Alzhemy: A cloud enabled machine learning model for Alzheimer’s disease prediction

Alzheimer’s disease (AD) is a generally well-known and common neurological illness that produces cognitive problems. Even though in medical and medical domains, it is one of the commonly studied disorders of the nervous system notwithstanding that it has no treatment or any technique to reduce or prevent its course. Nevertheless, there are numerous choices (drug or non-drug solutions) which may assist in treating indications of the AD at its various stages to enhance the patient’s life quality. As the AD develops with time, the individuals at its different phases need to be handled differently. For that reason, the early detection and classification of the phases of the AD can be very beneficial for the cure of the disease’s symptoms. In this work, we have developed a cloud based machine learning framework called Alzhemy for the early detection of AD. Experiments on the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset revealed that the logistic regression model performed superior to the Random Forest, k-Nearest Neighbour, Naive Bayes Classifier, and the AdaBoost Classifier in terms of accuracy.