Hybridized Deep Learning Approach for Detecting Alzheimer's Disease.

Area of Science:

• Neuroscience
• Artificial Intelligence
• Medical Imaging

Background:

• Alzheimer's disease (AD) is a neurodegenerative disorder characterized by neuronal atrophy and amyloid deposition.
• Early detection of mild cognitive impairment (MCI), particularly early MCI (EMCI), is crucial for timely intervention but remains challenging.
• Existing machine learning algorithms often struggle to differentiate between cognitively normal individuals and those with EMCI.

Purpose of the Study:

• To propose a hybrid Deep Learning Approach for the early detection of Alzheimer's disease.
• To develop a method for accurately distinguishing between cognitively normal individuals and those with early mild cognitive impairment (EMCI).

Main Methods:

• A hybrid deep learning model combining Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) algorithms was developed.
• The model utilized multimodal data, including magnetic resonance imaging (MRI), positron emission tomography (PET), and neuropsychological test scores.
• Adam's optimization algorithm was employed to update learning weights and enhance classification accuracy.

Main Results:

• The proposed deep learning system achieved an accuracy of 98.5% in classifying cognitively normal controls from individuals with EMCI.
• The methodology demonstrated the potential of deep neural networks to automatically discover imaging biomarkers for AD detection.

Conclusions:

• Deep neural networks can effectively identify subtle imaging biomarkers indicative of early Alzheimer's disease.
• The hybrid deep learning approach shows significant promise for accurate and early diagnosis of AD, facilitating prompt treatment.