Segmentation and visualization for cardiac ablation procedures
Massachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science.
MetadataShow full item record
In this thesis, we present novel medical image analysis methods to improve planning and outcome evaluation of cardiac ablation procedures. Cardiac ablation is a common medical procedure that consists of burning cardiac tissue causing atrial fibrillation, or irregular contractions of the heart's atria. We first propose a method for the automatic delineation of the left atrium in magnetic resonance (MR) images acquired during the procedure. The high anatomical variability of the left atrium shape and of the pulmonary veins that drain into it presents significant difficulties for cardiac ablation. Consequently, accurate visualization of the patient's left atrium promises to substantially improve intervention planning. We perform the segmentation using an automatic atlas-based method, which makes use of a set of example MR images in which the left atrium was manually delineated by an expert. We demonstrate that our approach provides accurate segmentations that are also robust to the high anatomical variability of the left atrium, while outperforming other comparable methods. We then present an approach to use this knowledge of the shape of the left atrium to aid in the subsequent automatic visualization of the ablation scars that result from the procedure and are visible in MR images acquired after the surgery. We first transfer the left atrium segmentation by aligning the pre and post-procedure scans of the same patient. We then project image intensities onto this automatically generated left atrium surface. We demonstrate that these projections correlate well with expert manual delineations. The goal of the visualization is to allow for inspection of the scar and improve prediction of the outcome of a procedure. This work has a potential to reduce the considerable recurrence rates that plague today's cardiac ablation procedures.
Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2011.Cataloged from PDF version of thesis.Includes bibliographical references (p. 67-70).
DepartmentMassachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science.
Massachusetts Institute of Technology
Electrical Engineering and Computer Science.