File size: 11,759 Bytes
243267c 2c0f341 243267c 2c0f341 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 |
---
widget:
- text: "The p53 protein plays a crucial role in tumor suppression."
- text: "Expression of BRCA1 gene was significantly upregulated in breast tissue."
- text: "The NF-kB pathway regulates inflammatory responses."
- text: "Activation of the STAT3 signaling pathway is observed in many cancers."
- text: "The experiment involved transfecting HeLa cells with a plasmid containing the target gene."
tags:
- token-classification
- named-entity-recognition
- biomedical-nlp
- transformers
- protein-recognition
- gene-recognition
- molecular-biology
- genomics
- dna
- rna
- cell_line
- cell_type
- protein
language:
- en
license: apache-2.0
---
# 𧬠[OpenMed-NER-DNADetect-SuperMedical-125M](https://huggingface.co/OpenMed/OpenMed-NER-DNADetect-SuperMedical-125M)
**Specialized model for Biomedical Entity Recognition - Proteins, DNA, RNA, cell lines, and cell types**
[](https://opensource.org/licenses/Apache-2.0)
[]()
[]()
[](https://huggingface.co/OpenMed)
## π Model Overview
This model is a **state-of-the-art** fine-tuned transformer engineered to deliver **enterprise-grade accuracy** for biomedical entity recognition - proteins, dna, rna, cell lines, and cell types. This specialized model excels at identifying and extracting biomedical entities from clinical texts, research papers, and healthcare documents, enabling applications such as **drug interaction detection**, **medication extraction from patient records**, **adverse event monitoring**, **literature mining for drug discovery**, and **biomedical knowledge graph construction** with **production-ready reliability** for clinical and research applications.
### π― Key Features
- **High Precision**: Optimized for biomedical entity recognition
- **Domain-Specific**: Trained on curated JNLPBA dataset
- **Production-Ready**: Validated on clinical benchmarks
- **Easy Integration**: Compatible with Hugging Face Transformers ecosystem
### π·οΈ Supported Entity Types
This model can identify and classify the following biomedical entities:
- `B-DNA`
- `B-RNA`
- `B-cell_line`
- `B-cell_type`
- `B-protein`
<details>
<summary>See 5 more entity types...</summary>
- `I-DNA`
- `I-RNA`
- `I-cell_line`
- `I-cell_type`
- `I-protein`
</details>
## π Dataset
JNLPBA corpus focuses on biomedical named entity recognition for protein, DNA, RNA, cell line, and cell type entities.
The JNLPBA (Joint Workshop on Natural Language Processing in Biomedicine and its Applications) corpus is a widely-used biomedical NER dataset derived from the GENIA corpus for the 2004 bio-entity recognition task. It contains annotations for five entity types: protein, DNA, RNA, cell line, and cell type, making it essential for molecular biology and genomics research applications. The corpus consists of MEDLINE abstracts annotated with biomedical entities relevant to gene and protein recognition tasks. It has been extensively used as a benchmark for evaluating biomedical NER systems and continues to be a standard evaluation dataset for developing machine learning models in computational biology and bioinformatics.
