https://doi.org/10.1140/epjs/s11734-024-01428-4
Regular Article
An automatic Alzheimer’s disease classifier based on reading task for Spanish language
1
Área Académica de Matemáticas y Física, Universidad Autónoma del Estado de Hidalgo, Carr. Pachuca-Tulancingo Km. 4.5, 42184, Mineral de la Reforma, Hidalgo, Mexico
2
Biología de Sitemas, Centro de Investigación y de Estudios Avanzados Unidad Monterrey, Vía del Conocimiento 201, 66600, Apodaca, Nuevo León, Mexico
Received:
30
August
2024
Accepted:
22
November
2024
Published online:
10
December
2024
This research develops and evaluates a deep learning model designed to identify Alzheimer’s disease (AD) and mild cognitive impairment (MCI) through Spanish language audio recordings. Addressing the scarcity of data on Spanish-speaking populations in Alzheimer’s research, our study presents a tailored strategy to bridge this gap. A convolutional neural network (CNN) was trained with a dataset comprising reading task audio from 361 participants, encompassing healthy individuals, MCI, and AD patients. The model’s accuracy was enhanced through data augmentation techniques and refined with an attention layer during fine-tuning. Our method achieved a classification accuracy of 73.03% in distinguishing between the three groups. Offering a cost-effective, non-invasive, and readily deployable solution for the early detection of Alzheimer’s and MCI, this approach shows promising results and potential for integration into clinical settings, especially in regions where such populations are largely underrepresented.
Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
© The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature 2024
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
