Pre-operative surgical planning based on pre-operative CT image, showing left anterior descending artery, robotic arms and intercostal spaces; (b) US images of left lung deflation stage with the aortic valve and mitral valve extracted; (c) US images of the heart following lung deflation and after thoracic insufflation and their corresponding segmented mitral and aortic valve annuli. Images adapted from Cho et al. MICCAI 2010.
A flowchart that illustrates the procedural planning and implementation of RA-CABG surgery.
Graphical representations of the heart migration during RA-CABG procedure displayed using whole-heart blood pool segmentions from CT data: (a) after intubation and anaesthesia (b) after left lung deflation, and (c) after chest insufflation with CO2.
Landmark extraction using a custom-developed landmark extraction tool on (a) CT and (b) US images.
(a) An heart phantom image showing the LAD path; and (b) peri-operative US image acquisition protocol showing imaging of the apex and coronary ostia using incrementally tracked 2D US images.
(a) LAD extracted from patients CT data (b) LAD with the axes of rotation.
Graphical representation of the target registration errors of the two registration techniques (robust ICP-based and Weighted Landmark-based) from peri-operative stages (Stage0, Stage1, and Stage2) to pre-operative stage (Stage CT ).
(a) Pre-operative heart phantom model showing the gold-standard LAD; (b) Visual display of the predicted LAD from the landmark-based approach, (c) and predicted via the robust ICP-based approach.
Estimated RMS TRE of the LAD from Monte Carlo simulations which were conducted for each peri-operative stage.
LAD TRE: Mean ± SD and RMS (mm) as observed in the in vitro study.
RMS landmark localization error (mm) of weighted landmark-based registration.
Maximum FRE of four landmarks—AVA, MVA, LCO, LVAp—employed in clinical validation.
Article metrics loading...
Full text loading...