Speech-language pathology ·
Mayo Clinic deep learning markers track articulatory impairment in motor speech disorders; framework structures velopharyngeal MRI reporting
Listen to this post
read by AI voices · all episodesTHE SHORT READ
Mayo Clinic researchers reported that phoneme-level outputs of a deep learning model were significantly reduced in 199 people with motor speech disorders, compared with 477 without, on a remotely captured word repetition task. The outputs fell further with greater severity. A University of Virginia and Children's Healthcare of Atlanta team published a framework for interpreting and reporting velopharyngeal magnetic resonance imaging.
THE DETAILS
- AI Oct 13, 2026Mayo Clinic researchers tested whether phoneme-level posterior probabilities from Phonet, an articulatory-informed deep learning model, can serve as markers of articulatory impairment in motor speech disorders, in the Journal of Speech, Language, and Hearing Research. Among 676 participants (199 with motor speech disorders, 477 without) who completed a remotely captured word repetition task, posterior probabilities were significantly reduced in the motor speech disorder group across all phoneme types. Within that group, they decreased progressively with greater severity on perceptual sound distortion ratings and a cumulative articulation error score. The abstract reports no accuracy figures.
- Guidelines Oct 13, 2026A team from the University of Virginia and Children's Healthcare of Atlanta published a literature-informed, consensus-based framework for interpreting velopharyngeal magnetic resonance imaging (MRI) in the American Journal of Speech-Language Pathology. The framework organizes interpretation across five domains and was successfully applied across pilot cases, supporting consistent interpretation and documentation, the authors wrote. They said it gives clinicians a practical method for integrating velopharyngeal MRI observations with speech assessment. The abstract reports no numerical results.
The weekly innovation roundup for speech-language pathology, published every Sunday for the week before; every item links to its sources. Latest speech-language pathology news · All specialties