Prediction of Alzheimer's Disease from Magnetic Resonance Imaging using a Convolutional Neural Network

Abstract Objectives: The primary goal of this study is to examine if a convolutional neural network (CNN) can be applied as a diagnostic tool predicting Alzheimer’s Disease (AD) from magnetic resonance imaging (MRI) using the MIRIAD-dataset (Minimal Interval Resonance Imaging in Alzheimer's Disease).Methods: The MIRIAD-dataset contains patients represented by a set of MRI scans of the brain and further diagnostic data. Hyperparameter and configurations of CNNs were optimized to determine the best-performing model. The CNN was implemented in Python with the deep learning library ‘Keras’ using Linux/Ubuntu as the operating system.Results: This study obtained the following best performance metrics on predicting Alzheimer’s Disease from MRI: Matthew's Correlation Coefficient (MCC) of 0.77; accuracy of 0.89; F1-score of 0.89; AUC of 0.92. The computational time for training of a CNN takes less than 30 seconds with a GPU (graphics processing unit).Conclusions: The study suggests that an axial MRI scan can be used to diagnose if a patient has Alzheimer's Disease with a performance of 0.92 AUC.