PURPOSE: New surgical approaches based on natural orifice transluminal surgery (NOTES) have the potential to further decrease morbidity and hospital stay. However, a number of key challenges have been identified preventing its clinical adoption, including inadequate instrument design and spatial disorientation. Furthermore, retroflexion, missing fixed anatomical references, and limited field-of-view are key factors contributing to disorientation in NOTES. METHODS: A hybrid approach of integrated orientation sensing and real-time vision processing is proposed to restore orientation cues for improved surgical navigation. The distal tip of an articulated robotic endoscope is equipped with an inertial measurement unit (IMU) enabling video images to be reoriented and stabilized with respect to the horizon. This is performed by measuring the direction of gravity in relation to the cameras. Dynamic view expansion is used to increase the field-of-view of the endoscope. The method registers past video images to the current image and creates an enlarged visualization of the anatomy through simultaneous localization and mapping (SLAM). RESULTS: The clinical potential of the system is demonstrated on a NOTES appendectomy procedure performed on the NOSsE phantom. This involves an articulated robotic endoscope navigating to visualize the appendix while retroflexed. The horizon stabilization is additionally evaluated quantitatively against known ground truth. CONCLUSIONS: The combination of horizon stabilization and dynamic view expansion presents a realistic approach for reintroducing orientation and navigation cues during NOTES. The platform allows real-time implementation, which is an important prerequisite for further clinical evaluation.
PURPOSE: New surgical approaches based on natural orifice transluminal surgery (NOTES) have the potential to further decrease morbidity and hospital stay. However, a number of key challenges have been identified preventing its clinical adoption, including inadequate instrument design and spatial disorientation. Furthermore, retroflexion, missing fixed anatomical references, and limited field-of-view are key factors contributing to disorientation in NOTES. METHODS: A hybrid approach of integrated orientation sensing and real-time vision processing is proposed to restore orientation cues for improved surgical navigation. The distal tip of an articulated robotic endoscope is equipped with an inertial measurement unit (IMU) enabling video images to be reoriented and stabilized with respect to the horizon. This is performed by measuring the direction of gravity in relation to the cameras. Dynamic view expansion is used to increase the field-of-view of the endoscope. The method registers past video images to the current image and creates an enlarged visualization of the anatomy through simultaneous localization and mapping (SLAM). RESULTS: The clinical potential of the system is demonstrated on a NOTES appendectomy procedure performed on the NOSsE phantom. This involves an articulated robotic endoscope navigating to visualize the appendix while retroflexed. The horizon stabilization is additionally evaluated quantitatively against known ground truth. CONCLUSIONS: The combination of horizon stabilization and dynamic view expansion presents a realistic approach for reintroducing orientation and navigation cues during NOTES. The platform allows real-time implementation, which is an important prerequisite for further clinical evaluation.
Authors: David P Noonan; Christopher J Payne; Jianzhong Shang; Vincent Sauvage; Richard Newton; Daniel Elson; Ara Darzi; Guang-Zhong Yang Journal: Med Image Comput Comput Assist Interv Date: 2010
Authors: Eric J Seibel; Robert E Carroll; Jason A Dominitz; Richard S Johnston; C David Melville; Cameron M Lee; Steven M Seitz; Michael B Kimmey Journal: IEEE Trans Biomed Eng Date: 2008-03 Impact factor: 4.538
Authors: Rosebet Miranda-Luna; Christian Daul; Walter C P M Blondel; Yahir Hernandez-Mier; Didier Wolf; François Guillemin Journal: IEEE Trans Biomed Eng Date: 2008-02 Impact factor: 4.538
Authors: James Clark; Mikael Sodergren; David Noonan; Ara Darzi; Guang-Zhong Yang Journal: J Laparoendosc Adv Surg Tech A Date: 2009-04 Impact factor: 1.878