Early Prediction of Alzheimer’s Disease Using Hybrid Neuro-Fuzzy System with Cognitive Test Data and Speech Analysis
Vidhyayana · 30.05.2026 · 11 (5)
DOI: 10.58213/6706an50 ↗Kurz gefasst
Fachartikel in der Zeitschrift Vidhyayana, online erschienen am 30.05.2026. Verfasst von Arti Patel. Die Arbeit nennt 13 Literaturangaben.
Themen: Demenz
Abstract (Originalsprache)
Alzheimer's Disease (AD) is the most common neurodegenerative condition and the primary cause of dementia in the elderly globally. A timely diagnosis is crucial; however, traditional diagnostic techniques mostly rely on neuroimaging (MRI/PET scans), biomarker assessment, and clinical evaluations, which are often intrusive, expensive, or need specialised medical apparatus. Recent research demonstrate that minor reductions in speech fluency, vocabulary diversity, cognitive reactivity, and hesitation patterns function as measurable indicators of Mild Cognitive Impairment (MCI), the preliminary discernible stage of Alzheimer's disease. This study presents a novel hybrid Neuro-Fuzzy inference diagnostic framework that combines speech pattern analysis with cognitive test scores (MMSE, MoCA) to forecast the advancement of Alzheimer's disease using cost-effective and non-invasive methods. We utilised voice datasets from DementiaBank (Pitt Corpus) and the ADReSS Challenge Dataset to get linguistic and temporal characteristics such as pause duration, speech rate, semantic density, and lexical variety. Cognitive inputs were recorded as fuzzy membership functions. The proposed Adaptive Neuro-Fuzzy Inference System (ANFIS) model was developed to categorise individuals as normal, exhibiting Mild Cognitive Impairment (MCI), or diagnosed with Alzheimer's Disease (AD). The experimental results indicate that ANFIS outperforms both ANN and fuzzy classifiers in predictive accuracy. The accuracy is 92.18%, the sensitivity is 90.47%, and the specificity is 93.10%. The findings demonstrate that soft computing techniques, particularly hybrid neuro-fuzzy systems, can be effectively employed for the early detection of Alzheimer's disease in clinical and residential settings. Future research will incorporate deep learning-based acoustic feature extraction and implement the model as a mobile application for real-time remote diagnosis..
Zitieren
APA
Patel, A. (2026). Early Prediction of Alzheimer’s Disease Using Hybrid Neuro-Fuzzy System with Cognitive Test Data and Speech Analysis. Vidhyayana, 11(5). https://doi.org/10.58213/6706an50
Vancouver
Patel A. Early Prediction of Alzheimer’s Disease Using Hybrid Neuro-Fuzzy System with Cognitive Test Data and Speech Analysis. Vidhyayana. 2026;11(5). doi:10.58213/6706an50
BibTeX
@article{Patel2026,
author = {Patel, Arti},
title = {Early Prediction of Alzheimer’s Disease Using Hybrid Neuro-Fuzzy System with Cognitive Test Data and Speech Analysis},
journal = {Vidhyayana},
year = {2026},
volume = {11},
number = {5},
doi = {10.58213/6706an50}
}
EUVIMED