AI-Driven Analysis of Biomarkers Using Machine Learning for Alzheimer's Research in Drosophila Melanogaster

Drosophila melanogaster, with its genetic similarity to humans, is a strong model organism for studying Alzheimer's disease. However, traditional research methods for exploring treatments and behavioral indicators of Alzheimer's in Drosophila have mainly relied on manual techniques. Leveraging Artificial Intelligence (AI) technology provides significant advantages in reliably analyzing biological behavior and processes. This paper develops a novel AI-driven analysis method using automatic video monitoring to investigate behavioral biomarkers associated with Alzheimer's in Drosophila and to implement machine learning models for predicting the presence of Alzheimer's. Five experimental groups of Drosophila-healthy, Alzheimer's, and Alzheimer's treated with 2 g, 4 g, and 6 g of Aloe vera-were used for AI analysis. Drosophila movement and behavior were tracked to analyze four biomarkers-negative geotaxis, velocity, directional heading changes, and grooming magnitude. Five behavioral parameters as features for machine learning were identified: speed, x-distance, y-distance, acceleration, and turning angle. Three machine learning models were developed to predict whether a Drosophila has Alzheimer's: two classical models (CatBoost and XGBoost) and a deep learning model (CNN-GRU). The three models achieved accuracies of 0.8654, 0.8638, and 0.8454 respectively. In addition, the AI-driven method was used to evaluate the effect of Aloe vera as a natural treatment for Alzheimer's model of Drosophila. Results indicated that Aloe as a natural treatment improved mobility and stability in the Alzheimer's model of Drosophila. The proposed AI-driven system for uncovering behavioral biomarkers in Drosophila with Alzheimer's offers a reliable and more effective alternative to traditional manual methods. Evaluating the potential of Aloe vera to be a therapeutic agent opens new doors to discovering new natural treatments for neurodegenerative disorders.