language:
- es
tags:
- biomedical
- clinical
- EHR
- spanish
- anonymization
license: apache-2.0
metrics:
- precision
- recall
- f1
base_model:
- PlanTL-GOB-ES/bsc-bio-ehr-es
model-index:
- name: BSC-NLP4BIA/bsc-bio-ehr-es-meddocan
results:
- task:
type: token-classification
dataset:
name: MEDDOCAN
type: MEDDOCAN
metrics:
- name: precision (micro)
type: precision
value: 0.95
- name: recall (micro)
type: recall
value: 0.972
- name: f1 (micro)
type: f1
value: 0.961
- task:
type: token-classification
dataset:
name: carmen-anonymization
type: carmen-anonymization
metrics:
- name: precision (micro)
type: precision
value: 0.804
- name: recall (micro)
type: recall
value: 0.818
- name: f1 (micro)
type: f1
value: 0.811
widget:
- text: >-
El diagnóstico definitivo de nuestro paciente fue de un Adenocarcinoma de
pulmón cT2a cN3 cM1a Estadio IV (por una única lesión pulmonar
contralateral) PD-L1 90%, EGFR negativo, ALK negativo y ROS-1 negativo.
- text: >-
Durante el ingreso se realiza una TC, observándose un nódulo pulmonar en
el LII y una masa renal derecha indeterminada. Se realiza punción biopsia
del nódulo pulmonar, con hallazgos altamente sospechosos de carcinoma.
- text: >-
Trombosis paraneoplásica con sospecha de hepatocarcinoma por imagen, sobre
hígado cirrótico, en paciente con índice Child-Pugh B.
Spanish RoBERTa-base biomedical model finetuned for the Named Entity Recognition (NER) task on the MEDDOCAN dataset.
Table of contents
Click to expand
Model description
A fine-tuned version of the bsc-bio-ehr-es model on the MEDDOCAN corpus.
For further information, check the official website.
Intended uses and limitations
TDB
How to use
⚠ We recommend pre-tokenizing the input text into words instead of providing it directly to the model, as this is how the model was trained. Otherwise, the results and performance might get affected.
A usage example can be found here.
Limitations and bias
At the time of submission, no measures have been taken to estimate the bias embedded in the model. However, we are well aware that our models may be biased since the corpora have been collected using crawling techniques on multiple web sources. We intend to conduct research in these areas in the future, and if completed, this model card will be updated.
Training
The model was trained using the Barcelona Supercomputing Center infrastructure.
Evaluation
F1 Score: 0.961 on MEDDOCAN
F1 Score: 0.811 on CARMEN-I Anonymization
Additional information
Authors
NLP4BIA team at the Barcelona Supercomputing Center ([email protected]).
Contact information
jan.rodriguez [at] bsc.es
Licensing information
Funding
This project was partially funded by the Spanish Plan for the Advancement of Language Technology (Plan TL) in collaboration with the Barcelona Supercomputing Center (BSC) and the Hospital Clinic de Barcelona (HCB). On the BSC's side, we acknowledge additional funding by the Spanish National AI4ProfHealth project (PID2020-119266RA-I00 MICIU/AEI/10.13039/501100011033) and EU Horizon projects (AI4HF 101080430 and DataTools4Heart 101057849). On the HCB's side, the project was also supported by the Instituto de Salud Carlos III (ISCIII).
Citing information
Please cite the following works:
@inproceedings{meddocan,
title={{Automatic De-identification of Medical Texts in Spanish: the MEDDOCAN Track, Corpus, Guidelines, Methods and Evaluation of Results}},
author={Marimon, Montserrat and Gonzalez-Agirre, Aitor and Intxaurrondo, Ander and Villegas, Marta and Krallinger, Martin},
booktitle="Proceedings of the Iberian Languages Evaluation Forum (IberLEF 2019)",
year={2019}
}
@misc{carmen_physionet,
author = {Farre Maduell, Eulalia and Lima-Lopez, Salvador and Frid, Santiago Andres and Conesa, Artur and Asensio, Elisa and Lopez-Rueda, Antonio and Arino, Helena and Calvo, Elena and Bertran, Maria Jesús and Marcos, Maria Angeles and Nofre Maiz, Montserrat and Tañá Velasco, Laura and Marti, Antonia and Farreres, Ricardo and Pastor, Xavier and Borrat Frigola, Xavier and Krallinger, Martin},
title = {{CARMEN-I: A resource of anonymized electronic health records in Spanish and Catalan for training and testing NLP tools (version 1.0.1)}},
year = {2024},
publisher = {PhysioNet},
url = {https://doi.org/10.13026/x7ed-9r91}
}
@article{physionet,
author = {Ary L. Goldberger and Luis A. N. Amaral and Leon Glass and Jeffrey M. Hausdorff and Plamen Ch. Ivanov and Roger G. Mark and Joseph E. Mietus and George B. Moody and Chung-Kang Peng and H. Eugene Stanley },
title = {PhysioBank, PhysioToolkit, and PhysioNet },
journal = {Circulation},
volume = {101},
number = {23},
pages = {e215-e220},
year = {2000},
doi = {10.1161/01.CIR.101.23.e215},
URL = {https://www.ahajournals.org/doi/abs/10.1161/01.CIR.101.23.e215}
}
Disclaimer
The models published in this repository are intended for a generalist purpose and are available to third parties. These models may have bias and/or any other undesirable distortions.
When third parties, deploy or provide systems and/or services to other parties using any of these models (or using systems based on these models) or become users of the models, they should note that it is their responsibility to mitigate the risks arising from their use and, in any event, to comply with applicable regulations, including regulations regarding the use of artificial intelligence.
Los modelos publicados en este repositorio tienen una finalidad generalista y están a disposición de terceros. Estos modelos pueden tener sesgos y/u otro tipo de distorsiones indeseables.
Cuando terceros desplieguen o proporcionen sistemas y/o servicios a otras partes usando alguno de estos modelos (o utilizando sistemas basados en estos modelos) o se conviertan en usuarios de los modelos, deben tener en cuenta que es su responsabilidad mitigar los riesgos derivados de su uso y, en todo caso, cumplir con la normativa aplicable, incluyendo la normativa en materia de uso de inteligencia artificial.