Developing a speech-based surface electromyography paradigm for automated objective assessment of bulbar involvement in amyotrophic lateral sclerosis: Considerations of task effects on motor and cognitive-linguistic functions

Objective This study aims to delineate task effects on craniofacial neuromuscular performance and their associations with a hallmark feature of amyotrophic lateral sclerosis (ALS)—bulbar involvement—and the frequently cooccurring cognitive-linguistic impairment. Methods Using a surface electromyography (sEMG) paradigm, 54 features were extracted from 16 individuals with ALS and 10 neurologically healthy controls during an unscripted speech task, a scripted speech task, and two speech-like (i.e., multi- and single-syllable repetitions) tasks. These features were factorized into 10 composite measures, reflecting the latent traits of craniofacial neuromuscular performance. The task effects on these measures, their associations with standardized bulbar and cognitive metrics, and their efficacy in ALS-control classification were evaluated. Results Significant task-related differences were found in craniofacial neuromuscular performance. The sEMG measures were associated with (1) bulbar involvement across all tasks; (2) cognitive-linguistic impairment during all tasks except for single-syllable repetition. Unscripted speech demonstrated the highest accuracy (0.96) for ALS-control classification, followed by scripted speech (0.85), whereas speech-like tasks were more responsive to subclinical bulbar involvement during the prodromal phase. Conclusions The single-syllable repetition task is recommended for bulbar assessment, which may improve screening and early detection of bulbar involvement while minimizing the impact of cooccurring cognitive-linguistic impairment in ALS. Unscripted speech is most suited for monitoring global declines in communicative function driven by the combined effects of both bulbar and cognitive-linguistic impairments. Significance The findings offer guidance for task selection to facilitate the translation of emerging speech-based paradigms into structured bulbar assessment tools.