| --- |
| dataset_info: |
| - config_name: anatomy-ct |
| features: |
| - name: series_id |
| dtype: string |
| - name: filename |
| dtype: string |
| - name: slice_location |
| dtype: int64 |
| - name: pixel_spacing_x |
| dtype: float64 |
| - name: pixel_spacing_y |
| dtype: float64 |
| - name: slice_thickness |
| dtype: float64 |
| - name: patient_id |
| dtype: string |
| - name: target |
| dtype: int64 |
| - name: modality |
| dtype: string |
| - name: raw_target |
| dtype: string |
| - name: original_slice_ID |
| dtype: string |
| - name: image |
| list: |
| list: float32 |
| - name: mask |
| list: |
| list: bool |
| splits: |
| - name: train |
| num_bytes: 9223954377 |
| num_examples: 36408 |
| - name: val |
| num_bytes: 462480337 |
| num_examples: 1867 |
| - name: test |
| num_bytes: 659601942 |
| num_examples: 2525 |
| download_size: 3737518544 |
| dataset_size: 10346036656 |
| - config_name: anatomy-mri |
| features: |
| - name: series_id |
| dtype: string |
| - name: filename |
| dtype: string |
| - name: slice_location |
| dtype: int32 |
| - name: pixel_spacing_x |
| dtype: float32 |
| - name: pixel_spacing_y |
| dtype: float32 |
| - name: slice_thickness |
| dtype: float32 |
| - name: patient_id |
| dtype: string |
| - name: target |
| dtype: int32 |
| - name: modality |
| dtype: string |
| - name: raw_target |
| dtype: string |
| - name: original_slice_ID |
| dtype: string |
| - name: image |
| list: |
| list: int16 |
| - name: mask |
| list: |
| list: bool |
| splits: |
| - name: train |
| num_bytes: 4090044515 |
| num_examples: 14197 |
| - name: val |
| num_bytes: 474039930 |
| num_examples: 1559 |
| - name: test |
| num_bytes: 374132018 |
| num_examples: 1259 |
| download_size: 2915576386 |
| dataset_size: 4938216463 |
| - config_name: covidx-ct |
| features: |
| - name: series_id |
| dtype: string |
| - name: filename |
| dtype: string |
| - name: slice_location |
| dtype: int64 |
| - name: pixel_spacing_x |
| dtype: float64 |
| - name: pixel_spacing_y |
| dtype: float64 |
| - name: slice_thickness |
| dtype: float64 |
| - name: patient_id |
| dtype: string |
| - name: target |
| dtype: int64 |
| - name: modality |
| dtype: string |
| - name: raw_target |
| dtype: string |
| - name: original_slice_ID |
| dtype: string |
| - name: mask_filename |
| dtype: string |
| - name: image |
| list: |
| list: float32 |
| - name: mask |
| dtype: 'null' |
| splits: |
| - name: train |
| num_bytes: 37812895523 |
| num_examples: 35748 |
| - name: val |
| num_bytes: 3547646273 |
| num_examples: 3367 |
| - name: test |
| num_bytes: 3581713974 |
| num_examples: 3374 |
| download_size: 9242613609 |
| dataset_size: 44942255770 |
| - config_name: deep-lesion-site |
| features: |
| - name: series_id |
| dtype: string |
| - name: filename |
| dtype: string |
| - name: slice_location |
| dtype: int64 |
| - name: pixel_spacing_x |
| dtype: float64 |
| - name: pixel_spacing_y |
| dtype: float64 |
| - name: slice_thickness |
| dtype: float64 |
| - name: modality |
| dtype: string |
| - name: patient_id |
| dtype: string |
| - name: target |
| dtype: int64 |
| - name: raw_target |
| dtype: string |
| - name: image |
| list: |
| list: float32 |
| - name: mask |
| list: |
| list: float32 |
| splits: |
| - name: train |
| num_bytes: 5505019129 |
| num_examples: 2610 |
| - name: val |
| num_bytes: 2566286625 |
| num_examples: 1220 |
| - name: test |
| num_bytes: 2565764632 |
| num_examples: 1221 |
| download_size: 1845165457 |
| dataset_size: 10637070386 |
| - config_name: emidec-classification-mask |
| features: |
| - name: series_id |
| dtype: string |
| - name: filename |
| dtype: string |
| - name: slice_location |
| dtype: int32 |
| - name: pixel_spacing_x |
