SMART FARMING PREDICTION SYSTEM USING EXPLAINABLE AI

Abstract

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder that primarily affects memory, thinking, and behavior. Early detection of Alzheimer's is critical for managing symptoms and improving the quality of life for patients. In this project, we build a computer-based system to predict the early stages of Alzheimer’s disease.

The system combines different kinds of information, such as brain scans (MRI and PET), lab tests (like blood or spinal fluid markers), genetic risk factors, and memory test results.

After cleaning and organizing the data, we train both traditional machine learning models and modern deep learning methods to find patterns linked to early Alzheimer’s. We then test how well these models can tell apart healthy people, those with mild cognitive problems, and those already showing early Alzheimer’s. We will first train the models using the large ADNI database and then check their accuracy on another independent dataset to make sure the results are reliable. Along with accuracy, we focus on making the predictions explainable so doctors can understand which features are most important. This work aims to support earlier and more reliable detection of Alzheimer’s disease in real-world healthcare.