Grant: Travel Awards - Institute Community Support

Healthcare is a critical field that attracts significant attention due to its importance and rapid advancements. Around the world, health institutes, hospitals, and similar organizations generate vast, diverse, and rich datasets continuously. This huge amount of data has the potential for disease diagnostics, healthcare quality improvement, and medication optimization. In recent decades, deep learning has demonstrated remarkable results in the healthcare domain. However, traditional deep learning methods typically require data to be centralized from various institutions on a powerful server for training deep neural networks. This centralization imposes high communication costs and raises significant privacy and security concerns for users. Federated Learning (FL) is a decentralized machine learning approach that collaboratively trains a global model without requiring users' data to be transferred to an external server. By keeping data local, FL significantly enhances privacy compared to traditional centralized methods, making it highly suitable for privacy-critical fields like healthcare. This year at the AI4H Summer Institute, we aim to explore and advance the applications of Federated Learning on electronic health data (EHD), focusing on its potential to transform healthcare while preserving privacy.