Alzheimer's disease, the most prevalent neurodegenerative illness, exhibits initially mild symptoms that worsen over time. It is a common form of dementia, and its lack of a cure poses treatment challenges. Diagnosis often occurs at later stages, making early prediction crucial for potentially slowing down the disease's progression. This study employs a machine learning algorithm, specifically a Convolutional Neural Network (CNN), to forecast the onset of Alzheimer's disease. The algorithm utilizes psychological indicators such as age, visits, MMSE scores, and educational level, along with MRI images, to make predictions. Detecting Alzheimer's at an early stage is vital to prevent severe brain damage, especially among individuals aged 65 and above, for whom the disease can become dangerous and even fatal. The primary objective of this research is to build a useful model for Alzheimer's prediction using machine learning techniques, including CNN and feature extraction/selection. Alzheimer's disease is now the sixth leading cause of mortality in the US, and projections suggest it may rank third among seniors, following heart disease and cancer. Early diagnosis and treatment are critical for forecasting the disease's progression and halting its advancement. However, the diagnosis of Alzheimer's relies on various medical tests and extensive multivariate heterogeneous data, making manual comparison, visualization, and interpretation of the data laborious and complex.