Hugging Face's logo

Datasets:
ppcnet-dataset
/
PPCNet

Tasks:
Image-to-3D
Other
Modalities:
3D
Image
Text
Formats:
text
Size:
1K - 10K
Tags:
medical-imaging
spine
point-cloud
lumbar
DRR
biplanar-radiographs
Libraries:
Datasets
License:
Dataset card Data Studio Files
xet
 Community
1
Dataset Viewer
Auto-converted to Parquet Duplicate
text
stringlengths
59
62
-2.04800000e+00 -3.41000000e-01 0.00000000e+00 3.16029255e+02
0.00000000e+00 -3.41000000e-01 -2.04800000e+00 -1.56550423e+03
0.00000000e+00 -6.66666667e-04 0.00000000e+00 5.52919403e-01
3.41000000e-01 -2.04800000e+00 0.00000000e+00 -1.39613518e+01
3.41000000e-01 0.00000000e+00 -2.04800000e+00 -1.51285226e+03
6.66666667e-04 0.00000000e+00 0.00000000e+00 6.55855801e-01
-2.04800000e+00 -3.41000000e-01 0.00000000e+00 4.40433500e+02
0.00000000e+00 -3.41000000e-01 -2.04800000e+00 1.34871906e+02
0.00000000e+00 -6.66666667e-04 0.00000000e+00 7.97000000e-01
3.41000000e-01 -2.04800000e+00 0.00000000e+00 7.35928000e+02
3.41000000e-01 0.00000000e+00 -2.04800000e+00 6.27504060e+01
6.66666667e-04 0.00000000e+00 0.00000000e+00 6.56000000e-01
-2.04800000e+00 -3.41000000e-01 0.00000000e+00 3.90293519e+02
0.00000000e+00 -3.41000000e-01 -2.04800000e+00 -4.35586956e+01
0.00000000e+00 -6.66666667e-04 0.00000000e+00 7.61693471e-01
3.41000000e-01 -2.04800000e+00 0.00000000e+00 6.32807904e+02
3.41000000e-01 0.00000000e+00 -2.04800000e+00 -9.22793458e+01
6.66666667e-04 0.00000000e+00 0.00000000e+00 6.66442933e-01
-2.04800000e+00 -3.41000000e-01 0.00000000e+00 3.98458127e+02
0.00000000e+00 -3.41000000e-01 -2.04800000e+00 1.74070654e+02
0.00000000e+00 -6.66666667e-04 0.00000000e+00 6.88062068e-01
3.41000000e-01 -2.04800000e+00 0.00000000e+00 3.98981846e+02
3.41000000e-01 0.00000000e+00 -2.04800000e+00 1.55382080e+02
6.66666667e-04 0.00000000e+00 0.00000000e+00 6.51525267e-01
-2.04800000e+00 -3.41000000e-01 0.00000000e+00 3.63068110e+02
0.00000000e+00 -3.41000000e-01 -2.04800000e+00 1.38228024e+02
0.00000000e+00 -6.66666667e-04 0.00000000e+00 7.85744731e-01
3.41000000e-01 -2.04800000e+00 0.00000000e+00 7.13274878e+02
3.41000000e-01 0.00000000e+00 -2.04800000e+00 8.37866590e+01
6.66666667e-04 0.00000000e+00 0.00000000e+00 6.79310000e-01
-2.04800000e+00 -3.41000000e-01 0.00000000e+00 4.10925891e+02
0.00000000e+00 -3.41000000e-01 -2.04800000e+00 2.43069758e+02
0.00000000e+00 -6.66666667e-04 0.00000000e+00 7.58754201e-01
3.41000000e-01 -2.04800000e+00 0.00000000e+00 6.20092768e+02
3.41000000e-01 0.00000000e+00 -2.04800000e+00 1.92166846e+02
6.66666667e-04 0.00000000e+00 0.00000000e+00 6.59237267e-01
-2.04800000e+00 -3.41000000e-01 0.00000000e+00 4.04745076e+02
0.00000000e+00 -3.41000000e-01 -2.04800000e+00 1.78487151e+02
0.00000000e+00 -6.66666667e-04 0.00000000e+00 7.82000600e-01
3.41000000e-01 -2.04800000e+00 0.00000000e+00 6.94514656e+02
3.41000000e-01 0.00000000e+00 -2.04800000e+00 1.18702656e+02
6.66666667e-04 0.00000000e+00 0.00000000e+00 6.65119867e-01
-2.04800000e+00 -3.41000000e-01 0.00000000e+00 3.49848561e+02
0.00000000e+00 -3.41000000e-01 -2.04800000e+00 -1.10652003e+03
0.00000000e+00 -6.66666667e-04 0.00000000e+00 4.78987732e-01
3.41000000e-01 -2.04800000e+00 0.00000000e+00 -2.53007012e+02
3.41000000e-01 0.00000000e+00 -2.04800000e+00 -1.02797957e+03
