Skip to content

Manha27/dental-cbct-3d-segmentation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 

Repository files navigation

dental-cbct-3d-segmentation

Automated segmentation of dental anatomy from 3D CBCT scans using a 3D U-Net. Built during a funded research internship at NIT Calicut in collaboration with KMCT Hospital.


What it does

Segments six anatomical structures from volumetric dental CT scans:

  • Edentulous (missing tooth) regions
  • Mandible & Mandibular nerve
  • Upper & lower teeth
  • Maxilla / upper skull

Outputs are compatible with 3D Slicer for clinical visualization and implant planning.


Results

Structure Dice Score
Mandible 0.65
Lower Teeth 0.63
Maxilla & Upper Skull 0.62
Upper Teeth 0.61
Other 0.59
Edentulous Region 0.58
Mandibular Nerve 0.55

Mean Dice: ~0.60 across 167 manually annotated CBCT scans.


Stack

Python PyTorch MONAI SimpleITK nibabel 3D Slicer


Pipeline

DICOM → NIfTI conversion
      → Preprocessing (resampling, normalization, augmentation)
      → 3D U-Net training (500 epochs, Dice + CE loss, Adam lr=1e-4)
      → Prediction export → 3D Slicer visualization

Dataset

167 CBCT scans with manual segmentation masks annotated in 3D Slicer.

Dataset not publicly available due to patient privacy constraints.


Possible Extensions

  • Transformer-based architectures (Swin-UNETR)
  • Larger, more diverse dataset
  • Bone density estimation for implant planning

About

3D U-Net segmentation of dental CBCT scans using MONAI and PyTorch

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors