In 1204 CT images we segmented 104 anatomical structures (27 organs, 59 bones, 10 muscles, 8 vessels) covering a majority of relevant classes for most use cases. The CT images were randomly sampled from clinical routine, thus representing a real world dataset which generalizes to clinical application. The dataset contains a wide range of different pathologies, scanners, sequences and institutions. s0720/segmentations/portal_vein_and_splenic_vein.nii.gz 187.74kB s0720/segmentations/pancreas.nii.gz 45.25kB s0720/segmentations/lung_upper_lobe_right.nii.gz 218.92kB s0720/segmentations/lung_upper_lobe_left.nii.gz 230.82kB s0720/segmentations/lung_middle_lobe_right.nii.gz 201.18kB s0720/segmentations/lung_lower_lobe_right.nii.gz 240.63kB s0720/segmentations/lung_lower_lobe_left.nii.gz 239.49kB s0720/segmentations/liver.nii.gz 273.08kB s0720/segmentations/kidney_right.nii.gz 198.91kB s0720/segmentations/kidney_left.nii.gz 197.82kB s0720/segmentations/inferi

In 616 MR images we segmented 50 anatomical structures covering a majority of relevant classes for most use cases. The MR images were randomly sampled from clinical routine, thus representing a real world dataset which generalizes to clinical application. The dataset contains a wide range of different pathologies, scanners, sequences and institutions. Moreover, it contains some images from IDC for further data diversity (see column "source" in meta.csv).

Link to a copy of this dataset on Dropbox for much quicker download: Dropbox Link

You can find a segmentation model trained on this dataset here.

More details about the dataset can be found in the corresponding paper. Please cite this paper if you use the dataset. The CT images described in the paper can be found here.

This dataset contains all 50 structures from the TotalSegmentator "total" task. It does not contain the structures of other TotalSegmentator MRI subtasks.

This dataset was created by the department of Research and Analysis at University Hospital Basel.

UPDATE: on 2025-01-21 we uploaded version 2.0.0 which increases the number of images from 298 to 616. It also contains slightly different structures.

About

This is a derivative of the TotalSegmentator dataset.

1228 CT images and corresponding segmentation mask of 117 structures We combined multiple segmentation masks into a single nii.gz file under the folder Masks, and moved all CT images to the folder Images. All images and masks are renamed according to case IDs.

This dataset is released under the CC-BY-4.0 license.

  Official Release

GitHub (official): https://github.com/wasserth/TotalSegmentator (Apache-2.0… See the full description on the dataset page: https://huggingface.co/datasets/YongchengYAO/TotalSegmentator-CT-Lite.

This dataset contributes volumetric segmentations of the anatomic regions in a subset of CT images available from NCI Imaging Data Commons [1] (https://imaging.datacommons.cancer.gov/) automatically generated using the TotalSegmentation model v1.5.6 [2]. The initial release includes segmentations for the majority of the CT scans included in the National Lung Screening Trial (NLST) collection [3], [4] already available in IDC. Direct link to open this analysis result dataset in IDC (available after release of IDC v18): https://portal.imaging.datacommons.cancer.gov/explore/filters/?analysis_results_id=TotalSegmentator-CT-Segmentations.

Specifically, for each of the CT series analyzed, we include segmentations as generated by TotalSegmentator, converted into DICOM Segmentation object format using dcmqi v1.3.0 [5], and first order and shape features for each of the segmented regions, as produced by pyradiomics v3.0.1 [6]. Radiomics features were converted to DICOM Structured Reporting documents following template TID1500 using dcmqi. TotalSegmentator analysis on the NLST cohort was executed using Terra platform [7]. Implementation of the workflow that was used for performing the analysis is available at https://github.com/ImagingDataCommons/CloudSegmentator [8].

Due to the large size of the files, they are stored in the cloud buckets maintained by IDC, and the attached files are the manifests that can be used to download the actual files.

If you use the files referenced in the attached manifests, we ask you to cite this dataset and the preprint describing how it was generated [9].

Download instructions

Each of the manifests include instructions in the header on how to download the included files.

To download the TotalSegmentator segmentations (in DICOM SEG format) and pyradiomics measurements (in DICOM SR format) files using .s5cmd manifests:

1. install idc-index package: pip install --upgrade idc-index
2. download the files referenced by manifests included in this dataset by passing the .s5cmd manifest file. E.g., idc download totalsegmentator_ct_segmentations_aws.s5cmd

Other files included in the record are:

1. firstorder and shape radiomics features extracted using pyradiomics, and organized one file per segmented structure (see README file in the zip file for details on how those are organized)
   1. pyradiomics_features_csv.zip: saved in CSV format
   2. pyradiomics_features_parquet.zip: saved in Parquet format

Support

If you have any questions about this dataset, or if you experience any issues, please reach out to Imaging Data Commons support via support@canceridc.dev or (preferred) IDC Forum at https://discourse.canceridc.dev.

The TotalSegmentator-V2 dataset is a publicly available dataset for 3D medical image segmentation. It contains 1,228 CT scans with annotations for 117 major anatomical structures in WBCT images.

Dataset

This dataset was created by Elahi

Contents

nnU-Net weights for TotalSegmentator lung vessels model (Task 258). See https://github.com/wasserth/TotalSegmentator for more details.

stranger47/totalsegmentator-mesh-dataset dataset hosted on Hugging Face and contributed by the HF Datasets community

Dataset

This dataset was created by Elahi

Contents

Dataset Card for TotalSegmentator

This is a FiftyOne dataset with 50 samples.

