ARTIFICIAL INTELLIGENCE’S SIX MODELS CROSS-VALIDATION ON ALZHEIMER PATTERNS RECOGNITION USING ARTIFICIAL INTELLIGENCE

The early detection of Alzheimer’s disease remains a critical challenge in medical research due to the complexity of its symptoms and the late-stage diagnosis commonly observed in clinical settings. This paper explores artificial intelligence models cross-validation, specifically a deep learning model based on the Faster-RCNN-ResNet-Coco 101 architecture versus others not using deep learning: decision Tree, random forest, support vector machine, XGBoost, and voting classifier to facilitate early diagnosis. By leveraging data from the ADNI open database, our study implements a comprehensive training process within a Python Conda environment, demonstrating promising classification accuracy and validation results. While various recognition systems exist, our approach provides an expert, ad hoc solution tailored to the early identification of Alzheimer’s patterns through tomography analysis based on Faster-RCNN-ResNet-Coco 101 architecture as the best choice in the tested models.