Structural MRI Classification In Alzheimer's Disease Using CNN-XGboost: An Early Diagnosis

The sixth most common cause of mortality worldwide is Alzheimer's disease (AD), a progressive neurological condition.Large healthcare data has been a topic of interest for the past ten years due to the digitization that has led to a rise in data captured in medical industries.Lately, various categorization problems have been addressed with deep learning and machine learning techniques, which have demonstrated notable gains in effectiveness.On the other hand, their limited generalization abilities and inadequate model variance provide a problem, as reported in the literature.By using a mixed model that combines machine learning methods and deep learning, this project aims to increase the accuracy of Alzheimer's image classification.Particularly in the area of computer vision, the deep learning approach to machine learning has outperformed traditional machine learning in terms of its capacity to identify complex structures in complex, high-dimensional data.To improve the detection ability to classify images of Alzheimer's disease, this work has presented a hybrid model using CNN a Deep Learning technology, and the XGBoost machine learning.Based on structural MRI images, our findings imply that AD can be accurately classified using CNN-XGBoost algorithms.This may result in the creation of tools for AD diagnosis and aid in enhancing the illness's early identification and management.Using a structural MRI image hold-out test set, we assessed our model's performance.In comparison to other techniques used to classify AD from MRI scans, our model's accuracy of 98.7% is much greater.