multiversity is a Python library for creating customized subsets of the MultiCaRe Dataset — an open-source multimodal clinical case dataset with 98,000+ clinical cases and 139,000+ medical images.
pip install multiversityThe MultiCaRe Dataset is an open-source collection of de-identified clinical cases and medical images built from PubMed Central open-access case reports. It includes:
- 98,000+ clinical cases across multiple medical specialties
- 139,000+ labeled medical images
- A hierarchical image taxonomy with 140+ classes
- Rich metadata: patient demographics, clinical narratives, image captions, and citations
from multiversity.multicare_dataset import MedicalDatasetCreator
# Downloads the dataset from Zenodo on first run (5–10 min)
mdc = MedicalDatasetCreator(directory='medical_datasets')filters = [
{'field': 'min_age', 'string_list': ['18']},
{'field': 'gender', 'string_list': ['Male']},
{'field': 'case_strings', 'string_list': ['tumor', 'cancer'], 'operator': 'any'},
{'field': 'label', 'string_list': ['mri', 'head']}
]mdc.create_dataset(
dataset_name='brain_tumor_dataset',
filter_list=filters,
dataset_type='multimodal' # Options: 'multimodal', 'text', 'image', 'case_series'
)mdc.display_example()| Field | Description | Example |
|---|---|---|
min_age |
Minimum patient age | ['18'] |
max_age |
Maximum patient age | ['65'] |
gender |
Patient gender | ['Male'], ['Female'] |
case_strings |
Keywords in the clinical narrative | ['tumor', 'cancer'] |
caption |
Keywords in image captions | ['mass', 'lesion'] |
label |
Image taxonomy labels | ['mri', 'head'] |
Use 'operator': 'any' to match any keyword, or 'operator': 'all' (default) to require all.
| Type | Description |
|---|---|
multimodal |
Cases with both text and images |
text |
Cases with clinical narratives only |
image |
Image classification dataset |
case_series |
Groups of related cases |
If you use this library or the MultiCaRe dataset, please cite:
Nievas Offidani, M., Roffet, F., González Galtier, M. C., Massiris, M., & Delrieux, C. (2025).
An Open-Source Clinical Case Dataset for Medical Image Classification and Multimodal AI Applications.
Data, 10(8), 123. https://doi.org/10.3390/data10080123Contributions are welcome! Open an issue or submit a pull request.
If you find this useful, consider giving it a ⭐ — it helps with visibility.
For questions or collaborations, reach out on LinkedIn.