6.66666667e-04 0.00000000e+00 0.00000000e+00 6.32537000e-01
-2.04800000e+00 -3.41000000e-01 0.00000000e+00 3.80094500e+02
0.00000000e+00 -3.41000000e-01 -2.04800000e+00 -1.91983281e+01
0.00000000e+00 -6.66666667e-04 0.00000000e+00 4.76573466e-01
3.41000000e-01 -2.04800000e+00 0.00000000e+00 -2.65665319e+02
3.41000000e-01 0.00000000e+00 -2.04800000e+00 5.53353373e+01
6.66666667e-04 0.00000000e+00 0.00000000e+00 6.22289332e-01
-2.04800000e+00 -3.41000000e-01 0.00000000e+00 3.75487937e+02
0.00000000e+00 -3.41000000e-01 -2.04800000e+00 1.15861342e+02
0.00000000e+00 -6.66666667e-04 0.00000000e+00 7.04364733e-01
3.41000000e-01 -2.04800000e+00 0.00000000e+00 4.54276707e+02
3.41000000e-01 0.00000000e+00 -2.04800000e+00 9.40470265e+01
6.66666667e-04 0.00000000e+00 0.00000000e+00 6.61717000e-01
-2.04800000e+00 -3.41000000e-01 0.00000000e+00 3.94079267e+02
0.00000000e+00 -3.41000000e-01 -2.04800000e+00 7.79497947e+01
0.00000000e+00 -6.66666667e-04 0.00000000e+00 7.78034668e-01
3.41000000e-01 -2.04800000e+00 0.00000000e+00 6.83769445e+02
3.41000000e-01 0.00000000e+00 -2.04800000e+00 2.16320074e+01
6.66666667e-04 0.00000000e+00 0.00000000e+00 6.67931467e-01
-2.04800000e+00 -3.41000000e-01 0.00000000e+00 -3.99705634e+00
0.00000000e+00 -3.41000000e-01 -2.04800000e+00 -4.72567150e+02
0.00000000e+00 -6.66666667e-04 0.00000000e+00 6.24066666e-01
3.41000000e-01 -2.04800000e+00 0.00000000e+00 2.63948051e+02
3.41000000e-01 0.00000000e+00 -2.04800000e+00 -3.96961996e+02
6.66666667e-04 0.00000000e+00 0.00000000e+00 7.71877330e-01
-2.04800000e+00 -3.41000000e-01 0.00000000e+00 4.09407943e+02
0.00000000e+00 -3.41000000e-01 -2.04800000e+00 -1.98193042e+03
0.00000000e+00 -6.66666667e-04 0.00000000e+00 5.35863800e-01
3.41000000e-01 -2.04800000e+00 0.00000000e+00 -8.33566504e+01
3.41000000e-01 0.00000000e+00 -2.04800000e+00 -1.93755499e+03
6.66666667e-04 0.00000000e+00 0.00000000e+00 6.22619268e-01
-2.04800000e+00 -3.41000000e-01 0.00000000e+00 2.34217038e+02
0.00000000e+00 -3.41000000e-01 -2.04800000e+00 6.29141070e+02
0.00000000e+00 -6.66666667e-04 0.00000000e+00 5.23652934e-01
3.41000000e-01 -2.04800000e+00 0.00000000e+00 -9.27384224e+01
3.41000000e-01 0.00000000e+00 -2.04800000e+00 7.07892360e+02
6.66666667e-04 0.00000000e+00 0.00000000e+00 6.77614400e-01
-2.04800000e+00 -3.41000000e-01 0.00000000e+00 4.69503890e+02
0.00000000e+00 -3.41000000e-01 -2.04800000e+00 1.35946124e+02
0.00000000e+00 -6.66666667e-04 0.00000000e+00 6.29515333e-01
3.41000000e-01 -2.04800000e+00 0.00000000e+00 2.02310644e+02
3.41000000e-01 0.00000000e+00 -2.04800000e+00 1.30388573e+02
6.66666667e-04 0.00000000e+00 0.00000000e+00 6.18650131e-01
-2.04800000e+00 -3.41000000e-01 0.00000000e+00 4.08586906e+02
0.00000000e+00 -3.41000000e-01 -2.04800000e+00 -2.94405503e+02
0.00000000e+00 -6.66666667e-04 0.00000000e+00 8.08578929e-01
3.41000000e-01 -2.04800000e+00 0.00000000e+00 7.77787191e+02
3.41000000e-01 0.00000000e+00 -2.04800000e+00 -3.66160903e+02
6.66666667e-04 0.00000000e+00 0.00000000e+00 6.68294667e-01
-2.04800000e+00 -3.41000000e-01 0.00000000e+00 3.64626866e+02
0.00000000e+00 -3.41000000e-01 -2.04800000e+00 -9.11008625e+01
0.00000000e+00 -6.66666667e-04 0.00000000e+00 6.85213000e-01
3.41000000e-01 -2.04800000e+00 0.00000000e+00 3.95619902e+02
End of preview. Expand in Data Studio

