wq2012/knee_3d_mri_segmentation_SKI10

Knee Bone & Cartilage Segmentation Models

This repository contains trained models for 3D MRI Knee Bone and Cartilage Segmentation on the SKI10 dataset, using the kneeseg Python package.

• GitHub: https://github.com/wq2012/kneeseg
• PyPI: https://pypi.org/project/kneeseg/

Model Details

• Architecture: Dense Random Forest with Auto-Context (2 Passes for Bones, 2 Passes for Cartilage).
• Dataset:
  • Training: The models are trained on all 100 cases from the SKI10/TrainingData folder.
  • Evaluation: The evaluations are performed on a subset of 20 cases of SKI10/TrainingData. Note that in this setup, the 20 evaluation cases are included in the training set, so there is train-eval overlap. For a non-overlap eval, see GitHub for a 80%-20% split experiment.
  • Inference: We have included the predicted labels for the 50 official SKI10 TestData cases in the TestData_labels/ directory. These were generated using the fully trained models but were not evaluated due to the lack of ground truth.
• Performance (evaluated on a subset of 20 cases):
  • Femur Bone: 0.9155 DSC (+/- 0.0303)
  • Tibia Bone: 0.9383 DSC (+/- 0.0194)
  • Femoral Cartilage: 0.7095 DSC (+/- 0.0478)
  • Tibial Cartilage: 0.6799 DSC (+/- 0.0429)

Usage

These models are designed to be used with the kneeseg library.

1. Install Library

pip install kneeseg

2. Load Models

from kneeseg.bone_rf import BoneClassifier
from kneeseg.rf_seg import CartilageClassifier

# Load Bone Model (Pass 1)
bone_p1 = BoneClassifier()
bone_p1.load("bone_rf_p1.joblib")

# ... (See library documentation for full pipeline)

Files

• bone_rf_p1.joblib: Bone Segmentation Pass 1 (Voxel Classifier)
• bone_rf_p2.joblib: Bone Segmentation Pass 2 (Auto-Context Refinement)
• cartilage_rf_p1.joblib: Cartilage Segmentation Pass 1 (Semantic Context Forest)
• cartilage_rf_p2.joblib: Cartilage Segmentation Pass 2 (Auto-Context Refinement)
• TestData_labels: Predicted labels of the SKI10 official TestData using our models.

Citation

Plain Text:

Quan Wang, Dijia Wu, Le Lu, Meizhu Liu, Kim L. Boyer, and Shaohua Kevin Zhou. "Semantic Context Forests for Learning-Based Knee Cartilage Segmentation in 3D MR Images." MICCAI 2013: Workshop on Medical Computer Vision.

Quan Wang. Exploiting Geometric and Spatial Constraints for Vision and Lighting Applications. Ph.D. dissertation, Rensselaer Polytechnic Institute, 2014.

BibTeX:

@inproceedings{wang2013semantic,
  title={Semantic context forests for learning-based knee cartilage segmentation in 3D MR images},
  author={Wang, Quan and Wu, Dijia and Lu, Le and Liu, Meizhu and Boyer, Kim L and Zhou, Shaohua Kevin},
  booktitle={International MICCAI Workshop on Medical Computer Vision},
  pages={105--115},
  year={2013},
  organization={Springer}
}

@phdthesis{wang2014exploiting,
  title={Exploiting Geometric and Spatial Constraints for Vision and Lighting Applications},
  author={Quan Wang},
  year={2014},
  school={Rensselaer Polytechnic Institute},
}