| dtype: float32 |
| - name: pixel_spacing_y |
| dtype: float32 |
| - name: slice_thickness |
| dtype: float32 |
| - name: patient_id |
| dtype: string |
| - name: target |
| dtype: int32 |
| - name: modality |
| dtype: string |
| - name: raw_target |
| dtype: string |
| - name: original_slice_ID |
| dtype: string |
| - name: image |
| list: |
| list: int16 |
| - name: mask |
| list: |
| list: uint8 |
| splits: |
| - name: train |
| num_bytes: 75412392 |
| num_examples: 562 |
| - name: val |
| num_bytes: 8854211 |
| num_examples: 73 |
| - name: test |
| num_bytes: 8279163 |
| num_examples: 72 |
| download_size: 45267403 |
| dataset_size: 92545766 |
| - config_name: ixi |
| features: |
| - name: series_id |
| dtype: string |
| - name: filename |
| dtype: string |
| - name: slice_location |
| dtype: int32 |
| - name: pixel_spacing_x |
| dtype: float32 |
| - name: pixel_spacing_y |
| dtype: float32 |
| - name: slice_thickness |
| dtype: float32 |
| - name: patient_id |
| dtype: string |
| - name: target |
| dtype: float32 |
| - name: modality |
| dtype: string |
| - name: raw_target |
| dtype: string |
| - name: original_slice_ID |
| dtype: string |
| - name: image |
| list: |
| list: int16 |
| - name: mask |
| dtype: 'null' |
| splits: |
| - name: train |
| num_bytes: 332222990 |
| num_examples: 4279 |
| - name: val |
| num_bytes: 68222176 |
| num_examples: 880 |
| - name: test |
| num_bytes: 80255956 |
| num_examples: 1034 |
| download_size: 328877695 |
| dataset_size: 480701122 |
| - config_name: kits |
| features: |
| - name: series_id |
| dtype: string |
| - name: filename |
| dtype: string |
| - name: slice_location |
| dtype: int32 |
| - name: pixel_spacing_x |
| dtype: float32 |
| - name: pixel_spacing_y |
| dtype: float32 |
| - name: slice_thickness |
| dtype: float32 |
| - name: patient_id |
| dtype: string |
| - name: target |
| dtype: int32 |
| - name: modality |
| dtype: string |
| - name: raw_target |
| dtype: string |
| - name: original_slice_ID |
| dtype: string |
| - name: image |
| list: |
| list: int16 |
| - name: mask |
| list: |
| list: bool |
| splits: |
| - name: train |
| num_bytes: 182285449 |
| num_examples: 324 |
| - name: val |
| num_bytes: 37042750 |
| num_examples: 66 |
| - name: test |
| num_bytes: 80820588 |
| num_examples: 144 |
| download_size: 190639279 |
| dataset_size: 300148787 |
| - config_name: kneeMRI |
| features: |
| - name: series_id |
| dtype: string |
| - name: filename |
| dtype: string |
| - name: slice_location |
| dtype: int32 |
| - name: pixel_spacing_x |
| dtype: float32 |
| - name: pixel_spacing_y |
| dtype: float32 |
| - name: slice_thickness |
| dtype: float32 |
| - name: patient_id |
| dtype: string |
| - name: target |
| dtype: int32 |
| - name: modality |
| dtype: string |
| - name: raw_target |
| dtype: string |
| - name: original_slice_ID |
| dtype: string |
| - name: image |
| list: |
| list: |
| list: int16 |
| - name: mask |
| list: |
| list: |
| list: int16 |
| splits: |
| - name: train |
| num_bytes: 1502718092 |
| num_examples: 733 |
| - name: val |
| num_bytes: 188416480 |
| num_examples: 92 |
| - name: test |
| num_bytes: 188659984 |
| num_examples: 92 |
| download_size: 430646992 |
| dataset_size: 1879794556 |
| - config_name: luna16 |
| features: |
| - name: series_id |
| dtype: string |
| - name: filename |
| dtype: string |
| - name: slice_location |
| dtype: int32 |
| - name: pixel_spacing_x |
| dtype: float32 |
| - name: pixel_spacing_y |
| dtype: float32 |
| - name: slice_thickness |
| dtype: float32 |
| - name: patient_id |
| dtype: string |
| - name: target |
| dtype: int32 |
| - name: modality |
| dtype: string |
| - name: raw_target |
| dtype: string |
