Method to control tongue position and open source image segmentation for cone-beam computed tomography of patients with large palatal defect to facilitate digital obturator design

Objective: To allow for easy segmentation in open source image processing software with minimal effort and human errors and facilitate the overall process of CAD based obturator design. 

Method: The patient's existing lower denture was modified with cross-arch tongue cribs/bars made of thermoplastic impression material which acted as a restrictive barrier for the tongue. A cone-beam computed tomography (CBCT) was done and the images were then carried over to an open source image processing software and the defect site was segmented from the DICOM and picked for conversion into 3D model. The 3D model could then be used to design the prosthesis in CAD software. 

Result: The method of control and segmentation produced better texture and morphological details. The digital obturator designed was able to engage peripheral margins on the 3D model. 

Conclusion: The use of tongue restriction technique and open-source image segmentation can be a cost effective and viable option for easing the overall process of digitally designing an obturator for palatal defect.