Early Detection of Alzheimer's Disease and Affective Disorders by Automated Voice and Speech Analysis (PLATA)

Centre Hospitalier Universitaire de Nice100 enrolled

Overview

PLATA aims to develop an algorithm to identify vocal biomarkers of Alzheimer's dementia. Using data collected as part of routine care, speech patterns will be compared to known biomarkers of Alzheimer's disease, such as amyloid 1-42 and p-Tau in CSF (cerebrospinal fluid). If biomarkers of speech can be identified in Alzheimer's disease, it is possible that patients and research participants will no longer need to undergo need to undergo the intensive and invasive baseline biomarker methods currently used, such as lumbar punctures and PET scans.

Study Type

OBSERVATIONAL

Enrollment

100

Conditions

Neurocognitive Disorders

Interventions

Series of cognitive tasks during a semi-automated callOTHER

Tasks: * Verbal learning recall (immediate) or Story Recall task (immediate) * Narrative Storytelling /free speech * Verbal fluency task * Verbal learning recall (delayed) or Story Recall task (delayed)

Eligibility

Medical Language ↔ Plain English

Inclusion Criteria: * Age ≥ 50 years * Diagnosis relevant biomarker and neuropsychological data already available * Cognitively healthy to very mild dementia (CDR score max. 0.5) * Sufficient knowledge of the study language to understand study information, non opposition form,and questionnaires * Expression of non opposition Exclusion Criteria: * Hearing problems * Patient protected by law, under guardianship or curator ship, or not able to participate in a clinical study according to the article L.1121-16 of the French Public Health Code

Outcomes

Primary Outcomes

Build and validate speech-based machine learning models for relevant Phenotype detection through access to phenotyped patients from reference memory center.

Speech biomarker algorithm(s)

Time frame: 20 minutes