To develop a system for real-time 3D visualization of spinal motion using robotic ultrasound and preoperative CBCT to evaluate spinal instability, emphasizing the critical need to minimize radiation exposure for patient safety.
Key Findings:
The integration of robotic ultrasound with a preoperative CBCT model allows for accurate 3D visualization of spinal motion, enhancing diagnostic capabilities.
This method significantly reduces radiation exposure compared to traditional dynamic X-ray imaging, addressing a critical concern in spinal assessments.
The kinematic model enables continuous visualization of spinal dynamics during bending, providing richer data for clinical evaluation.
Interpretation:
The proposed system offers a non-ionizing, real-time alternative for assessing spinal instability, enhancing diagnostic capabilities while ensuring patient safety through reduced radiation exposure, with potential implications for clinical practice.
Limitations:
Ultrasound imaging is limited to superficial structures and may not visualize the entire vertebral body, necessitating complementary imaging techniques.
Shape completion from partial observations can lead to reconstruction errors affecting diagnostic accuracy, highlighting the need for improved algorithms.
Conclusion:
The developed framework provides a promising approach for evaluating spinal instability, combining the strengths of robotic ultrasound and CBCT to achieve real-time, motion-aware visualization, while prioritizing patient safety and diagnostic accuracy.
Radiologists assigned to receive step-by-step explanations from a large language model achieved higher diagnostic accuracy in a randomized vignette study, while differential-diagnosis outputs may have increased inappropriate reliance on incorrect model suggestions.
