{"success":true,"data":{"author":"Evan Calabrese et al., University of California San Francisco (converted by MedOtter)","bucketAccessType":"public","bucketName":"MedOtter/UCSF-PDGM","bucketPrefix":"","category":"Health & Biomedical","cloudProvider":"huggingface","createdAt":"2026-11-14T14:15:12.096Z","description":"The University of California San Francisco Preoperative Diffuse Glioma MRI (UCSF-PDGM) dataset, published on The Cancer Imaging Archive, is a cohort of preoperative 3 T brain MRI from patients with histopathologically confirmed diffuse gliomas, with molecular and clinical annotation. This MedOtter repository is the BraTS-format structural subset: for each of 495 patients a folder with T1, post-contrast T1c, T2 and FLAIR volumes co-registered to the 240 x 240 x 155 BraTS grid (1 mm isotropic, skull-stripped) and a tumor segmentation with BraTS labels (necrosis or non-enhancing core, edema, enhancing tumor), produced by an ensemble of BraTS algorithms and corrected by neuroradiologists. The diffusion, perfusion, SWI and HARDI series of the full TCIA release are not included. UCSF-PDGM-metadata_v5.csv gives sex, age, WHO CNS grade, WHO 2021 diagnosis, IDH, MGMT, 1p/19q, survival, extent of resection and BraTS21 IDs; about 60 percent of cases overlap BraTS 2021, so exclude them when evaluating alongside BraTS. As measured 2026-11-14 the repository has 495 patient folders of NIfTI (.nii.gz) files (about 6.2 GB in all); read them with nibabel or SimpleITK and the metadata with pandas. TCIA releases the collection under CC BY 4.0; cite Calabrese et al., Radiology: Artificial Intelligence 2022.","displayTitle":"UCSF-PDGM: preoperative diffuse glioma MRI with tumor segmentations (BraTS-format structural subset)","doi":"10.7937/tcia.bdgf-8v37","id":"07ecad19-853c-5461-94e6-62242617659a","license":"CC-BY-4.0","licenseEvidenceUrl":"https://www.cancerimagingarchive.net/collection/ucsf-pdgm/","objectCountEstimate":2480,"probeStatus":"ok","referenceUrl":"https://doi.org/10.1148/ryai.220058","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"MedOtter/UCSF-PDGM\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"MedOtter/UCSF-PDGM\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=MedOtter, license:cc-by-4.0, size_categories:n<1K, format:parquet, downloads=4566"],"format":"mixed","generatedAt":"2026-11-14T14:15:16.508Z","layout":{"example":"UCSF-PDGM-metadata_glossary.csv","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"Term,Represents,Values\r\nID,\"DICOM (0010,0020) PatientID\",\r\nSex,\"DICOM (0010,0040) Patient Sex\",\"M,F\"\r\nAge at MRI,Age in years at time of MR imaging,\r\nWHO CNS Grade,Grade per the 2021 World Health Organization Classification of Tumors of the Central Nervous System (WHO CNS 2021) (https://doi.org/10.1093/neuonc/noab106),\"2,3,4\"\r\nFinal pathologic diagnosis (WHO 2021),Final (integrated) pathologic diagnosis per the 2021 World Health Organization Classification of Tumors of the Central Nervous System (WHO CNS 20","encoding":"utf8","key":"UCSF-PDGM-metadata_glossary.csv","truncatedAt":512}],"sampleObjects":["UCSF-PDGM-0004/UCSF-PDGM-0004_FLAIR.nii.gz","UCSF-PDGM-0004/UCSF-PDGM-0004_T1.nii.gz","UCSF-PDGM-0004/UCSF-PDGM-0004_T1c.nii.gz","UCSF-PDGM-0004/UCSF-PDGM-0004_T2.nii.gz","UCSF-PDGM-0004/UCSF-PDGM-0004_tumor_segmentation.nii.gz"],"sidecars":[{"bytes":"---\nlicense: cc-by-4.0\npretty_name: UCSF-PDGM (BraTS-format structural subset)\ntask_categories:\n- image-segmentation\ntags:\n- medical-imaging\n- mri\n- brain\n- glioma\n- brain-tumor\n- brats\n- tumor-segmentation\n- neuro-oncology\ndataset_info:\n  features:\n  - name: patient_id\n    dtype: string\n  - name: sex\n    dtype: string\n  - name: age_at_mri\n    dtype: string\n  - name: who_cns_grade\n    dtype: string\n  - name: diagnosis_who2021\n    dtype: string\n  - name: idh\n    dtype: string\n  - name: mgmt_status\n    dtype: string\n  - name: brats21_id\n    dtype: string\n  - name: os_days\n    dtype: string\n  - name: dead\n    dtype: string\n  - name: num_slices\n    dtype: int32\n  - name: tumor_slice_idx\n    dtype: int32\n  - name: tumor_voxels\n    dtype: int64\n  - name: T1\n    dtype: image\n  - name: T1c\n    dtype: image\n  - name: T2\n    dtype: image\n  - name: FLAIR\n    dtype: image\n  - name: tumor_mask\n    dtype: image\n  - name: overlay_T1c\n    dtype: image\n  splits:\n  - name: test\n    num_bytes: 38063859\n    num_examples: 495\n  download_size: 38045105\n  dataset_size: 38063859\nconfigs:\n- config_name: default\n  data_files:\n  - split: test\n    path: data/test-*\n---\n\n# UCSF-PDGM — BraTS-format structural subset (T1/T1c/T2/FLAIR + tumor segmentation)\n\n> **⚠️ This repository is the *BraTS-format structural subset* of UCSF-PDGM — NOT the full multi-sequence release.