Advances in Ultrasound-Guided Interventions

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.

Sources

DiffUS: Differentiable Ultrasound Rendering from Volumetric Imaging

Vibration-Based Energy Metric for Restoring Needle Alignment in Autonomous Robotic Ultrasound

PADReg: Physics-Aware Deformable Registration Guided by Contact Force for Ultrasound Sequences

Shape Completion and Real-Time Visualization in Robotic Ultrasound Spine Acquisitions

The Brain Resection Multimodal Image Registration (ReMIND2Reg) 2025 Challenge

Built with on top of