AI-Based Model for Detection and Classification of Alzheimer Disease

Millions of people worldwide suffer from by the deadly brain disorder known as Alzheimer's Disease (AD). Alzheimer's disease is an incurable neurological disorder that affects brain function. To improve patient quality of life and create efficient treatment options, early AD diagnosis is essential. In this research, we have proposed a novel method for AD detection using Convolutional Neural Networks (CNNs) using Magnetic Resonance Imaging (MRI) datasets. We use a pre-trained CNN model to extract features from brain MRI images and classify them into Alzheimer's or non-Alzheimer's groups. We compare the efficiency of our model with several classifiers, such as: SoftMax, Support Vector Machine (SVM), DNN, and some other classifiers. We evaluated our model on an MRI dataset. We achieve high accuracy on the MRI dataset: 99.56% on the training set and 99% on the validation set. The loss values of training and validation are 0.0146% and 0.153% respectively. Our method is fast, robust, and generalizable. We point out that CNN-based models can be a powerful tool for AD detection using MRI data.