| - name: original_slice_ID |
| dtype: string |
| - name: image |
| list: |
| list: int16 |
| - name: mask |
| list: |
| list: int16 |
| splits: |
| - name: train |
| num_bytes: 355835106 |
| num_examples: 338 |
| - name: val |
| num_bytes: 177917547 |
| num_examples: 169 |
| - name: test |
| num_bytes: 178970321 |
| num_examples: 170 |
| download_size: 230576160 |
| dataset_size: 712722974 |
| - config_name: luna16-3D |
| features: |
| - name: series_id |
| dtype: string |
| - name: filename |
| dtype: string |
| - name: slice_location |
| dtype: int32 |
| - name: pixel_spacing_x |
| dtype: float32 |
| - name: pixel_spacing_y |
| dtype: float32 |
| - name: slice_thickness |
| dtype: float32 |
| - name: patient_id |
| dtype: string |
| - name: target |
| dtype: int32 |
| - name: modality |
| dtype: string |
| - name: raw_target |
| dtype: string |
| - name: original_slice_ID |
| dtype: string |
| - name: image |
| list: |
| list: |
| list: int16 |
| - name: mask |
| list: |
| list: |
| list: int16 |
| splits: |
| - name: train |
| num_bytes: 4254440674 |
| num_examples: 338 |
| - name: val |
| num_bytes: 2127220331 |
| num_examples: 169 |
| - name: test |
| num_bytes: 2139807441 |
| num_examples: 170 |
| download_size: 2287903515 |
| dataset_size: 8521468446 |
| - config_name: oasis |
| features: |
| - name: series_id |
| dtype: string |
| - name: filename |
| dtype: string |
| - name: slice_location |
| dtype: int64 |
| - name: pixel_spacing_x |
| dtype: float64 |
| - name: pixel_spacing_y |
| dtype: float64 |
| - name: slice_thickness |
| dtype: float64 |
| - name: patient_id |
| dtype: string |
| - name: target |
| dtype: int64 |
| - name: modality |
| dtype: string |
| - name: raw_target |
| dtype: string |
| - name: original_slice_ID |
| dtype: string |
| - name: mask_filename |
| dtype: string |
| - name: image |
| list: |
| list: |
| list: float32 |
| - name: mask |
| dtype: 'null' |
| splits: |
| - name: train |
| num_bytes: 14653591629 |
| num_examples: 348 |
| - name: val |
| num_bytes: 1852752976 |
| num_examples: 44 |
| - name: test |
| num_bytes: 1852752955 |
| num_examples: 44 |
| download_size: 2989008001 |
| dataset_size: 18359097560 |
| configs: |
| - config_name: anatomy-ct |
| data_files: |
| - split: train |
| path: anatomy-ct/train-* |
| - split: val |
| path: anatomy-ct/val-* |
| - split: test |
| path: anatomy-ct/test-* |
| - config_name: anatomy-mri |
| data_files: |
| - split: train |
| path: anatomy-mri/train-* |
| - split: val |
| path: anatomy-mri/val-* |
| - split: test |
| path: anatomy-mri/test-* |
| - config_name: covidx-ct |
| data_files: |
| - split: train |
| path: covidx-ct/train-* |
| - split: val |
| path: covidx-ct/val-* |
| - split: test |
| path: covidx-ct/test-* |
| - config_name: deep-lesion-site |
| data_files: |
| - split: train |
| path: deep-lesion-site/train-* |
| - split: val |
| path: deep-lesion-site/val-* |
| - split: test |
| path: deep-lesion-site/test-* |
| - config_name: emidec-classification-mask |
| data_files: |
| - split: train |
| path: emidec-classification-mask/train-* |
| - split: val |
| path: emidec-classification-mask/val-* |
| - split: test |
| path: emidec-classification-mask/test-* |
| - config_name: ixi |
| data_files: |
| - split: train |
| path: ixi/train-* |
| - split: val |
| path: ixi/val-* |
| - split: test |
| path: ixi/test-* |
| - config_name: kits |
| data_files: |
| - split: train |
| path: kits/train-* |
| - split: val |
| path: kits/val-* |
| - split: test |
| path: kits/test-* |
| - config_name: kneeMRI |
| data_files: |
| - split: train |
| path: kneeMRI/train-* |
| - split: val |
| path: kneeMRI/val-* |
| - split: test |
| path: kneeMRI/test-* |
| - config_name: luna16 |
