Alzheimer's Disease Prediction Using Fisher Mantis Optimization and Hybrid Deep Learning Models.

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Neuroscience

Background:

• Alzheimer's disease (AD) is a progressive neurodegenerative disorder impacting memory and cognition.
• Early and accurate diagnosis of AD is crucial for effective treatment and management.
• Magnetic Resonance Imaging (MRI) is a key modality for AD assessment.

Purpose of the Study:

• To develop and validate a novel hybrid deep learning framework for improved Alzheimer's disease diagnosis using MRI data.
• To enhance the accuracy and efficiency of early AD detection.
• To explore the potential clinical applicability of advanced AI models in neurodegenerative disease diagnosis.

Main Methods:

• A hybrid deep learning framework combining Gray-Level Co-occurrence Matrix (GLCM) for texture features and VGG16 for spatial features was proposed.
• Fisher Mantis Optimization (FMO) was utilized for optimal feature selection.
• A Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) model classified selected features, capturing spatio-temporal patterns in MRI data.

Main Results:

• The proposed GLCM + VGG16 + FMO + CNN-LSTM model achieved high diagnostic performance: 98.63% accuracy, 98.69% sensitivity, 98.66% precision, and 98.67% F1-score.
• The framework significantly outperformed existing methods like CNN + SVM and 3D-CNN + BiLSTM.
• Comparative analyses demonstrated the superiority of FMO over other optimization algorithms and confirmed the model's robustness.

Conclusions:

• The GLCM + VGG16 + FMO + CNN-LSTM model demonstrates high efficacy and stability for accurate and early Alzheimer's disease diagnosis.
• The findings support the potential of this advanced deep learning approach for clinical application in AD detection.
• This study highlights the power of hybrid AI models in analyzing complex medical imaging data for neurological disorders.