Deep Learning-Based Alzheimer’s Disease Detection Using 2D MRI Images

About 50 million people in this world are affected by Alzheimer’s disease. Alzheimer’s disease is a type of dementia. It is a type of neurological disorder that causes changes in peoples’ memory, thinking ability, and behavior. The main reason for this is the destruction of human brain cells. Symptoms of this disease first appear in people of 60 years of age. People age 65 and older are at greatest risk. It is a progressive disease and where dementia symptoms gradually worsen. That is why it is very important to treat this disease if it is affected. Over the years deep learning models have shown significant results in medical image analysis including classification. This study presents a prediction model using MRI images of the brain, a deep learning model for the AD classification, specifically EfficientNetB0, and EfficientNetV2B. For classification, a dataset consisting of 2D MRI images of patients’ brains with varying stages of dementia, including nondemented individuals and those with very mild, mild, and moderate dementia were used. Finally, the performance of the models were measured using metrics such as accuracy, loss, and Scott’s Pi coefficient and were compared. The proposed model is able to accurately classify AD stages from brain MRI images. This model highlights deep learning models are as valuable tools for AD diagnosis and classification, paving the way for improved disease and management strategies.