Early Detecting Alzheimer Disease Using Deep Learning

Dementia can be considered as an “umbrella” in which different type of diseases associated with cognitive impairment lies underneath, Alzheimer disease represents the majority with 50%–75% of the cases. As of today, there is no cure from this disease and the only way to prevent any associated medical, economic, and financial impacts or losses is to early detect the disease and work closely with suspected patients to prevent any further progress. In this research we proposed a methodology consist of 4 modules, preprocessing, exemplar pyramid along with bi-linear interpolation followed by feature extraction using GLCM and LBP then concatenation of all extracted features and finally classification of Alzheimer disease stage using deep learning, Multi-Layer Perceptron, in particular. Our proposed method was tested using MPRAGE structural MRI dataset from Alzheimer Disease Neuro Imaging Initiative (ADNI) and outperformed state of the art methodologies in major performance evaluation parameters enclose accuracy. An accuracy result of 89.80 was reported for multi-class classification of 4 stages (CN, EMCI, LMCI and AD) of Alzheimer disease for both GM and WM. In term of binary-class classification, GM performance surpass the WM and CSF distinguishing between CN vs EMCI, EMCI vs AD and LMCI vs AD with accuracy results of 96.43%, 90.91% and 95.24% respectively. On the other hand, WM performance surpass the GM and CSF distinguishing between CN vs LMCI with 100% accuracy and EMCI vs LMCI with 95.65% accuracy. The reported results show that the WM performance distinguishing between CN vs AD was like the GM performance with an accuracy result of 96.15.