Home /Research /The Additional Accuracy Gained by Cone Beam CT in Shape-Sensing Robotic Bronchoscopy
OTHER

The Additional Accuracy Gained by Cone Beam CT in Shape-Sensing Robotic Bronchoscopy

Alberto Revelo, Jing Peng, Jianing Ma, Michael Woods, Christian Ghattas, Jasleen Pannu, Jeffrey C. Horowitz, Nicholas J. Pastis

Year
2025
Citations
1

Abstract

Cone beam CT (CBCT) scan provides an intraprocedural 3-dimensional image of the biopsy tool in relation to the lesion, which may enhance the accuracy of robotic-assisted bronchoscopy (RAB) through real-time bronchoscopy adjustments. Our goal is to evaluate the ability to accurately position, in real time, a biopsy needle in the center of a pulmonary nodule using shape-sensing RAB alone vs shape-sensing RAB + CBCT guidance. Does the addition of CBCT improve the accuracy of robotic bronchoscopy? A total of 102 nodules were biopsied using shape-sensing RAB and the position of the needle in relation to the center of the nodule identified using a ceiling-mounted CBCT scanner. Repositioning of the RAB after 1 or 2 CBCT adjustments was accomplished using information gathered from the 3-dimensional images and using an updated augmented fluoroscopy target. The primary end point was needle location and distance change of the needle tip in reference to the lesion center using RAB alone vs RAB + CBCT scan. Secondary end points were improvement in radial endobronchial ultrasound image and average number of CBCT spins required to land at the center of the target. Using RAB alone, the needle was placed in the center in 27 nodules (26.5%). The addition of CBCT scan to RAB greatly improved the distance of the needle toward the center of the lesion (mean ± SD, −4.08 ± 4.63 mm) in 46 nodules (61.3%) after 1 CBCT adjustment and (mean ± SD, −4.02 ± 4.21 mm) in an additional 17 nodules (58.6%) after a second CBCT adjustment ( P < .001). Radial endobronchial ultrasound image was also improved from eccentric to concentric. The addition of CBCT scan to RAB was shown to improve the position of a biopsy tool in relation to the center of a pulmonary nodule. To our knowledge, this study is the first to quantify the accuracy achieved using both technologies, which may translate into reliability to perform diagnostic and potentially future therapeutic interventions in guided bronchoscopy.

Keywords

BronchoscopyComputer visionFlexible bronchoscopyComputer scienceArtificial intelligenceMedicineRadiology

Related papers

Browse all OTHER papers