Houri Vorperian, PhD

Slide of the Week: Houri K. Vorperian, PhD

Title: A novel registration-based semi-automatic mandible segmentation pipeline using computed tomography images to study mandibular development

Legend: A semi-automatic mandible segmentation (SAMS) pipeline was developed to isolate 3D mandible/jaw models from CT studies for the purpose of studying the developmental changes in the size and shape of the mandible/jaw. The top panel, showcasing three male 3D mandible models at ages 2;10 (years;months), 10;11  and 18;10 depicts the characteristic downward and forward growth of the mandible. Note the increased length of the ramus, length of the mandible body, and protrusion of the mental protuberance. The bottom panel is a visual display assessing the successful implementation of our SAMS pipeline. The blue mandible (left) is that of a 6;02 (years; months) old female that was manually segmented, the gray (middle) mandible is the same mandible that was automatically segmented using the SAMS pipeline, and the final display (right) is that of the two models superimposed. This visual display shows that the SAMS pipeline was fairly accurate but requires minor edits in the region of the condyles and coronoid processes.

Citation: Chuang, Y. J., Doherty, B. M., Adluru, N., Chung, M. K. and Vorperian, H. K. (In Press, 2017). A novel registration-based semi-automatic mandible segmentation pipeline using computed tomography images to study mandibular development. Journal of Computer Assisted Tomography. NIHMSID: NIHMS894554.  SAMS pipeline documentation: http://www.waisman.wisc.edu/vocal/samsdoc

Abstract:

Objective: We present a registration-based semi-automatic mandible segmentation (SAMS) pipeline designed to process a large number of computed tomography studies to segment three dimensional (3D) mandibles. Method: The pipeline consists of a manual preprocessing step, an automatic segmentation step, and a final manual post-processing step. The automatic portion uses a nonlinear diffeomorphic method to register each preprocessed input CT test scan on 54 reference templates aged birth to 19 years. This creates 54 segmentations, which are then combined into a single composite mandible. Results: This pipeline was assessed using 20 mandibles from CT studies ages 1-19 years, segmented using both SAMS-processing and manual segmentation. Comparisons between the SAMS-processed and manually-segmented mandibles revealed 97% similarity agreement with comparable volumes. The resulting 3D mandibles were further enhanced with manual post-processing in specific regions. Conclusions: Findings are indicative of a robust pipeline that reduces manual segmentation time by 75% and increases the feasibility of large-scale mandibular growth studies.

About the Lab: The Vocal Tract Development Laboratory (VTLab) uses a combination of imaging, acoustics, and vocal tract modeling to understand the lifespan changes of the vocal tract anatomy in typically and atypically developing individuals, and to examine the relation of anatomic changes to speech acoustics.

Slide of the Week Archives