Abstract
Brain-based technologies for human augmentation face challenges in personalization and real-world translation. We present an AI-driven personalized Bayesian optimization algorithm that remotely adjusts neurostimulation parameters based on baseline ability and head anatomy to enhance sustained attention at home. Validated through in silico modeling and a double-blind, sham-controlled study, our approach aligns with MRI-based models and neurobiological theories, maximizing efficacy and enabling scalable, personalized cognitive enhancement and therapy in real-world settings.