This technique corrects for array motion in hyoid bone measurements, improving ultrasound accuracy and reliability.
Ultrasound imaging is able to track oropharyngeal structures used in swallowing, e.g., the hyoid bone. Studies using simultaneous B-mode and x-ray videofluoroscopic imaging have shown that accuracy of ultrasound-measured hyoid movements is highly dependent on motion of the ultrasound array in relation to vocal tract anatomy. Here, we address this challenge by correcting for array motion, measured either by videofluoroscopy or by an inertial measurement unit (IMU) integrated into a handheld, wireless scanner (Clarius C7 HD3). For videofluoroscopy, a Harris corner tracker is used in CASM (Computational Analysis of Swallowing Mechanics) MATLAB-based software for semi-automatic tracking of reference points on the array. For IMU measurements, time-dependent 3-axis acceleration and angular velocity measurements are smoothed and integrated to obtain time-dependent array displacement and angle. The measured hyoid trajectory is then corrected using coordinate transforms accounting for the array’s time-dependenttranslation and in-plane rotation. Hyoid trajectories and velocities, measured using CASM from synchronized B-mode and standard-of-care fluoroscopy recordings, are compared using intraclass correlation coefficients and root-mean-square error between the two modalities, with and without correction for array motion. Results indicate significant improvement in measurement accuracy when correcting for array motion, while also characterizing the magnitude of measurement errors for uncorrected handheld ultrasound imaging.
No takes yet. Share an insight, caveat, or question.
Schoenleb et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: