Previously submitted to: JMIR Mental Health (no longer under consideration since Dec 07, 2021)
Date Submitted: Dec 7, 2021
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Deep Learning of Speech Data for Early Detection of Alzheimer’s Disease in the Elderly: Preliminary Study
ABSTRACT
Background:
Alzheimer's disease is the most common form of dementia, and it is a disease that makes it difficult for patients and their families due to various symptoms. For these reasons, early detection is very important, and after early detection, symptoms can be alleviated through medication and treatment.
Objective:
Since Alzheimer's disease strongly induces language disorders, our research goal is detecting Alzheimer's disease quickly and easily through the analysis of language characteristics.
Methods:
Using the Mini-Mental State Examination for Dementia Screening (MMSE-DS), which is the most used in Korean public health centers, negative answers were obtained according to the questionnaire. Among the acquired voices, significant questionnaires and answers were selected, spectrogrammed, and converted into MFCC. After accumulating significant answers, training data was created, augmented, and then trained on various deep learning models and the results were observed.
Results:
Due to the lack of data, the results of the five-fold cross validation were more significant than the holdout method. the accuracy of separating AD patients from the control group using Densnet121 was 91.25%.
Conclusions:
In this regard, the potential for remote health care can be increased by simplifying the AD screening process. By facilitating remote health care, the proposed method is expected to enhancing the accessibility of AD screening and increase the rate of early AD detection. Clinical Trial: IRB No. CNUH2019-02-068
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.