## π Performance Metrics
### Current Model Performance
- **F1 Score**: `0.80`
- **Precision**: `0.76`
- **Recall**: `0.86`
- **Accuracy**: `0.93`
### π Comparative Performance on JNLPBA Dataset
| Rank | Model | F1 Score | Precision | Recall | Accuracy |
|------|-------|----------|-----------|--------|-----------|
| π₯ 1 | [OpenMed-NER-DNADetect-SuperClinical-434M](https://huggingface.co/OpenMed/OpenMed-NER-DNADetect-SuperClinical-434M) | **0.8188** | 0.7778 | 0.8643 | 0.9320 |
| π₯ 2 | [OpenMed-NER-DNADetect-SuperMedical-355M](https://huggingface.co/OpenMed/OpenMed-NER-DNADetect-SuperMedical-355M) | **0.8177** | 0.7716 | 0.8697 | 0.9318 |
| π₯ 3 | [OpenMed-NER-DNADetect-MultiMed-568M](https://huggingface.co/OpenMed/OpenMed-NER-DNADetect-MultiMed-568M) | **0.8157** | 0.7758 | 0.8599 | 0.9354 |
| 4 | [OpenMed-NER-DNADetect-BigMed-560M](https://huggingface.co/OpenMed/OpenMed-NER-DNADetect-BigMed-560M) | **0.8134** | 0.7723 | 0.8591 | 0.9346 |
| 5 | [OpenMed-NER-DNADetect-BioClinical-108M](https://huggingface.co/OpenMed/OpenMed-NER-DNADetect-BioClinical-108M) | **0.8071** | 0.7632 | 0.8562 | 0.9147 |
| 6 | [OpenMed-NER-DNADetect-MultiMed-335M](https://huggingface.co/OpenMed/OpenMed-NER-DNADetect-MultiMed-335M) | **0.8069** | 0.7642 | 0.8547 | 0.9185 |
| 7 | [OpenMed-NER-DNADetect-PubMed-335M](https://huggingface.co/OpenMed/OpenMed-NER-DNADetect-PubMed-335M) | **0.8056** | 0.7611 | 0.8556 | 0.9344 |
| 8 | [OpenMed-NER-DNADetect-SuperClinical-184M](https://huggingface.co/OpenMed/OpenMed-NER-DNADetect-SuperClinical-184M) | **0.8053** | 0.7548 | 0.8630 | 0.9259 |
| 9 | [OpenMed-NER-DNADetect-BioPatient-108M](https://huggingface.co/OpenMed/OpenMed-NER-DNADetect-BioPatient-108M) | **0.8052** | 0.7605 | 0.8555 | 0.9137 |
| 10 | [OpenMed-NER-DNADetect-SuperMedical-125M](https://huggingface.co/OpenMed/OpenMed-NER-DNADetect-SuperMedical-125M) | **0.8044** | 0.7589 | 0.8557 | 0.9284 |
*Rankings based on F1-score performance across all models trained on this dataset.*

*Figure: OpenMed (Open-Source) vs. Latest SOTA (Closed-Source) performance comparison across biomedical NER datasets.*
## π Quick Start
### Installation
```bash
pip install transformers torch
```
### Usage
```python
from transformers import pipeline
# Load the model and tokenizer
# Model: https://huggingface.co/OpenMed/OpenMed-NER-DNADetect-SuperMedical-125M
model_name = "OpenMed/OpenMed-NER-DNADetect-SuperMedical-125M"
# Create a pipeline
medical_ner_pipeline = pipeline(
model=model_name,
aggregation_strategy="simple"
)
# Example usage
text = "The p53 protein plays a crucial role in tumor suppression."
entities = medical_ner_pipeline(text)
print(entities)
token = entities[0]
print(text[token["start"] : token["end"]])
```
NOTE: The `aggregation_strategy` parameter defines how token predictions are grouped into entities. For a detailed explanation, please refer to the [Hugging Face documentation](https://huggingface.co/docs/transformers/en/main_classes/pipelines#transformers.TokenClassificationPipeline.aggregation_strategy).
Here is a summary of the available strategies:
- **`none`**: Returns raw token predictions without any aggregation.
- **`simple`**: Groups adjacent tokens with the same entity type (e.g., `B-LOC` followed by `I-LOC`).
- **`first`**: For word-based models, if tokens within a word have different entity tags, the tag of the first token is assigned to the entire word.
- **`average`**: For word-based models, this strategy averages the scores of tokens within a word and applies the label with the highest resulting score.
- **`max`**: For word-based models, the entity label from the token with the highest score within a word is assigned to the entire word.
