The field of ultrasound-guided interventions is moving towards more accurate and robust methods for image guidance and registration. Researchers are exploring new approaches to address the challenges posed by noise, artifacts, and poor alignment between preoperative and intraoperative images. One notable trend is the use of physics-based models and machine learning techniques to improve the accuracy of ultrasound image rendering and registration. Another area of focus is the development of real-time visualization and shape completion methods to enhance spinal visualization and navigation. Noteworthy papers in this area include DiffUS, which presents a differentiable ultrasound renderer that synthesizes realistic B-mode images from volumetric imaging, and PADReg, which proposes a physics-aware deformable registration framework guided by contact force. These innovative approaches have the potential to significantly improve the accuracy and effectiveness of ultrasound-guided interventions.