Speech-based detection of multi-class Alzheimer’s disease classification using machine learning

• Published in: International journal of data science and analytics Vol. 18; no. 1; pp. 83 - 96
• Main Authors: Tripathi, Tripti, Kumar, Rakesh
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.06.2024; Springer Nature B.V
• Subjects: Algorithms; Artificial Intelligence; Business Information Systems; Classification; Computational Biology/Bioinformatics; Computer Science; Data Mining and Knowledge Discovery; Database Management; Impairment; Machine learning; Public health; Regular Paper; Speech; Statistical tests; Alzheimer’s disease; Classification; Machine learning
• ISSN: 2364-415X; 2364-4168
• DOI: 10.1007/s41060-023-00475-9

Alzheimer’s disease, a significant global health concern, necessitates early detection for effective treatment and management. This research introduces an innovative method for classifying six types of cognitive impairment via speech-based analysis: probable AD, possible AD, MCI, memory impairments, vascular dementia, and control. Leveraging speech data from DementiaBank’s Pitt Corpus, we preprocess the data to extract relevant acoustic features. These features are then employed to train five machine learning algorithms: KNN, DT, SVM, XGBoost, and RF. The study's results indicate an overall accuracy of 75.59% in the six-class classification challenge. Furthermore, statistical tests establish the statistical significance of the differences in accuracy between XGBoost and the other algorithms, except for random forest. This approach has the potential to evolve into a non-invasive, cost-effective, and readily accessible diagnostic tool for early cognitive impairment detection and management.