3D Dynamic Visualization of Swallowing from Multi-Slice Computed Tomography

Authors

Andrew K. Ho, University of British ColumbiaFollow
Mark A. Nicosia, Widener University
Angela Dietsch, Walter Reed National Military Medical CenterFollow
William Pearson, Georgia Regents University
Jana Rieger, University of Alberta
Nancy Pearl Solomon, Walter Reed National Military Medical CenterFollow
Maureen Stone, University of Maryland
Yoko Inamoto, Fujita Health University
Eiichi Saitoh, Fujita Health University
Sheldon Green, University of British Columbia
Sidney Fels, University of British Columbia

ORCID IDs

Dietsch 0000-0003-4554-5365

Document Type

Article

Date of this Version

2014

Citation

Proceedings, SIGGRAPH '14, ACM SIGGRAPH 2014

Posters, Article number 103

Comments

United States government work

Abstract

Human swallowing and its disorders (dysphagia) are still poorly understood, and yet many speech-language pathologists (SLPs) need to be trained to recognize correct, incorrect, and potentially dangerous swallows. The anatomy of the head and neck region is notoriously complex and difficult to visualize and study. Currently, training programs that teach SLPs to recognize swallowing disorders use artistically derived animations of swallowing, rendered at fixed viewpoints, to help students visualize the anatomy of the head and neck region.

This work improves on these animations by using state-of-the-art medical images to create a dynamic, interactive, 3D simulation of human swallowing. Images of a male subject during swallow were captured in a single shot using a 320-slice CT scanner [Inamoto et al. 2011]. The images have very high spatial resolution (0:5 x 0:5 x 0:5 mm³), but low temporal resolution (10 Hz). The low temporal resolution resulted in blurring of the fluid being swallowed, making automatic segmentation and visualizations of the fluid difficult to generate.