An optimized hybrid deep learning model to detect Alzheimer disease.

Area of Science:

• Neuroscience
• Artificial Intelligence
• Medical Imaging

Background:

• Alzheimer's disease (AD) diagnosis requires early detection for effective treatment, but traditional methods struggle with identifying early stages like mild cognitive impairment (MCI).
• Limitations in feature extraction and classification hinder accurate early-stage AD detection.

Purpose of the Study:

• To develop and evaluate an optimized hybrid deep learning model for improved Alzheimer's disease detection.
• To enhance the accuracy and performance of early AD detection, particularly for MCI stages.

Main Methods:

• A hybrid deep learning model combining Inception v3 for feature extraction and ResNet 50 for classification was developed.
• Network parameters were optimized using the Adaptive Rider Optimization (ARO) algorithm.
• The model was evaluated on a benchmark dementia dataset.

Main Results:

• The proposed model achieved a high accuracy of 96.6%.
• The model demonstrated excellent performance with a precision of 98%, recall of 97%, and F1-score of 98%.
• Results surpassed those of existing state-of-the-art techniques.

Conclusions:

• The optimized hybrid deep learning model shows significant potential for accurate and early Alzheimer's disease detection.
• This approach offers a promising tool for improving the diagnosis of mild cognitive impairment (MCI).
• The model's superior performance suggests a valuable advancement in neurodegenerative disease diagnostics.