• Rama Lakshmi Boyapati
• PAVAN GUNDA

Alzheimer's disease (AD) is a long-term, irreversible brain illness for which there is now no known treatment. Nonetheless, current medications may halt its advancement. For this reason, stopping and managing the progression of AD depends greatly on early detection. The primary goal is to create a comprehensive framework for medical picture classification according to different phases of Alzheimer's disease and early identification of the condition. This work employs a deep learning approach, more precisely convolutional neural networks (CNN). The AD spectrum has four multi-classified stages. Moreover, distinct binary classifications of medical images are applied for every pair of AD stages. To identify AD and categorize the medical photos, two techniques are employed. The first approach makes use of straightforward CNN architectures based on 2D and 3D convolution to handle 2D and 3D structural brain scans from the Alzheimer's disease Neuroimaging Initiative (ADNI) dataset. The second approach makes use of pre-trained models, like the VGG19 model, for medical image classifications by applying the principle of transfer learning. We show that deep models perform better than shallow models, such as support vector machines, decision trees, random forests, and k-nearest neighbours, using the Alzheimer's disease neuroimaging initiative (ADNI) dataset. We further show that in terms of accuracy, precision, recall, and meanF1 scores, incorporating multi-modality data performs better than single modality models.

Medical image classification, Alzheimer’s disease , Convolutional neural network (CNN), Machine learning, Brain MRI.

"MULTIMODAL DEEP LEARNING METHOD FOR ALZHEIMER'S DISEASE SEVERITY IDENTIFICATION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 12, page no.e427-e434, December-2023, Available :http://www.jetir.org/papers/JETIR2312453.pdf

"MULTIMODAL DEEP LEARNING METHOD FOR ALZHEIMER'S DISEASE SEVERITY IDENTIFICATION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 12, page no. ppe427-e434, December-2023, Available at : http://www.jetir.org/papers/JETIR2312453.pdf