Sci Five Radsum23

🚀 Impression section Generator For Radiology Reports 🏥

This model is the outcome of the SINAI team's participation in Task 1B: Radiology Report Summarization at the BioNLP workshop during ACL 2023. The aim of this task is to promote the development of automatic radiology report summarization systems and broaden their applicability by including seven different modalities and anatomies in the provided data. We propose to automate the generation of radiology impressions using "sequence - to - sequence" learning, which harnesses the power of publicly available pre - trained models, both from the general domain and those specific to the biomedical domain. This repository offers access to our top - performing system, which was obtained by fine - tuning Sci - Five base, a T5 model trained for an additional 200k steps to optimize it for biomedical literature.

🚀 Quick Start

There is no specific quick start content provided in the original README.

✨ Features

• The model is the result of the SINAI team's participation in a specific radiology report summarization task at a well - known workshop.
• It uses "sequence - to - sequence" learning and fine - tunes a pre - trained T5 model for biomedical literature.
• It aims to automate the generation of radiology impressions.

📦 Installation

There is no installation steps provided in the original README.

💻 Usage Examples

There is no code example provided in the original README.

📚 Documentation

Results

The official evaluation results demonstrate that adapting a general - domain system for biomedical literature is advantageous for subsequent fine - tuning for the radiology report summarization task. The table below summarizes the official scores achieved by this model during the official evaluation. Team standings are available here.

System description paper and citation

The paper with a detailed description of the system is published in the Proceedings of the 22st Workshop on Biomedical Language Processing.

BibTeX citation:

@inproceedings{chizhikova-etal-2023-sinai,
    title = "{SINAI} at {R}ad{S}um23: Radiology Report Summarization Based on Domain-Specific Sequence-To-Sequence Transformer Model",
    author = "Chizhikova, Mariia  and
      Diaz-Galiano, Manuel  and
      Urena-Lopez, L. Alfonso  and
      Martin-Valdivia, M. Teresa",
    booktitle = "The 22nd Workshop on Biomedical Natural Language Processing and BioNLP Shared Tasks",
    month = jul,
    year = "2023",
    address = "Toronto, Canada",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2023.bionlp-1.53",
    pages = "530--534",
    abstract = "This paper covers participation of the SINAI team in the shared task 1B: Radiology Report Summarization at the BioNLP workshop held on ACL 2023. Our proposal follows a sequence-to-sequence approach which leverages pre-trained multilingual general domain and monolingual biomedical domain pre-trained language models. The best performing system based on domain-specific model reached 33.96 F1RadGraph score which is the fourth best result among the challenge participants. This model was made publicly available on HuggingFace. We also describe an attempt of Proximal Policy Optimization Reinforcement Learning that was made in order to improve the factual correctness measured with F1RadGraph but did not lead to satisfactory results.",
}

🔧 Technical Details

There is no technical implementation details provided in the original README.

📄 License

The model is licensed under the afl - 3.0 license.

Additional Information

Tags

• summarization
• t5
• medical
• clinical

Datasets

• MIMIC - III

Widget Examples

• again noted is the large intraparenchymal hemorrhage in the posterior right frontal lobe with extension into both lateral ventricles. the degree of surrounding edema and effacement of adjacent sulci is unchanged. there is minor contralateral shift of normal midline structures. the ventricular size is unchanged. subarachnoid blood is now seen in the left frontal and parietal lobes, likely due to recirculation of the ventricular blood.
• a least two attempts were made at imaging, however, the study remains severely limited by patient motion. minimal hyperdensity tracks along a left parietal sulcus (2a:18) is equivocal for a small subarachnoid hemorhage. there is no large mass effect detected. there is no shift of normally midline structures. a minimally displaced zygomatic fracture is present (2a:9). the middle ear cavities, mastoid air cells are clear. there is extensive soft tissue swelling overlying the right frontal calvarium with swelling extending to the right preseptal soft tissues (2a:12). there is mild - moderate mucosal thickening within the ethmoid and maxillary sinuses with some fluid and fluid mucosal thickening in the sphenoid sinus.

Inference Parameters

• max_length: 350

Metrics

• rouge - l