Biomedical pretrained language model for Spanish. For more details about the corpus, the pretraining and the evaluation, check the official repository and read our preprint.
Intended uses and limitations
The model is ready-to-use only for masked language modelling to perform the Fill Mask task (try the inference API or read the next section). However, it is intended to be fine-tuned on downstream tasks such as Named Entity Recognition or Text Classification.
How to use
python
1from transformers import AutoTokenizer, AutoModelForMaskedLM
2tokenizer = AutoTokenizer.from_pretrained("BSC-TeMU/roberta-base-biomedical-es")3model = AutoModelForMaskedLM.from_pretrained("BSC-TeMU/roberta-base-biomedical-es")4from transformers import pipeline
5unmasker = pipeline('fill-mask', model="BSC-TeMU/roberta-base-biomedical-es")6unmasker("El único antecedente personal a reseñar era la <mask> arterial.")
# Output
[
{
"sequence": " El único antecedente personal a reseñar era la hipertensión arterial.",
"score": 0.9855039715766907,
"token": 3529,
"token_str": " hipertensión"
},
{
"sequence": " El único antecedente personal a reseñar era la diabetes arterial.",
"score": 0.0039140828885138035,
"token": 1945,
"token_str": " diabetes"
},
{
"sequence": " El único antecedente personal a reseñar era la hipotensión arterial.",
"score": 0.002484665485098958,
"token": 11483,
"token_str": " hipotensión"
},
{
"sequence": " El único antecedente personal a reseñar era la Hipertensión arterial.",
"score": 0.0023484621196985245,
"token": 12238,
"token_str": " Hipertensión"
},
{
"sequence": " El único antecedente personal a reseñar era la presión arterial.",
"score": 0.0008009297889657319,
"token": 2267,
"token_str": " presión"
}
]
Training
Tokenization and model pretraining
This model is a RoBERTa-based model trained on a
biomedical corpus in Spanish collected from several sources (see next section).
The training corpus has been tokenized using a byte version of Byte-Pair Encoding (BPE)
used in the original RoBERTA model with a vocabulary size of 52,000 tokens. The pretraining consists of a masked language model training at the subword level following the approach employed for the RoBERTa base model with the same hyperparameters as in the original work. The training lasted a total of 48 hours with 16 NVIDIA V100 GPUs of 16GB DDRAM, using Adam optimizer with a peak learning rate of 0.0005 and an effective batch size of 2,048 sentences.
Training corpora and preprocessing
The training corpus is composed of several biomedical corpora in Spanish, collected from publicly available corpora and crawlers.
To obtain a high-quality training corpus, a cleaning pipeline with the following operations has been applied:
data parsing in different formats
sentence splitting
language detection
filtering of ill-formed sentences
deduplication of repetitive contents
keep the original document boundaries
Finally, the corpora are concatenated and further global deduplication among the corpora have been applied.
The result is a medium-size biomedical corpus for Spanish composed of about 963M tokens. The table below shows some basic statistics of the individual cleaned corpora:
Crawler of more than 3,000 URLs belonging to Spanish biomedical and health domains.
Clinical cases misc.
102,855,267
A miscellany of medical content, essentially clinical cases. Note that a clinical case report is a scientific publication where medical practitioners share patient cases and it is different from a clinical note or document.
Biomedical Abbreviation Recognition and Resolution (BARR2) containing Spanish clinical case study sections from a variety of clinical disciplines.
Wikipedia_life_sciences
13,890,501
Wikipedia articles crawled 04/01/2021 with the Wikipedia API python library starting from the "Ciencias_de_la_vida" category up to a maximum of 5 subcategories. Multiple links to the same articles are then discarded to avoid repeating content.
Patents
13,463,387
Google Patent in Medical Domain for Spain (Spanish). The accepted codes (Medical Domain) for Json files of patents are: "A61B", "A61C","A61F", "A61H", "A61K", "A61L","A61M", "A61B", "A61P".
Spanish-side articles extracted from a collection of Spanish-English parallel corpus consisting of biomedical scientific literature. The collection of parallel resources are aggregated from the MedlinePlus source.
PubMed
1,858,966
Open-access articles from the PubMed repository crawled in 2017.
Evaluation
The model has been evaluated on the Named Entity Recognition (NER) using the following datasets:
ICTUSnet: consists of 1,006 hospital discharge reports of patients admitted for stroke from 18 different Spanish hospitals. It contains more than 79,000 annotations for 51 different kinds of variables.
The evaluation results are compared against the mBERT and BETO models:
F1 - Precision - Recall
roberta-base-biomedical-es
mBERT
BETO
PharmaCoNER
89.48 - 87.85 - 91.18
87.46 - 86.50 - 88.46
88.18 - 87.12 - 89.28
CANTEMIST
83.87 - 81.70 - 86.17
82.61 - 81.12 - 84.15
82.42 - 80.91 - 84.00
ICTUSnet
88.12 - 85.56 - 90.83
86.75 - 83.53 - 90.23
85.95 - 83.10 - 89.02
Additional information
Author
Text Mining Unit (TeMU) at the Barcelona Supercomputing Center (bsc-temu@bsc.es)
This work was funded by the Spanish State Secretariat for Digitalization and Artificial Intelligence (SEDIA) within the framework of the Plan-TL.
Citation information
If you use our models, please cite our latest preprint:
bibtex
12@misc{carrino2021biomedical,
3 title={Biomedical and Clinical Language Models for Spanish: On the Benefits of Domain-Specific Pretraining in a Mid-Resource Scenario},
4 author={Casimiro Pio Carrino and Jordi Armengol-Estapé and Asier Gutiérrez-Fandiño and Joan Llop-Palao and Marc Pàmies and Aitor Gonzalez-Agirre and Marta Villegas},
5 year={2021},
6 eprint={2109.03570},
7 archivePrefix={arXiv},
8 primaryClass={cs.CL}
9}
10
If you use our Medical Crawler corpus, please cite the preprint:
bibtex
12@misc{carrino2021spanish,
3 title={Spanish Biomedical Crawled Corpus: A Large, Diverse Dataset for Spanish Biomedical Language Models},
4 author={Casimiro Pio Carrino and Jordi Armengol-Estapé and Ona de Gibert Bonet and Asier Gutiérrez-Fandiño and Aitor Gonzalez-Agirre and Martin Krallinger and Marta Villegas},
5 year={2021},
6 eprint={2109.07765},
7 archivePrefix={arXiv},
8 primaryClass={cs.CL}
9}
10
Disclaimer
Click to expand
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.
In no event shall the owner of the models (SEDIA – State Secretariat for Digitalization and Artificial Intelligence) nor the creator (BSC – Barcelona Supercomputing Center) be liable for any results arising from the use made by third parties of these models.
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.
En ningún caso el propietario de los modelos (SEDIA – Secretaría de Estado de Digitalización e Inteligencia Artificial) ni el creador (BSC – Barcelona Supercomputing Center) serán responsables de los resultados derivados del uso que hagan terceros de estos modelos.
