A respiratory correlated image guided surgery method: quantitative accuracy results in swine and human cadaver environments

Abstract

Intervention on small targets in the lung is difficult, leading our team to develop a highly accurate respiratory correlated image guided surgery (RCIGS) system. Simulated point source targets were implanted into ex vivo porcine and human cadaver lungs attached to a ventilator. The RCIGS system was utilized to guide intervention in the presence of respiratory motion using a commercially available electromagnetic tracking solution. After intervention, the lungs were imaged to determine the target registration error between the target and needle. The system tumor position modeling had sub-mm accuracy. The mean intervention error for 12 porcine targets was 3.8 mm. The mean target registration error on four targets in a human cadaver was 4.0 mm at a mean depth of 9 cm. The system provides an accuracy for intervention on targets of less than 1 cm in diameter at depths of up to 10 cm. A system of this accuracy outperforms current clinical standards.