Saved in:
Bibliographic Details
Main Authors: Myrgyyassov, Alisher, Wang, Bruce Xiao, Sun, Yu, Huang, Shuming, Song, Zhen, Wong, Min Ney, Zheng, Yongping
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.03350
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918368786776064
author Myrgyyassov, Alisher
Wang, Bruce Xiao
Sun, Yu
Huang, Shuming
Song, Zhen
Wong, Min Ney
Zheng, Yongping
author_facet Myrgyyassov, Alisher
Wang, Bruce Xiao
Sun, Yu
Huang, Shuming
Song, Zhen
Wong, Min Ney
Zheng, Yongping
contents Manual measurement of muscle morphology from ultrasound during speech is time-consuming and limits large-scale studies. We present SMMA, a fully automated framework that combines deep-learning segmentation with skeleton-based thickness quantification to analyze geniohyoid (GH) muscle dynamics. Validation demonstrates near-human-level accuracy (Dice = 0.9037, MAE = 0.53 mm, r = 0.901). Application to Cantonese vowel production (N = 11) reveals systematic patterns: /a:/ shows significantly greater GH thickness (7.29 mm) than /i:/ (5.95 mm, p < 0.001, Cohen's d > 1.3), suggesting greater GH activation during production of /a:/ than /i:/, consistent with its role in mandibular depression. Sex differences (5-8% greater in males) reflect anatomical scaling. SMMA achieves expert-validated accuracy while eliminating the need for manual annotation, enabling scalable investigations of speech motor control and objective assessment of speech and swallowing disorders.
format Preprint
id arxiv_https___arxiv_org_abs_2603_03350
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Automated Measurement of Geniohyoid Muscle Thickness During Speech Using Deep Learning and Ultrasound
Myrgyyassov, Alisher
Wang, Bruce Xiao
Sun, Yu
Huang, Shuming
Song, Zhen
Wong, Min Ney
Zheng, Yongping
Quantitative Methods
Machine Learning
Sound
Audio and Speech Processing
Manual measurement of muscle morphology from ultrasound during speech is time-consuming and limits large-scale studies. We present SMMA, a fully automated framework that combines deep-learning segmentation with skeleton-based thickness quantification to analyze geniohyoid (GH) muscle dynamics. Validation demonstrates near-human-level accuracy (Dice = 0.9037, MAE = 0.53 mm, r = 0.901). Application to Cantonese vowel production (N = 11) reveals systematic patterns: /a:/ shows significantly greater GH thickness (7.29 mm) than /i:/ (5.95 mm, p < 0.001, Cohen's d > 1.3), suggesting greater GH activation during production of /a:/ than /i:/, consistent with its role in mandibular depression. Sex differences (5-8% greater in males) reflect anatomical scaling. SMMA achieves expert-validated accuracy while eliminating the need for manual annotation, enabling scalable investigations of speech motor control and objective assessment of speech and swallowing disorders.
title Automated Measurement of Geniohyoid Muscle Thickness During Speech Using Deep Learning and Ultrasound
topic Quantitative Methods
Machine Learning
Sound
Audio and Speech Processing
url https://arxiv.org/abs/2603.03350