Project: Analysis and Processing of Dental CT Scans


We present an approach to accurately align the CT scans of a patient to a stone-cast model of his/her mandible or maxilla, and use the result of registration to clean up the patient's scans from artifacts and defects. The proposed approach assumes that the maxillofacial features are roughly symmetric with respect to a 3D plane. Then 3D volumetric models of both the patient and the stone-cast are reconstructed from the input data using a marching cube algorithm. The planes of symmetry are extracted using an improved Extended Gaussian Images method. After an initial alignment of the two volumes guided by the plane of symmetry due to 3D homology, we minimize a global cost function that depends on the sum of square differences (SSD) of patient data with the stone-cast model to finally recover the rigid transformation between the two scans.

Keywords: CT Scan Alignment, Inpainting, Outlier Detection, Medical Imaging, 3D Visualization, Registration, Artifact Reduction, 3D Symmetry.


  • Murat Balci, Mais Alnasser, and Hassan Foroosh, "Alignment of Maxillofacial CT Scans to Stone-Cast Models Using 3D Symmetry for Backscattering Artifact Reduction", Medical Image Understanding and Analysis Conference (MIUA), 2006.

How does it work? 

The Process: (a) Original Patient Model (b) Corresponding Stone Cast Model (c) Initial Models with planes of symmetry (d) Models after planes of symmetry are matched (Top View) (e) A section of zoomed Patient Data (f) Partial zoom view of corrected data by use of stone-cast matching (g) Patient Model after clean up (h) Stone Cast model and cleaned up patient data merged together.

The Results: (a) Histogram of complete CT Scan data set (b) Part of the histogram cut by outlier threshold. Please note that there are just a few hundreds of voxels are marked as outlier in ~ 70 millions (c) An example of sample data slice. (d) Corresponding outliers marked as green. (e) Result after outlier removal is shown. Note that the color range is also improved by outlier removal as well.