The field of federated learning and medical image analysis is rapidly advancing, with a focus on developing innovative methods to improve model performance and address challenges related to data heterogeneity and privacy. Researchers are exploring new approaches to federated learning, such as layer skipping and personalized learning, to reduce communication costs and improve model accuracy. Additionally, there is a growing interest in using self-supervised learning methods to enhance fine-grained anatomical discrimination in radiographic images. These advances have the potential to significantly improve the accuracy and efficiency of medical image analysis, leading to better patient outcomes. Notable papers in this area include: Federated Learning with Layer Skipping, which proposes a novel approach to reduce communication costs in federated learning. AFiRe: Anatomy-Driven Self-Supervised Learning, which introduces a new framework for enhancing fine-grained representation in radiographic images. X2BR: High-Fidelity 3D Bone Reconstruction, which presents a hybrid neural implicit framework for accurate 3D bone reconstruction from a single planar X-ray image.