Acoustic prosodic parameters associated with Parkinson's disease cognitive impairment

Area of Science:

• Neurology
• Speech-language pathology
• Biomedical acoustics

Background:

• Acoustic prosodic analysis shows promise for identifying mild cognitive impairment (MCI) and dementia in Alzheimer's disease (AD).
• The study hypothesizes that acoustic analysis can also differentiate cognitive impairment in Parkinson's disease (PD).

Purpose of the Study:

• To investigate the utility of acoustic parameters in differentiating cognitive impairment in Parkinson's disease (PD).
• To evaluate the importance and clinical implications of acoustic parameters for identifying PD with cognitive impairment.

Main Methods:

• Investigated acoustic parameters in 90 subjects: 30 PD with normal cognition (PD-NC), 30 PD with mild cognitive impairment (PD-MCI), and 30 PD with dementia (PDD).
• Utilized reading ('Supermarket Passage') and picture description ('Cookie Theft') tasks.
• Performed feature selection and modeling to assess acoustic parameters' role in identifying cognitive impairment.

Main Results:

• Acoustic parameters were independently associated with cognitive impairment (PDD and PD-MCI), with an AUC of 0.826 for detection.
• Speech rate, pre-verb pause (≥1s), and between-utterance pause (≥2s) in the 'Cookie Theft' task were key factors.
• Feature selection showed high AUC for discriminating PDD (0.944) and PD-MCI (0.753) from PD-NC.

Conclusions:

• Acoustic parameters effectively differentiate PD patients with cognitive impairment from those with normal cognition, independent of age.
• Acoustic parameters show potential for automated screening of cognitive decline risk in PD patients.
• Further investigation into acoustic analysis for PD cognitive assessment is warranted.