Machine Learning-Based Prediction Models for Cognitive Decline Progression: A Comparative Study in Multilingual Settings Using Speech Analysis

• DOI: 10.14283/jarlife.2024.6
• Publication Date: 2024-05-17T02:31:37Z
• Authors: B. Ceyhan; S. Bek; T. Onal-Suzek
• Keywords: Cognitive Decline

BACKGROUND: Mild cognitive impairment (MCI) is a condition commonly associated with dementia.Therefore, early prediction of progression from MCI to dementia essential for preventing or alleviating decline.Given that affects functions like language and speech, detecting disease through speech analysis can provide costeffective solution patients caregivers.DESIGN-PARTICIPANTS: In our study, we examined spontaneous (SS) written Mini Mental Status Examination (MMSE) scores 60-patient dataset obtained the Mugla University Dementia Outpatient Clinic (MUDC) 153-patient Alzheimer 's Recognition Spontaneous Speech (ADRess) challenge.Our first time, analyzed impact audio features extracted SS in distinguishing between different degrees using both an Indo-European Turkic language, which exhibit distinct word order, agglutination, noun cases, grammatical markers.RESULTS: When each machine learning model was tested on its respective trained attained 95% accuracy random forest classifier ADRess 94% MUDC employing multilayer perceptron (MLP) neural network algorithm.In second experiment, evaluated effectiveness languagespecific other language.We achieved accuracies 72% English 76% Turkish, respectively.CONCLUSION: These findings underscore cross-language potential automated tracking patients, offering convenient cost-effective option clinicians patients.