Towards a Videobronchoscopy Localization System from Airway Centre Tracking Carles S´ anchez 1 , Antonio Esteban-Lansaque 1 , Agn` es Borr´ as 1 , Marta Diez-Ferrer 2 , Antoni Rosell 2 and Debora Gil 1 1 Computer Vision Center, Computer Science Department, UAB, Bellaterra, Spain 2 Pneumology Unit, Hospital Universitari Bellvitge, IDIBELL, CIBERES, Barcelona, Spain Keywords: Video-bronchoscopy, Lung Cancer Diagnosis, Airway Lumen Detection, Region Tracking, Guided Bron- choscopy Navigation. Abstract: Bronchoscopists use fluoroscopy to guide flexible bronchoscopy to the lesion to be biopsied without any kind of incision. Being fluoroscopy an imaging technique based on X-rays, the risk of developmental problems and cancer is increased in those subjects exposed to its application, so minimizing radiation is crucial. Alternative guiding systems such as electromagnetic navigation require specific equipment, increase the cost of the clinical procedure and still require fluoroscopy. In this paper we propose an image based guiding system based on the extraction of airway centres from intra-operative videos. Such anatomical landmarks are matched to the airway centreline extracted from a pre-planned CT to indicate the best path to the nodule. We present a feasibility study of our navigation system using simulated bronchoscopic videos and a multi-expert validation of landmarks extraction in 3 intra-operative ultrathin explorations. 1 INTRODUCTION Lung cancer is one of the most incidental neoplasms in Europe with around 437, 773 new cases estimated for 2015 1 . Screening programs increase lung can- cer early diagnosis and, thus, may significantly reduce the risk of death (Aberle et al., 2011). Screening pro- grams are based on the detection of small pulmonary lesions with low dose chest computed tomography (CT) and its pathological confirmation. Pathologi- cal confirmation requires either transthoracic needle aspiration or an endoscopic examination. Transtho- racic needle aspiration has some complications (Man- hire et al., 2003), like pneumothorax (20%). Conven- tional bronchoscopic diagnostic procedures are visu- ally guided using radiating fluoroscopy and render a suboptimal 34% of positive results for lesions < 2 cm (Donnelly and Edwin, 2012). New endoscopy tech- niques (like virtual bronchoscopy or electromagnetic techniques) are expensive, require either manual in- tervention or special gadgets, only increase diagnos- tic yield to 70% (Aberle et al., 2011), and still radiate the patient. The 30% undiagnosed pulmonary lesions need CT follow-up or futile surgery procedures such 1 globocan.iarc.fr as thoracoscopies (Aberle et al., 2011). Diagnostic yield could be improved reducing radiation and costs if imaging technology could better detect and guide the bronchoscopist to the target lesion. A main limitation of flexible bronchoscopy is the difficulty to determine the correct pathway to periph- erial lesions. CT Virtual Bronchoscopy (VB) can pre- cede bronchoscopy to assess the optimal path using virtual navigation. VB is a computer simulation of the video bronchoscope image from the bronchoscope camera (Reynisson et al., 2014) which is created from the 3D CT volume, with the same view angle and zoom settings. Despite different approaches to VB, existing systems fail due to several aspects at plan- ning and exploration time. During exploration, VBs should accurately guide the operator across the planned path to the biopsy point. Standard protocols relying on fluoroscopy have a diagnostic yield around 40%, last 20 min per intervention and require 5-10 min of repetitive patient and medical staff radiation (Asano et al., 2013). Existing alternatives like image based systems (LungPoint, NAVI) or electromagnetic navigation (in- ReachTM,SPinDriveA) are far from meeting clini- cian expectations. Image systems are based on multi- modal registration of CT virtual projections to bron- 352 Sà ˛ anchez C., Esteban Lansaque A., BorrÃ˘ as A., Diez-Ferrer M., Rosell A. and Gil D. Towards a Videobronchoscopy Localization System from Airway Centre Tracking. DOI: 10.5220/0006115803520359 In Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2017), pages 352-359 ISBN: 978-989-758-225-7 Copyright c  2017 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved