An Enhanced Deep Convolution Neural Network Model to Diagnose Alzheimer's Disease Using Brain Magnetic Resonance Imaging

Milon Biswas, Md. Kawsher Mahbub, Md. Abdul Mozid Miah

Abstract: Alzheimer’s disease (AD) is a brain ailment that is irreversible and has an initial warning sign, such as memory cognitive functioning loss. The precise and early diagnosis of AD is exceedingly vital for patient care. The study proposed a deep convolutional neural network (CNN) model for diagnosing AD state using brain magnetic resonance imaging (MRI). The authors’ concentrated on a binary classification decision for brain MRI and observed better results compared to the other state-of-the-art studies, with an accuracy of 0.9938, sensitivity 0.9890, specificity 0.9974, precision 0.9970 and F1 score of 0.9932. The experiment was conducted on a 4800 image dataset (Kaggle) using Google collaboratory GPU, Keras library with TensorFlow backend.

External IDs:dblp:conf/rtip2r/BiswasMM21