Aizheimer's disease prediction analysis using Machine Learning and Deep Learning methods

Abstract

Introduction: Alzheimer’s Disease (AD) is a progressive neurodegenerative disorder that affects memory, cognition and daily functioning severely. It is critical to make early and accurate prediction in order to be effective in management and intervention. In this research, the use of ML and DL techniques for AD prediction using clinical and neuroimaging data is investigated. And then, I implemented and tested four algorithms, which are: “Support Vector Machine (SVM), Random Forest (RF), Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM).” The dataset is MRI scans with preprocessed and cognitive assessment scores used. The results of experiments show that DL is superior to traditional ML methods. The best results for CNN are 93.6, for LSTM 92.4, that outperformed SVM (88.3) and RF (89.7). CNN model’s other two performances of sensitivity (94:1%) and precision (92.8%) were also better. Due to its improved performance and robustness, the proposed models are confirmed by a comparative analysis with existing studies. The most apparent point from this study is that increasing prediction accuracy requires feature extraction, data preprocessing, and model selection. The study supports the use of AI driven diagnostic tools in clinical practice and equips doctors with the means to start diagnosing Alzheimer’s Disease early and begin treatment. Future work will extend to use of more data sources, longitudinal data, and more interpretable model to aid broader clinical adoption