{"success":true,"data":{"author":"Christian Mayer, Antonio Pepe, Jan Egger, Heinrich Mächler et al. (Medical University of Graz); mirrored by MedOtter","bucketAccessType":"public","bucketName":"MedOtter/Aortic-Dissection-Graz","bucketPrefix":"","category":"Health & Biomedical","cloudProvider":"huggingface","createdAt":"2026-11-17T19:29:33.302Z","description":"40 computed tomography angiography (CTA) volumes of type B aortic dissection collected in clinical routine between 2005 and 2021 across the Medical University of Graz regional hospital network, each with expert segmentations of the true lumen and the false lumen. Scans are arterial phase and not ECG-gated, mostly 512x512xZ with in-plane spacing 0.525 to 0.965 mm and slice spacing 1.25 to 3.0 mm; head and face were cropped for anonymization. Thrombosed sections, calcifications and atherosclerotic changes were not segmented. The repository holds the image volumes and masks as NRRD with surface meshes (STL) under dataset/, the untouched source masks under original_masks/, and train.jsonl with per-case shape, spacing and true and false lumen volumes in mL; about 3.8 GB (measured 2026-11-17). Use for lumen segmentation and false-lumen volumetry. Read NRRD with SimpleITK or pynrrd and meshes with trimesh. Mirrored on Hugging Face by MedOtter from the authors' release, with a card documenting provenance and any normalization. The original figshare record is CC BY 4.0; cite Mayer et al., Type B Aortic Dissection CTA Collection with True and False Lumen Expert Annotations, Scientific Data 11, 596 (2024).","displayTitle":"Type B aortic dissection CTA (Graz): 40 volumes with expert true-lumen and false-lumen annotations","doi":"10.6084/m9.figshare.22269091","huggingFaceUrl":"https://huggingface.co/datasets/MedOtter/Aortic-Dissection-Graz","id":"bb06af73-ea6e-52ee-9fa2-227911784ec5","license":"CC-BY-4.0","licenseEvidenceUrl":"https://api.figshare.com/v2/articles/22269091","objectCountEstimate":284,"probeStatus":"ok","referenceUrl":"https://doi.org/10.1038/s41597-024-03284-2","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"MedOtter/Aortic-Dissection-Graz\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"MedOtter/Aortic-Dissection-Graz\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=MedOtter, license:cc-by-4.0, size_categories:n<1K, format:parquet, downloads=1117"],"format":"mixed","generatedAt":"2026-11-17T19:40:24.971Z","layout":{"example":"preview/train-00000-of-00001.parquet","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"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","encoding":"hex","key":"preview/train-00000-of-00001.parquet","truncatedAt":512}],"sampleObjects":["dataset/cta01s/cta01s.nrrd","dataset/cta01s/falselumen01.seg.nrrd","dataset/cta01s/mask01.nrrd","dataset/cta01s/mesh01.stl","dataset/cta01s/truelumen01.seg.nrrd"],"sidecars":[{"bytes":"---\nlicense: cc-by-4.0\ntask_categories:\n- image-segmentation\ntags:\n- medical\n- cta\n- ct\n- aorta\n- aortic-dissection\n- vascular\n- 3d\npretty_name: Aortic Dissection (Graz) - true/false lumen CTA\nsize_categories:\n- n<1K\n---\n\n# Aortic Dissection (Graz) - CTA with true / false lumen expert annotations\n\nRe-host of **\"Aortic Dissection Dataset and Segmentations\"**\n([figshare 10.6084/m9.figshare.22269091](https://doi.org/10.6084/m9.figshare.22269091),\nCC BY 4.0) - 40 **type-B aortic dissection** CTA volumes collected in clinical\nroutine 2005-2021 across the Medical University of Graz regional hospital\nnetwork, with expert **true-lumen** and **false-lumen** annotations.\n\n> Mayer C, Pepe A, Hossain S, Karner B, Arnreiter M, Kleesiek J, Schmid J,\n> Janisch M, Fuchsjaeger M, Deutschmann H, Zimpfer D, Egger J, Maechler H.\n> *Type B Aortic Dissection CTA Collection with True and False Lumen Expert\n> Annotations for the Development of AI-based Algorithms.*\n> **Scientific Data 11, 596 (2024).