10Classification of Dementia Using Statistical First-Order and Second-Order Features
Deepika Bansal1* and Rita Chhikara2
1Department of Information Technology, Maharaja Agrasen Institute of Technology, Rohini, India
2Department of Computer Science Engineering, The NorthCap University, Gurugram, India
Abstract
A neurological disorder called dementia causes memory loss, which interferes with a person’s ability to live a normal life. Adults 65 years of age and older are starting to experience it as a global health issue. An early diagnosis can be very helpful to slow down the progress of disease. The purpose of this study is to predict the presence of dementia using magnetic resonance imaging data, which have become a significant tool for the diagnosis of dementia. The publicly available Open Access Series of Imaging Studies cross-sectional MRI data were analyzed. This paper investigates an ensemble of clinical features, first-order features and second-order wavelet features of the magnetic resonance images for classification. Discrete wavelet transform with Haar transform is applied on images for feature extraction. The MRIs of 200 individuals were used for this study, which includes 30 demented, 70 very mild demented, and 100 normal controls. Randomly 75% of the data are used for training and 25% are used for testing purposes. The results with the approximation coefficient using the Haar wavelet and ensembled features outperform with 99.27% accuracy for the two-class (dementia/normal) ...
Get Blockchain and Deep Learning for Smart Healthcare now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
