Alzheimer’s Disease Prediction from MRI image using Convolutional Deep Neural Network

Alzheimer’s disease is one sort of neurological disorder which affects the daily activities and it is incurable. But, if it can be detected early, then it may be prevented or made to progress slowly. For detection, deep learning has played a great role now a days. In this paper, we have taken 6400 MRI images of both dementia and non-dementia subjects. The dataset is spilt into ${8 0 \%}$ for training a model and ${2 0 \%}$ for validating and accessing the achievement of the trained model. We have recommended a two-dimensional convolutional neural network architecture in order to apply on the MRI dataset to generate a model for classifying and predicting Alzheimer’s disease patients. The model generated using the proposed two-dimensional architecture has given approximately $98 \%$ classification accuracy. Besides, we have presented roc-auc curve, f1-score, recall, support, precision and confusion matrix to show the execution of the model. Again, at training procedure, model loss and accuracy also depicted according to epoch times. The researches already performed on prediction of Alzheimer’s disease has given poor performance of ${9 1 \%}$ classification accuracy in comparison to our model.