### Batch Processing
For efficient processing of large datasets, use proper batching with the `batch_size` parameter:
```python
texts = [
"The p53 protein plays a crucial role in tumor suppression.",
"Expression of BRCA1 gene was significantly upregulated in breast tissue.",
"The NF-kB pathway regulates inflammatory responses.",
"Activation of the STAT3 signaling pathway is observed in many cancers.",
"The experiment involved transfecting HeLa cells with a plasmid containing the target gene.",
]
# Efficient batch processing with optimized batch size
# Adjust batch_size based on your GPU memory (typically 8, 16, 32, or 64)
results = medical_ner_pipeline(texts, batch_size=8)
for i, entities in enumerate(results):
print(f"Text {i+1} entities:")
for entity in entities:
print(f" - {entity['word']} ({entity['entity_group']}): {entity['score']:.4f}")
```
### Large Dataset Processing
For processing large datasets efficiently:
```python
from transformers.pipelines.pt_utils import KeyDataset
from datasets import Dataset
import pandas as pd
# Load your data
# Load a medical dataset from Hugging Face
from datasets import load_dataset
# Load a public medical dataset (using a subset for testing)
medical_dataset = load_dataset("BI55/MedText", split="train[:100]") # Load first 100 examples
data = pd.DataFrame({"text": medical_dataset["Completion"]})
dataset = Dataset.from_pandas(data)
# Process with optimal batching for your hardware
batch_size = 16 # Tune this based on your GPU memory
results = []
for out in medical_ner_pipeline(KeyDataset(dataset, "text"), batch_size=batch_size):
results.extend(out)
print(f"Processed {len(results)} texts with batching")
```
### Performance Optimization
**Batch Size Guidelines:**
- **CPU**: Start with batch_size=1-4
- **Single GPU**: Try batch_size=8-32 depending on GPU memory
- **High-end GPU**: Can handle batch_size=64 or higher
- **Monitor GPU utilization** to find the optimal batch size for your hardware
**Memory Considerations:**
```python
# For limited GPU memory, use smaller batches
medical_ner_pipeline = pipeline(
model=model_name,
aggregation_strategy="simple",
device=0 # Specify GPU device
)
# Process with memory-efficient batching
for batch_start in range(0, len(texts), batch_size):
batch = texts[batch_start:batch_start + batch_size]
batch_results = medical_ner_pipeline(batch, batch_size=len(batch))
results.extend(batch_results)
```
## π Dataset Information
- **Dataset**: JNLPBA
- **Description**: Biomedical Entity Recognition - Proteins, DNA, RNA, cell lines, and cell types
### Training Details
- **Base Model**: roberta-base
- **Training Framework**: Hugging Face Transformers
- **Optimization**: AdamW optimizer with learning rate scheduling
- **Validation**: Cross-validation on held-out test set
## π¬ Model Architecture
- **Base Architecture**: roberta-base
- **Task**: Token Classification (Named Entity Recognition)
- **Labels**: Dataset-specific entity types
- **Input**: Tokenized biomedical text
- **Output**: BIO-tagged entity predictions
## π‘ Use Cases
This model is particularly useful for:
- **Clinical Text Mining**: Extracting entities from medical records
- **Biomedical Research**: Processing scientific literature
- **Drug Discovery**: Identifying chemical compounds and drugs
- **Healthcare Analytics**: Analyzing patient data and outcomes
- **Academic Research**: Supporting biomedical NLP research
## π License
Licensed under the Apache License 2.0. See [LICENSE](https://www.apache.org/licenses/LICENSE-2.0) for details.
## π€ Contributing
We welcome contributions of all kinds! Whether you have ideas, feature requests, or want to join our mission to advance open-source Healthcare AI, we'd love to hear from you.
Follow [OpenMed Org](https://huggingface.co/OpenMed) on Hugging Face π€ and click "Watch" to stay updated on our latest releases and developments.
## Citation
If you use this model in your research or applications, please cite the following paper:
```latex
@misc{panahi2025openmedneropensourcedomainadapted,
title={OpenMed NER: Open-Source, Domain-Adapted State-of-the-Art Transformers for Biomedical NER Across 12 Public Datasets},
author={Maziyar Panahi},
year={2025},
eprint={2508.01630},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2508.01630},
}
```
Proper citation helps support and acknowledge my work. Thank you! |