Federated learning for healthcare data analytics is a rapidly growing research area that enables collaborative model training across multiple medical institutions without sharing sensitive patient data, thereby preserving privacy and complying with regulatory requirements. This approach leverages decentralized learning frameworks to analyze electronic health records, medical imaging, genomics, and wearable sensor data for predictive modeling, disease diagnosis, treatment recommendation, and patient outcome prediction. Research explores advanced federated optimization techniques, personalization strategies for heterogeneous clients, privacy-preserving mechanisms such as differential privacy and secure multi-party computation, and integration with deep learning architectures including CNNs, RNNs, and transformer-based models. Applications span early disease detection, personalized treatment planning, epidemic monitoring, and clinical decision support systems. Recent studies also focus on handling non-i.i.d. data distributions, improving communication efficiency, robustness to adversarial attacks, and enabling cross-institutional transfer learning, establishing federated learning as a transformative paradigm for privacy-aware, scalable, and collaborative healthcare analytics.