Automated Language Analysis in Neurodegenerative Diseases: Application of Artificial Intelligence Models

This multicenter study aims to characterize the language profiles of patients with various neurodegenerative diseases, including Alzheimer's disease (AD), Primary Progressive Aphasia (PPA), behavioral variant Frontotemporal Dementia (bvFTD), Dementia with Lewy Bodies (DLB), and Parkinson's disease (PD), using artificial intelligence (AI) models. By analyzing speech and language data collected during remote neuropsychological assessments, the researchers intend to identify objective linguistic biomarkers suitable for early and differential diagnosis, overcoming the limitations of traditional, invasive, and costly diagnostic tools. The project will develop and validate an AI-driven pipeline to provide scalable and individualized linguistic profiles, ultimately supporting personalized treatment strategies and improving clinical workflow efficiency in the Italian population.

Participants with neurodegenerative diseases Inclusion criteria Diagnosis of mild cognitive impairment due to Alzheimer's disease (AD), primary progressive aphasia (PPA), behavioral variant frontotemporal dementia (bvFTD), dementia with Lewy bodies (DLB), or Parkinson's disease (PD). Age 40-85 years. Italian native speaker. Able to provide informed consent. Exclusion criteria Mini-Mental State Examination (MMSE) score below 15. Other neurological or psychiatric diagnoses, including cerebrovascular disease documented by MRI or CT. Severe, uncorrected hearing or visual impairment. Clinical Dementia Rating (CDR) score above 1. Healthy controls Inclusion criteria Italian native speaker. Matched to the patient groups for age, sex, and education level. Exclusion criteria History of systemic, neurological, or psychiatric disease. Substance abuse. Family history of neurodegenerative disease. MMSE score below 27/30.