AUTHOR=Zhao Ming , Li Jie , Xiang Liuqing , Zhang Zu-hai , Peng Sheng-Lung 

TITLE=A diagnosis model of dementia via machine learning

JOURNAL=Frontiers in Aging Neuroscience

VOLUME=Volume 14 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/aging-neuroscience/articles/10.3389/fnagi.2022.984894

DOI=10.3389/fnagi.2022.984894

ISSN=1663-4365

ABSTRACT=As the aging population poses a serious challenge to families and the society, the problem of dementia has also gradually received attention. In this paper, Bagging method is used to establish a model, and principal component analysis (PCA) is adopted to extract important features. By using the above two methods to select the important questions in the dementia test questionnaire, it could combine similar questions and establish an appropriate diagnostic model. Finally, this model will be verified according to the confusion matrix. The accuracy of the diagnostic model established by Bagging and PCA methods can reach 80% of the real diagnosis result. This model can significantly reduce the medical cost and time cost of dementia diagnosis. It can be applied to assist doctors in dementia diagnosis in reality.