PPCNet Dataset

Biplanar DRRs + Projection Matrices + Ground-Truth Point Clouds

License Patients Points Size

A curated lumbar spine dataset for 3D point cloud reconstruction from biplanar radiographs.

Overview

This dataset provides paired biplanar DRRs, calibrated 3×4 projection matrices, and dense ground-truth point clouds for 1,037 patients with complete L1–L5 lumbar vertebrae, derived from the publicly available VerSe'19, VerSe'20, and CTSpine1K collections.

📁 Dataset Structure

Each of the 1,037 patient folders contains:

File Description
ct.nii.gz CT volume (LPS orientation)
seg.nii.gz Segmentation labels (L1=20, L2=21, L3=22, L4=23, L5=24)
gt_ppc.vtk Ground-truth point cloud (5,120 points, world-mm)
AP_0/drr_AP_0.png Anteroposterior DRR (512×512)
AP_0/P_AP_0.txt 3×4 AP projection matrix
LP_90/drr_LP_90.png Lateral DRR (512×512)
LP_90/P_LP_90.txt 3×4 lateral projection matrix

Label mapping: L1=20, L2=21, L3=22, L4=23, L5=24

Data Split

Fixed split in dataset_split.json (seed=42):

Split Patients Usage
Train 829 Model training
Validation 103 Hyperparameter tuning
Test 105 Final evaluation
Total 1,037

DRR Generation

DRRs are generated using Plastimatch's ray-casting algorithm with the following parameters:

Parameter Value
Source-to-Axis Distance (SAD) 1,000 mm
Source-to-Imager Distance (SID) 1,500 mm
Detector Size 500 × 500 mm
Image Resolution 512 × 512 px
Views AP (0°) + Lateral (90°)
Bone Enhancement 2.5× for HU > 300
Post-processing CLAHE normalisation

Each DRR comes with a calibrated 3×4 perspective projection matrix encoding the exact source–detector geometry.

Ground-Truth Generation

The ground-truth point clouds are constructed through a five-step pipeline:

  1. Surface extraction — marching cubes on per-vertebra segmentation masks
  2. Coordinate transform — vertices mapped to physical (mm) space via NIfTI affine
  3. Concatenation — all five vertebrae (L1–L5) merged into a single cloud
  4. Alignment — rigid ICP with four flip initialisations to match DRR projection space
  5. Sampling — farthest-point sampling to a uniform 5,120-point representation

⬇️ Download 

from huggingface_hub import snapshot_download

# Download full dataset (~69.2 GB)
snapshot_download(
    repo_id="ppcnet-dataset/PPCNet",
    repo_type="dataset",
    local_dir="./PPCNet_Dataset"
)

Or use the Hugging Face CLI:

pip install huggingface_hub
huggingface-cli download ppcnet-dataset/PPCNet --repo-type dataset --local-dir ./PPCNet_Dataset

Source Datasets

This dataset builds upon:

  • 📦 VerSe — A Vertebrae Labelling and Segmentation Benchmark for Multi-Detector CT Images
  • 📦 CTSpine1K — A Large-Scale Dataset for Spinal Vertebrae Segmentation in CT

License

This dataset is released under the MIT License.
Downloads last month
18,280
Number of rows:
6,222
Total file size:
69.2 GB