Modelpublic

DT4H/CardioBERTa.nl_P_enriched

sourceHugging Faceupdated 1mo agoView on Hugging Face

0likes8downloads

Model Card

DT4HCardioBERTaparentsnlenriched

DT4H_CardioBERTa_parents_nl_enriched is a Dutch biomedical terminology encoder for clinical concept normalization and entity linking. It is initialized from [UMCU/CardioBERTa.nl] and specialized using CUI-supervised terminology pairs and metric learning.

Backbone

The backbone belongs to the CardioBERTa family from CardioLM - a multilingual suite of small language models for the cardiology domain. CardioBERTa comprises language-specific encoder models adapted to cardiology through continued pretraining on monolingual biomedical and cardiology-related corpora using Masked Language Modeling (MLM). The family covers Czech, Dutch, English, Italian, Romanian, Spanish and Swedish.

Training


Language	Dutch (`nl`)
Triplet collection	`enriched`
Strategy	`parents`
Objective	Multi-Similarity Loss
Mining	All triplets, margin 0.2
Pooling	CLS
Epochs	1
Batch size	256
Learning rate	2e-5
Max. length	25

CUI-supervised terminology pairs enriched with parent-level ontology relations.

Terminology statistics

Strategy	Triplets	CUIs	Unique terms	Unique positives	Terms/CUI	Δ terms
synonyms	79,985	79,985	156,019	79,788	1.98	0
parents	1,677,259	476,525	543,012	405,646	3.98	+386,993
grandparents	4,898,584	476,972	543,296	463,560	9.94	+387,277

This model uses 1,677,259 triplets, covering 476,525 CUIs and 543,012 unique normalized terms.

The training terminology is not distributed with this repository because it contains resources subject to UMLS licensing conditions. Only aggregate statistics are released.

Intended use

The model is intended for terminology embedding, biomedical candidate retrieval, concept normalization and entity linking, particularly in cardiology and clinical NLP pipelines. It is not intended for direct clinical decision-making.

Usage

python

import torch
import torch.nn.functional as F
from transformers import AutoModel, AutoTokenizer

model_id = "DT4H/DT4H_CardioBERTa_parents_nl_enriched"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModel.from_pretrained(model_id)

inputs = tokenizer(
    "clinical concept",
    return_tensors="pt",
    truncation=True,
    max_length=25,
)

with torch.no_grad():
    output = model(**inputs)

embedding = F.normalize(
    output.last_hidden_state[:, 0, :],
    p=2,
    dim=1,
)

Reference

Danu et al. CardioLM - a multilingual suite of small language models for the cardiology domain.

Developed within the DataTools4Heart (DT4H) project, Grant Agreement 101057849.