** doi:10.1038/s41597-024-03284-2\n\nEthics: Medical University of Graz EK-34-161 ex 21/22; consent waived\n(retrospective), head and face cropped for anonymisation.\n\n## Contents\n\n| | |\n|---|---|\n| Modality | CTA, arterial phase (**not** ECG-gated) |\n| Cases | 40, one volume each |\n| Classes | `0` background, `1` true lumen, `2` false lumen |\n| Split | **none** - single cohort of 40 (`split=\"train\"` for all rows) |\n| Shape | mostly 512x512xZ; in-plane spacing 0.525-0.965 mm, through-plane 1.25-3.0 mm |\n| Foreground | TL 125,516-659,055 voxels; FL 25,242-496,428 voxels; **no empty masks** |\n\n**There is no thrombus class.** Per the paper, \"clearly thrombosed sections have\nnot been segmented\"; calcifications and atherosclerotic changes are likewise\nexcluded. Ambiguous dark regions judged to be late-filling artifact were\nsegmented *as lumen*. If you need a thrombus label, see ImageTBAD instead.\n\n## Layout\n\n```\ndataset/ctaNNs/ctaNNs.nrrd           image, int16 HU - byte-identical to figshare\n               truelumenNN.seg.nrrd  true lumen,  uint8 {0,1}\n               falselumenNN.seg.nrrd false lumen, uint8 {0,1}\n               maskNN.nrrd           merged {0=bg, 1=TL, 2=FL}\n               meshNN.stl            surface mesh - byte-identical to figshare\noriginal_masks/ctaNNs/*.seg.nrrd     untouched source masks (~13 MB total)\ntrain.jsonl                          40 rows (schema below)\n```\n\n`train.jsonl` per row: `case_id`, `case_num`, `image`, `mask`,\n`mask_true_lumen`, `mask_false_lumen`, `mesh`, `original_mask_*`, `shape`,\n`spacing_mm`, `coarse_axis`, `tl_voxels`, `fl_voxels`, `tl_volume_ml`,\n`fl_volume_ml`, `tl_fl_overlap_voxels`, `fl_label_value_raw`,\n`resampled_to_image_grid`, `annotator_tier`, `split`.\n\n## What was normalised here (and why)\n\nEverything below is flagged per-case in `train.jsonl`, and the untouched source\nmasks ship under `original_masks/`.\n\n1. **False-lumen masks binarised to `{0,1}`.** 17 of 40 cases store the false\n   lumen with **label value 2**, the other 23 with 1 - cases\n   `4, 5, 11, 12, 13, 14, 15, 16, 17, 18, 22, 25, 26, 30, 35, 36, 38`. Code\n   doing `mask == 1` against the raw files silently drops those 17 false\n   lumens. Original value kept in `fl_label_value_raw`.\n2. **Case 40 masks resampled onto the image grid.** As shipped, case 40's masks\n   sit on a grid rotated ~6.85 deg about Z (512x1103x187 @0.545 mm) while its\n   image is axis-aligned (512x512x401 @0.602 mm) - index-wise pairing raises\n   `IndexError`. Nearest-neighbour world-space resampling retains **100.01% /\n   100.04%** of physical volume, and mean HU inside the resampled masks is\n   **409 HU (TL) / 339 HU (FL)**, i.e. contrast-filled arterial lumen and in\n   line with the other 39 cases. Flagged by `resampled_to_image_grid`.\n   Every other case's masks already match their image exactly.\n3. **Merged `maskNN.nrrd` added**, painted FL first then TL, so the **true\n   lumen wins ties** - matching the authors' own converter\n   [`apepe91/AD_NRRD_TO_STL`](https://github.com/apepe91/AD_NRRD_TO_STL) (cited\n   in the paper's Code Availability), which subtracts a dilat","encoding":"utf8","key":"README.md","name":"README.md","size":7853,"truncated":true}],"sourceObjectKey":"preview/train-00000-of-00001.parquet","starterCell":"from datasets import load_dataset\nds = load_dataset(\"MedOtter/Aortic-Dissection-Graz\")\nprint(ds)","starterSamplePath":"dataset/cta01s/cta01s.nrrd","totalObjects":284},"summary":"40 computed tomography angiography (CTA) volumes of type B aortic dissection collected in clinical routine between 2005 and 2021 across the Medical University of Graz regional…","title":"Type B aortic dissection CTA (Graz): 40 volumes with expert true-lumen and false-lumen annotations","updatedAt":"2026-11-17T23:06:34.199Z","visibility":"public","websiteUrl":"https://huggingface.co/datasets/MedOtter/Aortic-Dissection-Graz"}}