Classification of Alzheimer disease using DenseNet-201 based on deep transfer learning technique

Area of Science:

• Neuroimaging
• Artificial Intelligence
• Medical Diagnostics

Background:

• Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized by memory loss.
• Early detection of AD is crucial due to the lack of definitive treatments or cures.
• Magnetic Resonance Imaging (MRI) is a key neuroimaging modality for AD diagnosis.

Purpose of the Study:

• To develop and evaluate a deep learning model for classifying different stages of Alzheimer's disease.
• To leverage transfer learning with the DenseNet-201 architecture for AD diagnosis using MRI scans.
• To improve the accuracy and reliability of AD detection through advanced computational methods.

Main Methods:

• Utilized a DenseNet-201 based transfer learning approach for AD classification.
• Employed a dataset of MRI scans categorized into five classes: Non-Demented (ND), Moderate Demented (MOD), Mild Demented (MD), Very Mild Demented (VMD), and Severe Demented (SD).
• Implemented data augmentation techniques to enhance dataset size and model performance.

Main Results:

• The proposed deep learning model achieved a high classification accuracy of 98.24%.
• The model demonstrated superior performance compared to existing and state-of-the-art methods in AD diagnosis.
• The study confirmed the model's practicality and reliability for identifying various Alzheimer's disease stages.

Conclusions:

• The developed DenseNet-201 model offers a highly accurate and effective tool for Alzheimer's disease diagnosis using MRI.
• This deep learning approach shows significant promise for early and reliable detection of AD stages.
• The findings support the integration of advanced AI techniques in clinical neuroimaging for neurodegenerative disease management.