English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Ultrasound imaging is safe, relatively affordable, and capable of real-timeperformance. One application of this technology is to visualize and tocharacterize human tongue shape and motion during a real-time speech to studyhealthy or impaired speech production. Due to the noisy nature of ultrasoundimages with low-contrast characteristic, it might require expertise fornon-expert users to recognize organ shape such as tongue surface (dorsum). Toalleviate this difficulty for quantitative analysis of tongue shape and motion,tongue surface can be extracted, tracked, and visualized instead of the wholetongue region. Delineating the tongue surface from each frame is a cumbersome,subjective, and error-prone task. Furthermore, the rapidity and complexity oftongue gestures have made it a challenging task, and manual segmentation is nota feasible solution for real-time applications. Employing the power ofstate-of-the-art deep neural network models and training techniques, it isfeasible to implement new fully-automatic, accurate, and robust segmentationmethods with the capability of real-time performance, applicable for trackingof the tongue contours during the speech. This paper presents two novel deepneural network models named BowNet and wBowNet benefits from the ability ofglobal prediction of decoding-encoding models, with integrated multi-scalecontextual information, and capability of full-resolution (local) extraction ofdilated convolutions. Experimental results using several ultrasound tongueimage datasets revealed that the combination of both localization andglobalization searching could improve prediction result significantly.Assessment of BowNet models using both qualitatively and quantitatively studiesshowed them outstanding achievements in terms of accuracy and robustness incomparison with similar techniques.

BowNet: Dilated Convolution Neural Network for Ultrasound Tongue Contour Extraction