YongchengYAO/DEEP-PSMA-Lite

About

This is a preprocessed redistribution of DEEP-PSMA (Zenodo), which is released under the CC BY-NC 4.0 license.

Dataset summary: 100 cases with paired PSMA and FDG PET scans and total-tumour-burden (TTB) masks for each tracer.

Contents of this repository:

• Images-PSMA/ — 100 files
• Masks-PSMA/ — 100 files
• Images-FDG/ — 100 files
• Masks-FDG/ — 100 files

📝 Landmark annotations, visualization figures and the benchmark plan files live in 🔥MedVision🔥, where you can load the complete images and annotations from dataset configs.

Relation to the source dataset

In the source	100 cases, each with PET + CT + totseg_24 for BOTH the PSMA and FDG tracers
Excluded here	the CT and totseg_24 volumes for both tracers
In this repo	100 Images-PSMA + 100 Masks-PSMA + 100 Images-FDG + 100 Masks-FDG

All 100 cases and both tracers are included, but the CT and totseg_24 volumes are not redistributed. The total-tumour-burden annotation is defined by SUV thresholding on the PET and is delivered on the PET grid (e.g. 192x192x335 at 2.87 x 2.87 x 3.27 mm). The CT of a PET/CT is acquired at roughly 1 mm for attenuation correction, so using it as the image would require resampling the mask onto a ~3x finer grid — inventing lesion boundary detail that was never annotated and changing the physical measurements.

Why -Lite? The suffix marks this as a derived redistribution rather than a copy of the source. These are preprocessed volumes — every case has been format-converted, geometry-normalised and reoriented to RAS+ — and for some sources cases or modalities are excluded as well (see the table above). Use it to reproduce MedVision, not as a substitute for the original release. See Preprocessing below for exactly what was changed.

Preprocessing

• PET (SUV) volumes and their TTB masks converted to nii.gz and standardized to RAS+ orientation.

• The two tracers are kept in separate image/mask folders so that the subject-level train/test split cannot place the same patient's PSMA and FDG scans on opposite sides.

Segmentation Labels

labels_map = {
    "1": "total tumor burden"
}

Landmarks

landmarks_map = {
    "P1": "most right/anterior/superior endpoint of the major axis",
    "P2": "most left/superior/inferior endpoint of the major axis",
    "P3": "most right/anterior/superior endpoint of the minor axis",
    "P4": "most left/superior/inferior endpoint of the minor axis"
}

News

• [25 Jul, 2026] Initial release. This dataset is integrated into 🔥MedVision🔥, where you can use these config names to load data in python:

  • DEEP-PSMA_BoxSize_Task01_Axial_Test
  • DEEP-PSMA_BoxSize_Task01_Axial_Train
  • DEEP-PSMA_BoxSize_Task02_Axial_Test
  • DEEP-PSMA_BoxSize_Task02_Axial_Train
  • DEEP-PSMA_MaskSize_Task01_Axial_Test
  • DEEP-PSMA_MaskSize_Task01_Axial_Train
  • DEEP-PSMA_MaskSize_Task02_Axial_Test
  • DEEP-PSMA_MaskSize_Task02_Axial_Train
  • DEEP-PSMA_TumorLesionSize_Task01_Axial_Test
  • DEEP-PSMA_TumorLesionSize_Task01_Axial_Train
  • DEEP-PSMA_TumorLesionSize_Task02_Axial_Test
  • DEEP-PSMA_TumorLesionSize_Task02_Axial_Train

Data Usage Agreement

By using the dataset, you agree to the terms as follow.

• You must comply with the original CC BY-NC 4.0 license terms of the source dataset.
• You are recommended to refer to the source of this dataset in any publication: https://huggingface.co/datasets/YongchengYAO/DEEP-PSMA-Lite
• You must cite the original publication(s):
  • https://doi.org/10.5281/zenodo.15281784

Official Release

For more information, please go to the official site: https://deep-psma.grand-challenge.org/

Download from Huggingface

# python
from huggingface_hub import snapshot_download
snapshot_download(repo_id="YongchengYAO/DEEP-PSMA-Lite", repo_type='dataset', local_dir="/your/local/folder")