AI detects Alzheimer's from speech using acoustic features

Researchers have developed a lightweight method for detecting Alzheimer's disease using only spontaneous speech audio. This approach avoids the need for transcripts or computationally intensive deep learning models, instead focusing on handcrafted acoustic-temporal features like pause and fluency statistics, as well as Mel Frequency Cepstral Coefficients (MFCCs). Using a Support Vector Machine with an RBF kernel on the DementiaBank Pitt corpus, the system achieved an average AUC of 0.674, demonstrating the potential for spectro-temporal and fluency cues in early Alzheimer's screening. AI

IMPACT Offers a potential low-cost, non-invasive method for early Alzheimer's disease screening, reducing reliance on expensive neuroimaging or language-based tools.

RANK_REASON Academic paper detailing a novel method for disease detection. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• Alzheimer's disease
• Cookie Theft Picture Description: Linguistic and Neural Correlates
• DementiaBank Pitt corpus
• Mel Frequency Cepstral Coefficients
• random forest
• Rbf Kernel
• support vector machine
• WebRTC

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →