SESSION TITLE: Interventional Pulmonology Investigations SESSION TYPE: Original Investigations PRESENTED ON: 10/10/2018 03:45 PM - 04:45 PM PURPOSE: To develop and characterize a novel navigation approach for endobronchial procedures based on Cone Beam Computed Tomography (CBCT) and Augmented Fluoroscopy from bench to bedside. METHODS: An endobronchial navigation software prototype was developed that can generate a 3D airway segmentation, tumor segmentation and 3D airway navigation path planning based on CBCT (XperCT, Philips), without electromagnetic (EM) tracking. All models (airways, targets, pathways) can be superimposed on X-ray to generate an augmented fluoroscopy view with real-time updates based upon table motion, magnification, or C-arm rotation. The navigation prototype was characterized and optimized using non-bronchoscopy devices (endovascular guidewires, catheters and steerable sheaths) in bench models (a 3D printed airway and an ex vivo breathable swine lung) and in an animal model (swine, n=3, approved by Animal Care and Use Committee). Two operators were asked to navigate to predefined lung segments and locations were confirmed with CBCT and 2D X-ray at two orthogonal angles. 9 complex clinical transbronchial biopsies of parenchymal lung lesions were retrospective reported (under an IRB-approved clinical protocol) using combinations of fluoroscopy, augmented fluoroscopy, CBCT, CBCT derived 3D models, EM navigation, and finally off-label miniature hardware without direct optical endoscopy. Clinical cases used previously FDA cleared software (OncoSuite, Philips), off-label for lung applications, whereas pre-clinical studies utilized a prototype software specific for lung. RESULTS: In the bench model, 10/10 (100%) navigation tasks were successful. In the animal model, navigation success was 21/24 targets (87%). In the clinical case series, CBCT and augmented fluoroscopy enabled correct guidance of the bronchoscope to peripheral and central lung lesions for biopsy. 8/9 (88.9%) patients, had accurate on-target navigation confirmed via CBCT and/or by histopathological or microbiological results. Pulmonologists found that the lesion segmentation was helpful in 9 of 9 cases (100%), including in 3 cases with fluoroscopic invisible/obscured lesions. The 9th case that failed with EM tracking with conventional guidance and CBCT guidance was successfully biopsied (confirmed with CBCT) using an off-label endovascular steerable sheath and a standard bronchoscopy aspirating needle without bronchoscopic navigation. CONCLUSIONS: CBCT derived 3D road mapping navigation for bronchoscopic procedures using augmented fluoroscopy is feasible and may be used with or without optical bronchoscopic tools. The impact upon workflow, outcomes, and cost are undefined as are the potential roles for these tools in clinical care. CLINICAL IMPLICATIONS: Real time derived navigation tools, such as CBCT image guidance and augmented fluoroscopy, may complement bronchial navigation for small and difficult to access peripheral lung nodules. DISCLOSURES: PI on grant to UCI relationship with Philips Healthcare Please note: $20001 - $100000 Added 06/03/2018 by Nadine Abi-Jaoudeh, source=Web Response, value=money to UCI Pi on grant to UCI relationship with Teclison Cheery Pharma Please note: >$100000 Added 06/03/2018 by Nadine Abi-Jaoudeh, source=Web Response, value=grant to UCI PI on grant to UCI relationship with Sillajen Please note: $5001 - $20000 Added 06/03/2018 by Nadine Abi-Jaoudeh, source=Web Response, value=grant money to UCI no disclosure on file for Robert Browning; No relevant relationships by Quirina de Ruiter, source=Web Response no disclosure on file for Joseph Fontana; No relevant relationships by john karanian, source=Web Response CRADA between NIH and Philips relationship with Philips Healthcare Please note: >$100000 Added 06/11/2018 by William Pritchard, source=Web Response, value=Grant/Research Support No relevant relationships by Alessandro Radaelli, source=Web Response No relevant relationships by Stephanie Schampaert, source=Web Response Employee relationship with Philips Please note: >$100000 Added 06/04/2018 by Imramsjah van der Bom, source=Web Response, value=Salary no disclosure on file for Bradford Wood; no disclosure on file for Sheng Xu