Multi-atlas segmentation in pediatric MRI (Kang et al., 2026)
Year: 2026
Type: other
Abstract: Purpose: Segmentation of velopharyngeal and vocal tract anatomy is crucial for quantifying structures and analyzing dynamics during speech. Automatic methods are needed to replace labor-intensive and poorly reproducible manual segmentation. To address this, we compared four multi-atlas–based methods to determine which provides the most accurate segmentation of the tongue, velum, and adenoid and how segmentation accuracy varies with the number of temporal frames.Method: Five spatiotemporal atlases of speech tasks were built, each with around 80 frames. We performed diffeomorphic registration of 10–40 frames of each atlas, then propagated labels to evaluate how the number of frames impacts performance. Four label fusion techniques—majority voting label fusion (MVLF), simultaneous truth and performance level estimation (STAPLE), multi-atlas label fusion (MALF), and corrective learning (CL)—were applied to evaluate segmentation accuracy. Performance was assessed using dice similarity coefficient (DSC) values across increasing numbers of input frames (10–40).Results: CL consistently produced the highest segmentation accuracy, with average DSC improving from 0.89 at 10 frames to 0.92 at 40 frames. In contrast, MVLF, STAPLE, and MALF showed relatively flat trends, with DSC values ranging from 0.82 to 0.87. Statistical analysis confirmed that only CL exhibited a significant positive association between frame count and DSC.Conclusions: CL demonstrates superior performance in segmenting pediatric vocal tract structures in dynamic magnetic resonance imaging, particularly as the number of temporal frames increases. These findings support the use of CL in analyzing speech-related anatomy and highlight its potential to replace manual segmentation in small specialized data sets.Supplemental Material S1. Dynamic Cookie atlas across 48 consecutive time frames (tf01–tf48). Each time frame is displayed in three planes: left = midsagittal, right = coronal, and bottom = axial. Manual segmentations for the tongue, velum, and adenoid are overlaid to illustrate frame to-frame articulatory motion and spatiotemporal changes.Supplemental Material S2. Comparison of segmentation methods on the Sissie atlas at five time frames (tf01, tf11, tf21, tf31, tf41). Each frame is shown across five conditions, from left to right: Manual, MVLF, STAPLE, MALF, and CL. This video enables direct visual comparison of segmentation quality across methods.Kang, H., Xing, F., Gilbert, I. R., Kim, J., Kostolansky, A., Sutton, B., Woo, J., & Perry, J. L. (2026). Multi-atlas–based segmentation of pediatric vocal tract anatomy in dynamic magnetic resonance imaging. Journal of Speech, Language, and Hearing Research. Advance online publication. https://doi.org/10.1044/2026_JSLHR-25-00495
Source: figshare ASHA Publications
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7202262289