The Treasure Trove Hidden in Plain Sight: The Utility of GPT-4 in Chest Radiograph Evaluation

Prerequisites

The software is developed in Python 3.8.

Main Python modules required for the software can be installed from ./requirements:

$ conda env create -f requirements.yaml
$ conda activate radada

Note: This might take a few minutes.

Code structure

Our source code is available here.

1. ./main_RADADA.py: includes all the training and alysis codes.
2. ./statistics_RADADA.py: includes all the statistical analysis and evaluation metrics.

---

In case you use this repository, please cite the original paper:

S. Tayebi Arasteh, R. Siepmann, M. Huppertz, et al. The Treasure Trove Hidden in Plain Sight: The Utility of GPT-4 in Chest Radiograph Evaluation. 2024.

BibTex

@article {radada_arasteh,
  author = {Tayebi Arasteh, Soroosh and Siepmann, Robert and Huppertz, Marc and Lotfinia, Mahshad and Puladi, Behrus and Kuhl, Christiane and Truhn, Daniel and Nebelung, Sven},
  title = {The Treasure Trove Hidden in Plain Sight: The Utility of GPT-4 in Chest Radiograph Evaluation},
  year = {2024},
}