Early Onset Alzheimer Disease Classification Using Convolution Neural Network

Abstract

Alzheimer’s disease is one of the major causes of death. The disease treatment is highly recommended in its early stage as it is difficult to treat it in later stage. The diagnosis of this slow growing disease is difficult as it does not show any symptoms in the early stage. As the deep neural networks have shown its success to process the medical images, the paper uses convolution neural network for early detection of the Alzheimer’s disease using binary classification. The network model uses $T_{2}$ magnetic resonance images from Alzheimer’s disease neuroimaging initiative dataset. The preprocessing extracts the slices with hippocampal region from the three-dimensional images and removes non-brain region of the slice. The proposed method achieves 71.13% accuracy. It performs better than AlexNet in terms of loss and prediction time.

Data used in preparation of this article were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.usc.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf.