C-arm positioning for standard projections during spinal implant placement

Area of Science:

• Medical Imaging
• Surgical Technology
• Orthopedic Surgery

Background:

• Fluoroscopy-guided surgeries require precise C-arm positioning for optimal results.
• Manual C-arm adjustments lead to increased radiation exposure and longer procedure times.

Purpose of the Study:

• To develop an automated C-arm repositioning method using X-ray images.
• To eliminate the need for patient-specific CT scans and additional equipment.
• To improve accuracy and efficiency in fluoroscopy-guided orthopedic procedures.

Main Methods:

• Training a deep learning model on digitally reconstructed radiographs (DRRs) with simulated surgical elements.
• Incorporating clinical decision-making steps like ROI localization and landmark detection.
• Validating the method on a cadaver study simulating real surgical scenarios.

Main Results:

• Achieved superior C-arm positioning accuracy with an average improvement of 8.8°±4.2°.
• Demonstrated robustness and generalization capabilities compared to existing methods.
• Successfully transferred simulated training data to real-world X-ray images.

Conclusions:

• The proposed automated method significantly enhances C-arm positioning accuracy and efficiency.
• This approach offers a radiation-sparing and time-effective alternative for orthopedic surgeries.
• The model's design, mimicking surgeon decision-making, improves interpretability and clinical relevance.