FUMPE: CT pulmonary angiography (CTPA) of 35 patients with pulmonary embolism masks (Ferdowsi University of Mashhad)

Provider
Masoudi et al. (Ferdowsi University of Mashhad); mirrored by MedOtter
License
CC-BY-4.0 (source)
Size
72 files
Format
NIfTI (in sampled files)
Stored at
Hugging Face (https://huggingface.co/datasets/MedOtter/FUMPE)
Access
Free and public. No account with the data provider is needed.
DOI
10.6084/m9.figshare.c.4107803.v1
Category
Health & Biomedical
Source
https://doi.org/10.1038/sdata.2018.180

About this dataset

FUMPE (Ferdowsi University of Mashhad's PE dataset) is a public benchmark for computer-aided detection of pulmonary embolism, blood clots blocking the lung arteries, in CT pulmonary angiography (CTA/CTPA). It holds three-dimensional contrast CT scans of 35 subjects, 8,792 slices in total, with ground-truth clot regions drawn by two expert radiologists using a semi-automated tool. There are 3,438 embolus regions, about two thirds of them in small peripheral arteries, which makes detection hard. The publisher also reports the Qanadli obstruction score per scan for mortality and morbidity risk analysis. The mirror stores the scans as NIfTI: images/PAT001.nii.gz to PAT035.nii.gz and matching binary masks under masks/, about 4 GB (measured 2026-11-21); read with nibabel or SimpleITK. Use for PE segmentation and detection, and for testing chest CT angiography models on peripheral emboli. Original data on figshare (collection DOI 10.6084/m9.figshare.c.4107803), released under CC BY 4.0. Cite Masoudi et al., Scientific Data 5:180180 (2018), doi 10.1038/sdata.2018.180.

How to load this dataset in Python

from huggingface_hub import hf_hub_download
path = hf_hub_download(repo_id="MedOtter/FUMPE", filename="images/PAT001.nii.gz", repo_type="dataset")
print(path)

Data structure

File types in sampled files: NIfTI

Files

Example file: images/PAT001.nii.gz

Analyze this dataset without downloading it

On Scigantic, FUMPE: CT pulmonary angiography (CTPA) of 35 patients with pulmonary embolism masks (Ferdowsi University of Mashhad) opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.

This page is also available as Markdown and as JSON.

Related Health & Biomedical datasets