Fangxu Xing1, Riwei Jin2, Imani Gilbert3, Xiaofeng Liu1, Georges El Fakhri1, Jamie Perry3, Bradley Sutton2, and Jonghye Woo1
1Department of Radiology, Massachusetts General Hospital, Boston, MA, United States, 2Department of Bioengineering, University of Illinois at Urbana-Champaign, Champaign, IL, United States, 3Department of Communication Sciences and Disorders, East Carolina University, Greenville, NC, United States
1Department of Radiology, Massachusetts General Hospital, Boston, MA, United States, 2Department of Bioengineering, University of Illinois at Urbana-Champaign, Champaign, IL, United States, 3Department of Communication Sciences and Disorders, East Carolina University, Greenville, NC, United States
Temporal alignment on a total number of 93 image volumes
showed an increased normalized cross-correlation score on average, indicating
improved similarity between the reference target sequence and source subject
sequences after processed by the proposed pipeline.
Figure 2. Mid-sagittal
slices at four selected time frames from the reference subject and one test
subject (before and after time alignment) during the pronunciation of “hamper”.
Figure 1. Wave
form matching of two test subjects to the target reference subject.