The field of medical imaging and surgery is witnessing significant advancements with the integration of deep learning and physics-based models. Researchers are exploring innovative approaches to improve the accuracy and efficiency of surgical procedures, such as smoke removal in laparoscopic surgery and bronchoscopy navigation. The development of versatile frameworks for real-time 3D super-resolution and ultra-high-definition image dehazing is also gaining traction. Furthermore, updates to existing libraries, such as PYRO-NN, are enhancing the compatibility and efficiency of differentiable CT operators. Noteworthy papers include: SurgiATM, which proposes a physics-guided plug-and-play model for deep learning-based smoke removal, and 4KDehazeFlow, which introduces a novel method for ultra-high-definition image dehazing via flow matching. BronchOpt is also notable for its vision-based pose optimization framework for accurate bronchoscopy navigation.