This advanced course provides a rigorous technical foundation in Geospatial Artificial Intelligence (GeoAI), focusing on deep learning applications to complex public health challenges. Students learn to design, train, and validate deep learning models for spatial and spatiotemporal data. Topics include Convolutional Neural Networks (CNNs) for extracting environmental health determinants from high-resolution satellite imagery, and Recurrent Neural Networks (RNNs) and Long Short-Term Memory (LSTM) models for forecasting disease outbreaks and epidemiological trends. Through hands-on Python projects using PyTorch or TensorFlow, students will address model interpretability, overfitting, and data sparsity, and build end-to-end AI pipelines for predictive analytics and early warning systems.