Multi-View Separable Pyramid Network for AD Prediction at MCI Stage by ¹⁸F-FDG Brain PET Imaging.

Area of Science:

• Neuroimaging
• Artificial Intelligence
• Gerontology

Background:

• Alzheimer's Disease (AD) is a leading cause of death in the elderly, with Mild Cognitive Impairment (MCI) as a prodromal stage.
• Identifying MCI patients who will progress to AD is crucial for timely intervention.

Purpose of the Study:

• To develop a deep learning model for early identification of AD progression in MCI patients.
• To leverage 18F-FDG PET neuroimaging for enhanced diagnostic accuracy.

Main Methods:

• A Multi-view Separable Pyramid Network (MiSePyNet) was designed using 18F-FDG PET scans.
• The network learns representations from axial, coronal, and sagittal views, employing separable convolutions for efficiency.
• The model was trained and evaluated on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset.

Main Results:

• MiSePyNet achieved a classification accuracy of 83.05% in predicting MCI progression to AD.
• The proposed method outperformed traditional and existing deep learning algorithms.

Conclusions:

• The MiSePyNet model demonstrates significant potential for early AD diagnosis using 18F-FDG PET.
• This deep learning approach offers a promising tool for identifying individuals at high risk of developing Alzheimer's Disease.