# Dataset Preprocessing Sources

## MIMIC-CXR
- Source: [MIMIC-CXR-JPG](https://physionet.org/content/mimic-cxr-jpg/2.1.0/)

## ChestXDet10, ChestXray14, CheXpert, OpenI, PadChest
- Preprocessed files from [CARZero (CVPR 2024)](https://github.com/laihaoran/CARZero) with minor path modifications in CSV files

## SIIM
- Preprocessing from [MGCA (NeurIPS 2022)](https://github.com/HKU-MedAI/MGCA/blob/main/mgca/preprocess/siim.py)

## RSNA
- Preprocessed file from [MedKLIP (ICCV 2023)](https://github.com/MediaBrain-SJTU/MedKLIP/blob/main/Sample_Zero-Shot_Grounding_RSNA/data_sample/test.csv)
- Only file paths in CSV files were modified

## MS-CXR
- Dataset: [MS-CXR](https://physionet.org/content/ms-cxr/0.1/)
- Test split: [MedRPG (MICCAI 2023)](https://github.com/eraserNut/MedRPG/tree/master/data/MS_CXR)