README.md

metadata

language: en
pipeline_tag: text-classification
tags:
  - biogpt
  - medical
  - radiology
  - birads
  - classification
license: mit
metrics:
  - accuracy
  - f1
model-index:
  - name: BioGPT BI-RADS Classifier
    results:
      - task:
          type: text-classification
          name: BI-RADS Classification
        metrics:
          - type: accuracy
            value: 0.9736
            name: Accuracy
          - type: f1
            value: 0.9066
            name: F1 (Macro)

BioGPT BI-RADS Classifier

This model is a fine-tuned version of microsoft/biogpt-large for BI-RADS classification of radiology reports.

Model Description

• Base Model: microsoft/biogpt-large
• Task: Multi-class text classification (BI-RADS categories 0-6)
• Training Data: Radiology reports with BI-RADS annotations
• Accuracy: 97.36%
• F1-Score (Macro): 90.66%

Performance

Overall Metrics

• Accuracy: 97.36%
• F1-Score (Macro): 90.66%
• F1-Score (Weighted): 97.34%
• Precision (Macro): 92.65%
• Recall (Macro): 88.96%

Per-Class Performance

BI-RADS	Precision	Recall	F1-Score	Support
0	0.9946	0.9482	0.9708	193
1	0.9504	0.9664	0.9583	119
2	0.9740	0.9943	0.9840	527
3	1.0000	0.8333	0.9091	18
4	0.9000	0.8182	0.8571	11
5	0.6667	0.6667	0.6667	3
6	1.0000	1.0000	1.0000	1

Usage

from transformers import AutoTokenizer, BioGptForSequenceClassification
import torch

# Load model and tokenizer
model = BioGptForSequenceClassification.from_pretrained("ishro/biogpt-aura")
tokenizer = AutoTokenizer.from_pretrained("ishro/biogpt-aura")

# Prepare input
report_text = "Your radiology report text here..."
inputs = tokenizer(report_text, return_tensors="pt", padding=True, truncation=True, max_length=512)

# Get prediction
with torch.no_grad():
    outputs = model(**inputs)
    predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
    predicted_class = torch.argmax(predictions, dim=-1).item()

# Map to BI-RADS label
birads_label = model.config.id2label[predicted_class]
print(f"Predicted BI-RADS: {birads_label}")
print(f"Confidence: {predictions[0][predicted_class].item():.4f}")

Training Details

Training Hyperparameters

• Learning Rate: 2e-5
• Batch Size: 4 per device (2 GPUs)
• Gradient Accumulation Steps: 8
• Effective Batch Size: 64
• Epochs: 3
• Optimizer: AdamW (fused)
• Mixed Precision: BF16
• Hardware: 2x NVIDIA L40S (46GB each)

Training Data

The model was trained on radiology reports with the following features:

• Report observations
• Conclusions
• Recommendations
• Patient metadata (age, hormonal therapy, family history, etc.)

Limitations

• Performance on BI-RADS categories 5 and 6 is lower due to limited training samples
• Model is trained on specific radiology report format
• May not generalize well to reports from different institutions without fine-tuning

Ethical Considerations

• This model is intended for research purposes and should not be used as the sole basis for clinical decisions
• Always consult with qualified medical professionals for diagnosis and treatment
• The model may have biases based on the training data distribution

Citation

If you use this model, please cite:

@misc{biogpt-birads-classifier,
  author = {Your Name},
  title = {BioGPT BI-RADS Classifier},
  year = {2025},
  publisher = {HuggingFace},
  url = {https://huggingface.co/ishro/biogpt-aura}
}

Model Card Authors

ishro

Model Card Contact

For questions or issues, please open an issue on the model repository.