AUTHOR=Sun Dazhong , Peng Haojun , Wu Zhibing 

TITLE=Establishment and Analysis of a Combined Diagnostic Model of Alzheimer's Disease With Random Forest and Artificial Neural Network

JOURNAL=Frontiers in Aging Neuroscience

VOLUME=Volume 14 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/aging-neuroscience/articles/10.3389/fnagi.2022.921906

DOI=10.3389/fnagi.2022.921906

ISSN=1663-4365

ABSTRACT=Alzheimer’s disease (AD) is a chronic neurodegenerative disease. Due to the limitations of existing diagnostic techniques for AD, it is necessary to develop novel diagnostic models to supplement existing methods. Few studies, however, have attempted to develop a diagnostic model based on gene biomarkers. To identify gene biomarkers and construct a diagnostic model, we used a computational method that combined two machine learning algorithms, including random forest (RF) and artificial neural network (ANN). We collected AD gene expression data from Gene Expression Omnibus (GEO) database; four datasets were utilized, including two training dataset for screening differentially expressed genes (DEGs) and two validation datasets. Firstly, based on RF, 6 key genes(NFKBIA, SST, KLF15, NDUFA7, MAFF, and UCHL1) in 177 key DEGs were identified to be vital for classification of AD and normal samples. The weights of these key genes were calculated and the diagnostic models were developed using ANN. Finally, two validation datasets were used to test and compare the performance by area under curve (AUC). Our model achieved an AUC of 0.822 in GSE109887, and 0.814 in GSE132903. To conclude, we uncovered gene biomarkers and successfully constructed a new diagnostic model of AD using an artificial neural network and verified its diagnostic efficacy in public datasets, which would be helpful for diagnosis.