CT-ORG: multi-organ segmentation in whole-body CT (liver, lungs, kidneys, bladder, bone, brain)

Provider
Blaine Rister, Darvin Yi, Kaushik Shivakumar, Tomomi Nobashi, Daniel Rubin (Stanford); converted by MedOtter
License
CC-BY-3.0 (source)
Size
283 files
Format
NIfTI, JSON Lines (in sampled files)
Stored at
Hugging Face (https://huggingface.co/datasets/MedOtter/ct-org)
Access
Free and public. No account with the data provider is needed.
DOI
10.7937/tcia.2019.tt7f4v7o
Category
Health & Biomedical
Source
https://doi.org/10.1038/s41597-020-00715-8

About this dataset

CT-ORG is a diverse multi-organ segmentation dataset of 140 computed tomography scans published by Stanford researchers in Scientific Data (2020) and distributed by The Cancer Imaging Archive. Scans come from several clinical sources and include contrast and non-contrast, abdominal, chest and whole-body studies. Each volume has voxel labels for six organ classes: liver, bladder, lungs, kidneys, bone and brain. This MedOtter repository holds the volumes and label maps as NIfTI plus a JSONL index (image path, mask path, labels, modality, official split, patient ID). As measured 2026-11-14 there are 140 files under volumes/ (about 18.7 GB) and 140 under labels/ (about 0.4 GB); open them with nibabel, SimpleITK or MONAI, and read train.jsonl with datasets or pandas. Used for multi-organ CT segmentation, organ localisation for radiotherapy and nuclear medicine dosimetry, and as a task in medical segmentation foundation model benchmarks. TCIA releases CT-ORG under CC BY 3.0; cite Rister et al. 2020.

How to load this dataset in Python

from datasets import load_dataset
ds = load_dataset("MedOtter/ct-org")
print(ds)

Data structure

File types in sampled files: NIfTI, JSON Lines

Files

Example file: train.jsonl

Analyze this dataset without downloading it

On Scigantic, CT-ORG: multi-organ segmentation in whole-body CT (liver, lungs, kidneys, bladder, bone, brain) 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