels**: Medical entities (Drug, Dosage, Route, etc.) |
|
|
|
|
|
## Performance |
|
|
|
|
|
This model was trained on the i2b2 2018 dataset for medical named entity recognition tasks. |
|
|
|
|
|
## Citation |
|
|
|
|
|
If you use this model, please cite the i2b2 2018 dataset: |
|
|
|
|
|
```bibtex |
|
|
@article{krallinger2015chemdner, |
|
|
title={The CHEMDNER corpus of chemicals and drugs and its annotation principles}, |
|
|
author={Krallinger, Martin and Rabal, Obdulia and Leitner, Florian and Vazquez, Miguel and Salgado, David and Lu, Zhiyong and Leaman, Robert and Lu, Yanan and Ji, Donghong and Lowe, Daniel M and others}, |
|
|
journal={Journal of cheminformatics}, |
|
|
volume={7}, |
|
|
number={1}, |
|
|
pages={1--17}, |
|
|
year={2015}, |
|
|
publisher={BioMed Central} |
|
|
} |
|
|
``` |
|
|
|
