Four-Way Classification of Alzheimer's Disease using Deep Siamese Convolutional Neural Network with Triplet-Loss Function

Abstract

Alzheimer's disease (AD) is a neurodegenerative disease that causes irreversible damage to several brain regions, including the hippocampus causing impairment in cognition, function, and behaviour. Early diagnosis of the disease will reduce the suffering of the patients and their family members. Towards this aim, in this paper, we propose a Siamese Convolutional Neural Network (CNN) architecture that employs the triplet-loss function for the representation of input MRI images as k-dimensional embeddings. We used both pre-trained and non-pre-trained CNNs to transform images into the embedding space. These embeddings are subsequently used for the 4-way classification of Alzheimer's Disease. The model efficacy was tested using the ADNI and OASIS datasets which produced an accuracy of 91.83% and 93.85%, respectively. Furthermore, obtained results are compared with similar methods proposed in the literature.

Data Complexity-Based Evaluation Using Brain Magnetic Resonance Images to Determine Alzheimer’s Disease

“…https://www.indjst.org/ • VGG16: The Siamese CNN classifier in the research was implemented using the VGG16 architecture, a widely recognized and preeminent deep learning model known for its capability to filter intricate features from images. With its deep layers and complex architecture, VGG16 has shown better performance in different image categorization tasks, making it an ideal choice for the model's classification of AD cases depending on MRI images (22) . • DCNN Classifier: The DCNN classifier implemented in this study is instrumental in its capability to extract intricate features from medical imaging data.…”

“…https://www.indjst.org/ Siamese CNN (VGG16) (22) achieved an accuracy of 85.31% with high recall and precision, making it a suitable model for AD diagnosis. (23) 3D-DCNN (DCNN) demonstrated impressive performance with an accuracy of 97.53%, suggesting its capability for AD staging.…”

“…In the realm of Alzheimer's Disease (AD) identification, recent studies have explored various deep-learning techniques. Notably, Siamese CNN models have demonstrated promising results, achieving high accuracies of 91.83% and 93.85% (22) . Additionally, the integration of DCNN and 3D-DCNN systems has shown superior performance in AD staging, surpassing pre-trained models (23) .…”