David Schwartzman1, Hua Zhong. 1. Cardiovascular Institute, University of Pittsburgh, Pittsburgh, Pennsylvania 15213-2582, USA. schwartzmand@upmc.edu
Abstract
INTRODUCTION: The utility of "virtual" imaging systems for left atrial (LA) navigation has been hampered by inadequate spatial detail, as well as inaccurate integration of more detailed preoperative images, such as those generated by computed tomography (CT). CartoSound is an intracardiac echocardiography (ICE)-based technology that promises to ameliorate these problems. Our objective was to examine the capabilities and optimal use of CartoSound, both as a stand-alone tool and as a facilitator of CT image integration. METHODS AND RESULTS: In 10 patients, CartoSound models of the LA were generated using each of 4 ICE transducer locations: LA, right atrium (RA), coronary sinus (CS), and esophagus (ESO). Each of these models was used to register CT-derived LA models into the operative workspace. We correlated the comprehensiveness of LA imaging from each transducer location with the quality of the CT registration, as well as the accuracy of mock circumferential antral ablation guided by the CartoSound model alone or by the CT model. The LA transducer location provided the most comprehensive rendering of the LA, which was associated with higher quality CT registration and greater CT-guided mock ablation accuracy. Mock ablation guided by the CartoSound model alone was at least as accurate as CT, although the models were less intuitive. CONCLUSIONS: For LA navigation, optimal use of CartoSound may require LA transducer location, which is effective for stand-alone use and as a facilitator of CT image integration.
INTRODUCTION: The utility of "virtual" imaging systems for left atrial (LA) navigation has been hampered by inadequate spatial detail, as well as inaccurate integration of more detailed preoperative images, such as those generated by computed tomography (CT). CartoSound is an intracardiac echocardiography (ICE)-based technology that promises to ameliorate these problems. Our objective was to examine the capabilities and optimal use of CartoSound, both as a stand-alone tool and as a facilitator of CT image integration. METHODS AND RESULTS: In 10 patients, CartoSound models of the LA were generated using each of 4 ICE transducer locations: LA, right atrium (RA), coronary sinus (CS), and esophagus (ESO). Each of these models was used to register CT-derived LA models into the operative workspace. We correlated the comprehensiveness of LA imaging from each transducer location with the quality of the CT registration, as well as the accuracy of mock circumferential antral ablation guided by the CartoSound model alone or by the CT model. The LA transducer location provided the most comprehensive rendering of the LA, which was associated with higher quality CT registration and greater CT-guided mock ablation accuracy. Mock ablation guided by the CartoSound model alone was at least as accurate as CT, although the models were less intuitive. CONCLUSIONS: For LA navigation, optimal use of CartoSound may require LA transducer location, which is effective for stand-alone use and as a facilitator of CT image integration.
Authors: Douglas N Stephens; Uyen T Truong; Amin Nikoozadeh; Omer Oralkan; Chi Hyung Seo; Jonathan Cannata; Aaron Dentinger; Kai Thomenius; Alan de la Rama; Tho Nguyen; Feng Lin; Pierre Khuri-Yakub; Aman Mahajan; Kalyanam Shivkumar; Matt O'Donnell; David J Sahn Journal: J Ultrasound Med Date: 2012-02 Impact factor: 2.153
Authors: David E Haines; Salwa Beheiry; Joseph G Akar; Janice L Baker; Doug Beinborn; John F Beshai; Neil Brysiewicz; Christine Chiu-Man; Kathryn K Collins; Matthew Dare; Kenneth Fetterly; John D Fisher; Richard Hongo; Samuel Irefin; John Lopez; John M Miller; James C Perry; David J Slotwiner; Gery F Tomassoni; Esther Weiss Journal: Heart Rhythm Date: 2014-05-07 Impact factor: 6.343