Differentiating upper- and lower motor neuron diseases using automated acoustic analysis

Area of Science:

• Neurology
• Speech Science
• Biomarker Discovery

Background:

• Motor neuron diseases (MNDs) cause motor impairments, significantly impacting speech through dysarthria.
• Speech metrics are emerging as crucial biomarkers for MND diagnosis and phenotyping.

Purpose of the Study:

• To characterize the acoustic features of upper motor neuron (UMN) and lower motor neuron (LMN) dysarthria in MNDs.
• To explore the relationship between bulbar disease severity and acoustic speech features for personalized management.

Main Methods:

• Acoustic analysis of speech tasks (passage reading, syllable repetition, vowel phonation) from individuals with primary lateral sclerosis (PLS - UMN), spinal and bulbar muscular atrophy (SBMA - LMN), and healthy controls.
• Extracted 52 acoustic features covering articulation, phonation, prosody, resonance, and speech timing.
• Compared acoustic features between groups and correlated them with clinical bulbar disease severity scores.

Main Results:

• Articulatory and prosodic speech features significantly differentiated PLS, SBMA, and control groups.
• In PLS, correlations were found between clinical scores and articulatory features, particularly those related to tongue and jaw movements.

Conclusions:

• Acoustic assessment can identify characteristic speech 'fingerprints' of dysarthria in PLS and SBMA.
• Remote speech analysis holds potential for characterizing diverse dysarthria profiles and informing personalized clinical care and trial strategies.