| data_files: |
| - split: train |
| path: luna16/train-* |
| - split: val |
| path: luna16/val-* |
| - split: test |
| path: luna16/test-* |
| - config_name: luna16-3D |
| data_files: |
| - split: train |
| path: luna16-3D/train-* |
| - split: val |
| path: luna16-3D/val-* |
| - split: test |
| path: luna16-3D/test-* |
| - config_name: oasis |
| data_files: |
| - split: train |
| path: oasis/train-* |
| - split: val |
| path: oasis/val-* |
| - split: test |
| path: oasis/test-* |
| license: cc-by-nc-sa-4.0 |
| --- |
| |
|
|
|
|
| <div align="center"> |
| <img src="https://cdn-uploads.huggingface.co/production/uploads/62cdea59a9be5c195561c2b8/JaS4YslW9wFR8dZ7LMawz.png" width="40%" alt="Raidium" /> |
| </div> |
| <hr> |
|
|
|
|
| <p align="center"> |
| <a href="https://github.com/raidium-med/curia"><b>🌟 Github</b></a> | |
| <a href="https://arxiv.org/abs/2509.06830"><b>📄 Paper Link</b></a> | |
| <a href="https://raidium.eu/blog/curia-benchmark/"><b>🌐 Blog post</b></a> |
| |
| </p> |
| <h2> |
| <p align="center"> |
| <h1 align="center">CuriaBench; Benchmark for the paper Curia: A Multi-Modal Foundation Model for Radiology</h1> |
| </p> |
| </h2> |
|
|
| <div align="center"> |
| <img src="https://cdn-uploads.huggingface.co/production/uploads/62cdea59a9be5c195561c2b8/RJLhOQBQSyqgkoDWd5pLt.png" width="40%" alt="Raidium" /> |
| </div> |
| <hr> |
|
|
| ## Usage |
|
|
| This repo contains evaluation datasets fro mthe Curia paper https://arxiv.org/abs/2509.06830. |
|
|
| You can run the curia evaluation pipeline using our github repo at https://github.com/raidium-med/curia. |
|
|
| The Curia-B model is available on huggingface: https://huggingface.co/raidium/curia |
|
|
| ## Acknowledgments |
|
|
| The benchmark has been built from the following public datasets: |
|
|
|
|
| - CT Organ Recognition — TotalSegmentator (CT) — Wasserthal et al., Nat. Commun., 2023. https://github.com/wasserth/TotalSegmentator |
| - MRI Organ Recognition — TotalSegmentator MRI — D’Antonoli et al., arXiv:2405.19492, 2024. https://arxiv.org/abs/2405.19492 |
| - Cross-Modality Organ Recognition — TotalSegmentator CT & MRI — Wasserthal et al., Nat. Commun., 2023. https://github.com/wasserth/TotalSegmentator |
| - Neuroimaging Age Estimation — IXI Dataset — Brain Development Project. https://brain-development.org/ixi-dataset/ |
| - Lung Nodule Malignancy — LUNA16 — Setio et al., Med. Image Anal., 2017. https://luna16.grand-challenge.org/ |
| - Kidney Lesion Malignancy — KiTS23 — Heller et al., arXiv:2307.01984, 2023. https://kits-challenge.org/ |
| - Tumor Localisation — DeepLesion — Yan et al., J. Med. Imaging, 2018. https://nihcc.app.box.com/v/DeepLesion |
| - ACL Tear — KneeMRI dataset — Štajduhar et al., Comput. Methods Programs Biomed., 2017. https://zenodo.org/records/14789903 |
| - Myocardial Infarction — EMIDEC — MICCAI 2020 Challenge. https://emidec.com/ |
| - Stroke — ATLAS v2.0 — Liew et al., Front. Neuroinform., 2022. https://fcon_1000.projects.nitrc.org/indi/retro/atlas.html |
| - Alzheimer’s Disease — OASIS-1 — Marcus et al., J. Cogn. Neurosci., 2007. https://www.oasis-brains.org/ |
| - Pulmonary Infections — COVIDx CT — Gunraj et al., Front. Med., 2021. https://github.com/haydengunraj/COVIDNet-CT/blob/master/docs/dataset.md |
| |
| |
| |
| ## Cite our paper |
| ``` |
| @article{dancette2025curia, |
| title={Curia: A Multi-Modal Foundation Model for Radiology}, |
| author={Dancette, Corentin and Khlaut, Julien and Saporta, Antoine and Philippe, Helene and Ferreres, Elodie and Callard, Baptiste and Danielou, Th{\'e}o and Alberge, L{\'e}o and Machado, L{\'e}o and Tordjman, Daniel and others}, |
| journal={arXiv preprint arXiv:2509.06830}, |
| year={2025} |
| } |
| ``` |
