Abstract
Virtual automatic bronchoscopic navigation (VBN) systems to determine the route to peripheral pulmonary lesions (PPLs) in lung cancer can improve diagnostic biopsy yields. However, compared with VBN, drawing manual routes using computed tomography images, especially with oblique methods, can identify more routes. The Ziostation2 VBN system combines the benefits of these 2 methods; we evaluated this performance by comparing 3 different route-determining methods.We retrospectively collected data from 50 patients with PPLs measuring <30 mm who underwent transbronchial biopsy with an ultrathin bronchoscope at the Osaka International Cancer Institute during January to December 2018. We compared automatic VBN (Ziostation2), manual route modification using an oblique method after automatic VBN, and manual navigation using a general application computed tomography viewer. Concordance between predicted and actual branching were determined. We also compared the predicted relationship between the terminal bronchi and the lesion by 2 of the methods with ultrasonographic images (radial-probe endobronchial ultrasonography [radial-EBUS]).Manual modification after automatic VBN significantly increased the rate of determining routes to the target (66%) versus with the automatic VBN alone (32%) (P < .001). Expected route bifurcations were exact matches with actual branching in 45/48 of the patients using manual modification after automatic VBN. The predicted relationship between the terminal bronchi and the lesion using manual modification after VBN matched the radial-EBUS images in 35/50 of the patients.Manual modification of routes to PPLs using an oblique method after automatic VBN predicted actual radial-EBUS route imaging and could help determine appropriate patients for bronchoscopy.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have