Artificial Intelligence Based on Modelling for Prediction of Alzheimer's Disease for Optimal Solution

Alzheimer's disease (AD) is an incurable neurodegenerative disorder that causes progressive brain deterioration. Beginning with mild symptoms, the disease gradually worsens over time, leading to brain tissue damage and eventual death, particularly affecting memory function. This paper introduces a multimodal approach in medical image processing, using machine learning for the classification and detection of AD. The procedure begins with image acquisition, where MRI images may suffer from noise and contrast issues. To address this, a Contrast Limited Adaptive Histogram Equalization algorithm is employed for pre-processing, enhancing image quality. For image segmentation, the k-means algorithm is used, leading to the identification of regions of interest. Feature extraction is accomplished through the Principal Component Analysis (PCA) algorithm. Finally, machine learning methods are used for the classification of processed images. The Support Vector Machine (SVM) with Radial Basis Function (RBF) classifier demonstrates the highest accuracy for diagnosing AD, followed by Artificial Neural Networks (ANN) and Iterative Dichotomiser 3 (ID3). The ANN algorithm exhibits higher sensitivity and F-score compared to other classifiers.