A COMPARATIVE SURVEY OF FEATURE EXTRACTION AND CLASSIFICATION TECHNIQUES FOR EARLY DIAGNOSIS OF ALZHIMER’S DISEASE

sherin A, Dr. R. Rajeswari

Abstract


Alzheimer’s disease (AD) is a neurodegenerative disorder of the brain in the elderly population AD is most severe and common form of Dementia that affects memory and cognitive functions of the elder people with behavioral impairment. Various Computer-Aided Diagnosis (CAD) systems have been developed for early diagnosis of AD available in the literature. All CAD techniques select and extract some feature vectors such as Principal Component Analysis (PCA), Partial Least Square (PLS), random forest etc. for AD diagnosis. In this paper, various feature extraction and classification approaches using the three imaging modalities, MRI, SPECT and PET along with their merits and demerits are discussed. In particular, this paper mainly focuses on feature extraction and classification approaches for early diagnosis of AD. A tabular demonstration of all approaches is presented to facilitate the comparison. Some discussion about the future enhancement in this direction by identifying some open research problems is also presented.

Keywords


Alzheimer’s disease, CSF, EM algorithm, ICA, Multimodal classification and Non-negative matrix factorization.

Full Text:

PDF


DOI: https://doi.org/10.26483/ijarcs.v9i2.5353

Refbacks

  • There are currently no refbacks.




Copyright (c) 2018 International Journal of Advanced Research in Computer Science