Objective: To develop a digital speech assessment tool to aid in identifying different neurological disorders.

Background: Speech impairment is a prevalent symptom in neurological disorders, however the mechanism and extent of speech impairment varies significantly in different disorders, and even within patients with the same disease. Thus, there is a need for scalable solutions to assess speech impairment as a potential tool for diagnosing and tracking neurological and neuromuscular disorders.

Method: Passage reading speech assessments were performed in four different studies using BioDigit Speech: 1) amyotrophic lateral sclerosis (ALS) (n=12, 40 recordings) 2) progressive supranuclear palsy (PSP) (n=18, 35 recordings), 3) Parkinson’s disease (PD) (n=13, 13 recordings) 4) Huntington’s Disease (HD) (n=18, recordings=42), and 5) age- and sex-matched control (n=16, recordings=24). Group differences and correlations with clinical scores were explored. A machine learning classifier was trained to automatically differentiate different patient populations based on their speech features.

Results: Machine learning classification models achieved a weighted accuracy of 90% in identifying the disease from simple speech tasks with a sensitivity of 79% for PSP, 90% for PD, 100% for ALS, 86% for HD and 90% for controls. In ALS, bulbar dysfunction as measured by the ALSFRS-R was associated with reduced articulatory rate and intelligibility. Similar observations were made in PSP and HD. In addition, multiple speech measures correlated with the MoCA, including similarity and intelligibility in PD, HD and PSP. Machine learning models demonstrated strong capabilities in predicting clinical diagnoses and outcomes with high accuracy and sensitivity.

Conclusion: Our findings highlight the potential of BioDigit Speech as a valuable tool to aid in identifying and tracking multiple neuromuscular and neurodegenerative disorders.

« Back to 2024 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/speech-analysis-and-machine-learning-can-aid-in-identifying-different-neurological-disorders/