@inproceedings{scholars15389,
       publisher = {Institute of Electrical and Electronics Engineers Inc.},
         journal = {2021 International Conference on Decision Aid Sciences and Application, DASA 2021},
           title = {Diagnosis of Alzheimer's Disease Using Pearson's Correlation and ReliefF Feature Selection Approach},
           pages = {578--582},
            note = {cited By 3; Conference of 2021 International Conference on Decision Aid Sciences and Application, DASA 2021 ; Conference Date: 7 December 2021 Through 8 December 2021; Conference Code:176623},
            year = {2021},
             doi = {10.1109/DASA53625.2021.9682409},
          author = {Sadiq, A. and Yahya, N. and Tang, T. B.},
            isbn = {9781665416344},
        keywords = {Computer aided diagnosis; Feature extraction; Magnetic resonance imaging; Nearest neighbor search; Neurodegenerative diseases; Neuroimaging, Alzheimers disease; Brain connectivity; Condition; Connectivity pattern; Features selection; Functional organization; Neuroimaging techniques; Pearson correlation; ReliefF; Resting-state functional magnetic resonance imaging, Correlation methods},
             url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85125770255&doi=10.1109\%2fDASA53625.2021.9682409&partnerID=40&md5=e91525edd19250503d05b14cabb34e89},
        abstract = {The study of brain connectivity patterns reveal important information in the understanding of the brain's functional organization. Resting-state functional magnetic resonance imaging (rs-fMRI) is a type of neuroimaging technique that can be used to diagnose a variety of neurological conditions. In this study, Pearson's correlation connectivity (PCC) and the feature selection algorithm ReliefF are used to distinguish Alzheimer's disease (AD) patients from normal controls (NC). PCC is a common measure to find the correlation between regions and ReliefF is known to perform well with high dimensional feature vectors so the combination of two gives a good accuracy. Using a k-nearest neighbor (KNN) classifier, the proposed method achieved a classification accuracy of 93.5 percent, showing the good potential of the proposed approach. {\^A}{\copyright} 2021 IEEE.}
}