**\n> Per patient it provides the **4 structural MRI modalities** (T1, T1c, T2, FLAIR) co-registered to the\n> **240 × 240 × 155 BraTS grid** (1 mm isotropic, skull-stripped) plus the **expert-corrected tumor segmentation**.\n> The full UCSF-PDGM release additionally contains DWI, ADC, ASL, SWI, HARDI/DTI and derived diffusion maps, N4\n> bias-corrected (`_bias`) images, and whole-brain / parenchyma masks — **all intentionally excluded here** to keep\n> a compact, BraTS-compatible glioma-segmentation dataset.\n\nThe University of California San Francisco Preoperative Diffuse Glioma MRI (UCSF-PDGM) dataset is a cohort of\npreoperative 3 T brain MRI from patients with histopathologically confirmed diffuse gliomas.\n\n## Contents\n\n- **495 patients** — one folder each, `UCSF-PDGM-XXXX/`, with **5 NIfTI volumes**.\n- `UCSF-PDGM-metadata_v5.csv` — official clinical metadata (sex, age, WHO CNS grade, final WHO-2021 diagnosis,\n  MGMT status/index, 1p/19q, IDH, overall survival, extent of resection, and the **BraTS21 ID / cohort columns**).\n- `UCSF-PDGM-metadata_glossary.csv` — glossary for the metadata columns.\n\n> Note: the metadata table has **501 rows** (imaging studies) while there are **495 image folders** (unique patients);\n> 6 follow-up/duplicate studies are not included as images.\n\n## Per-patient series layout (do not infer — this is the authoritative mapping)\n\nEvery `UCSF-PDGM-XXXX/` folder contains exactly these five files:\n\n| File                                   | Series | Description                          | Role        |\n|----------------------------------------|--------|--------------------------------------|-------------|\n| `UCSF-PDGM-XXXX_T1.nii.gz`             | T1     | Pre-contrast T1-weighted             | image       |\n| `UCSF-PDGM-XXXX_T1c.nii.gz`            | T1c    | Post-contrast (gadolinium) T1-w      | image       |\n| `UCSF-PDGM-XXXX_T2.nii.gz`             | T2     | T2-weighted                          | image       |\n| `UCSF-PDGM-XXXX_FLAIR.nii.gz`          | FLAIR  | T2-FLAIR                             | image       |\n| `UCSF-PDGM-XXXX_tumor_segmentation.nii.gz` | seg | Tumor segmentation                  | **ground truth mask** |\n\nAll five volumes share identical geometry — shape `(240, 240, 155)`, 1 mm isotropic — and are voxel-aligned.\n\n## Tumor segmentation labels (BraTS convention)\n\n| Label | Region                                            |\n|-------|---------------------------------------------------|\n| 0     | Background                                         |\n| 1     | Non-enhancing tumor core / necrosis (NCR)         |\n| 2     | Peritumoral edema (ED)                            |\n| 4     | Enhancing tumor (ET)                           ","encoding":"utf8","key":"README.md","name":"README.md","size":6167,"truncated":true}],"sourceObjectKey":"UCSF-PDGM-metadata_glossary.csv","starterCell":"import pandas as pd\ndf = pd.read_csv(\"hf://datasets/MedOtter/UCSF-PDGM/UCSF-PDGM-metadata_glossary.csv\")\ndf.head()","starterSamplePath":"UCSF-PDGM-0004/UCSF-PDGM-0004_FLAIR.nii.gz","totalObjects":2480},"summary":"The University of California San Francisco Preoperative Diffuse Glioma MRI (UCSF-PDGM) dataset, published on The Cancer Imaging Archive, is a cohort of preoperative 3 T brain MRI…","title":"UCSF-PDGM: preoperative diffuse glioma MRI with tumor segmentations (BraTS-format structural subset)","updatedAt":"2026-11-14T17:54:21.860Z","visibility":"public","websiteUrl":"https://www.cancerimagingarchive.net/collection/ucsf-pdgm/"}}