  Installation

If you haven't already, install FiftyOne: pip install -U fiftyone

  Usage

import fiftyone as fo import fiftyone.utils.huggingface as fouh

Load the dataset

Note: other available arguments include 'max_samples', etc

dataset = fouh.load_from_hub("dgural/Total-Segmentator-50")

Launch the App

session = fo.launch_app(dataset)

  Dataset Details… See the full description on the dataset page: https://huggingface.co/datasets/dgural/Total-Segmentator-50.

nnU-Net weights for TotalSegmentator body segmentation (Task 273). See https://github.com/wasserth/TotalSegmentator for more details.

stranger47/totalsegmentator-mesh-dataset2-debug dataset hosted on Hugging Face and contributed by the HF Datasets community

nnU-Net weights for TotalSegmentator intracerebral hemorrhage model (Task 150). See https://github.com/wasserth/TotalSegmentator for more details.

A live porcine CBCT-Scan was segmentated using TotalSegmentator (3.00mm resolution)

there are some dataset error, please see discussion: https://www.kaggle.com/competitions/rsna-2023-abdominal-trauma-detection/discussion/436096

Apply total segmentator[1] on rsna 2023 abdominal trauma dataset[2]. The command used is based on public notebook[3]

!TotalSegmentator \
-i /kaggle/input/rsna-2023-abdominal-trauma-detection/train_images/10104/27573 \
-o /kaggle/temp/masks \
-ot 'nifti' \
-rs spleen kidney_left kidney_right liver esophagus colon duodenum small_bowel stomach

NOTE: there are probably error (about 5%?) in the total segmentator results. Please do a check before using this dataset!!!

[1] https://github.com/wasserth/TotalSegmentator

[2] https://www.kaggle.com/competitions/rsna-2023-abdominal-trauma-detection

[3] https://www.kaggle.com/code/enriquezaf/totalsegmentator-offline

Whole-Spine Anatomical MRI dataset & B0 simulations

Dataset Description

This dataset includes structural MRI (T1-weighted) and simulated ΔB0 field maps for sixty volunteers. Participants were scanned using two Siemens 3T MRI scanners (MAGNETOM Tim Trio and Verio) equipped with head, neck, and spine coils. The scans cover anatomical regions extending from the head to the torso and include lateral torso encompassing most of both lungs.

All data is organized in BIDS format and is available on OpenNeuro.

Participants

• Total Participants: 60
• Males: 32
• Females: 18
• Undisclosed sex: 10
• Age: Mean = 27.1 years, SD = 6.5, Range = 21-56 years
• Weight: Mean = 66.7 kg, SD = 9.5, Range = 45-90 kg
• Height: Mean = 175.6 cm, SD = 8.8, Range = 155-192 cm

MRI Acquisition

• Scanner Models: Siemens MAGNETOM Tim Trio and Verio (3T)
• Coils Used: Head, neck, and spine coils
• Structural Images: T1-weighted MPRAGE
• Resolution: 1 mm³
• Field of View (FOV): From head to torso, including lateral regions of both lungs
• Data Processing

Structural Data Segmentation

1. Automated Segmentation Tools:

   • TotalSegmentator MRI: Used for full-body, sinuses, trachea, ear canal, and lungs based on training with 10 manually segmented subjects.
   • Samseg: Used for segmenting brain, eyes, and skull.
   • TotalSpineSeg: Used for segmenting spinal cord, vertebrae, and intervertebral disks.

2. Post-Processing Steps:

   • Tissue islands were removed, holes were closed, and tissue masks for specific regions (skull, brain, eyes, sinus, and ear canal) were smoothed using a custom pipeline (GitHub repo, release v1.1, commit: 4f3c471db542fa9b12f308aaeece401323980965).
   • Tissue masks were then merged into a single NIfTI file with the following voxel assignments: background (air), body, brain, spine, lungs, skull, trachea, sinus, ear canal, and eyes.

Susceptibility Assignment

Each anatomical label in the segmentation volumes was assigned a specific susceptibility value (χ) as defined in this Github repository:

• Air: 0.35 ppm
• Sinus & Ear Canals: -2 ppm
• Trachea & Lungs: -4.2 ppm
• Brain: -9.04 ppm
• Body & Eyes: -9.05 ppm
• Spinal Canal & Disks: -9.055 ppm
• Skull & Vertebrae: -11 ppm

Field Map Simulation

Field maps (ΔB0) were generated by applying a convolution in the Fourier domain between the susceptibility maps and an analytical dipole distribution. Key parameters:

• Implementation: Python (GitHub repo)
• Padding:
  • Edge-value padding applied on five volume surfaces
  • Constant-value padding applied on the dorsal surface
  • Padding Size: 50 voxels per surface

Dataset Files and Structure

This dataset is organized according to the BIDS format. Key directories and files include:

• /sub-
• /derivatives: Includes simulated ΔB0 field maps and segmentation labels

Dataset Card for VertebralBodiesCT-Labels

This dataset contains labels for the thoracic and lumbar vertebral bodies from 1460 CT scans, designed for deep learning applications in anatomical landmark identification.

  Dataset Details

VertebralBodiesCT-Labels is a dataset including segmentation labels for the vertebral bodies of the thoracic and lumbar spine, and the sacrum. Derived from 1460 CT scans originally published in the TotalSegmentator and VerSe datasets, labels… See the full description on the dataset page: https://huggingface.co/datasets/fhofmann/VertebralBodiesCT-Labels.

LUNA25 TS Segmentations

The LUNA25 challenge serves as a benchmark for evaluating lung nodule detection in low-dose CT scans. This repository includes segmentations generated using TotalSegmentator for the total, lung_vessels, and lung_nodules tasks. Note: Segmentation volumes have not been independently verified and are supplied "as is".

  Steps to recreate

Use scripts/convert_nifti.py and scripts/convert_nifti.sh to convert LUNA25 .mha files to .nii.gz. Install… See the full description on the dataset page: https://huggingface.co/datasets/farrell236/LUNA25_ts_seg.