Lumen Contouring and Retrospective Gating of Coronary OCT

More Info


Cardiovascular disease causes decreased blood flow to the heart and the brain. Due to the buildup of plaque the coronary arteries become narrow. The plaque can suddenly disrupt and partially or completely occlude a vessel, this can lead to myocardial infarction and sudden death. Angioplasty is a surgical procedure during which stents are placed inside the coronary arteries to keep these vessels open. Quantitative analysis can be performed on an imaging technique called OCT to study the lumen areas over time and the effects of implanted stents. The lumen areas are the part of an artery where the blood flow through. Contour detection is necessary for calculating the lumen areas. To our knowledge the lumen contours of the coronary arteries from OCT data-sets are currently only manually detected, no method for automatic contour detection exists. The lumen volume can be calculated using the lumen areas and is also used in quantitative OCT analysis. However, motion artifacts are introduced due to the cardiac cycle while acquiring an OCT data-set. These motion artifacts reduce the accuracy of the calculated lumen volume. The focus of this thesis is on automatic contour detection and selecting a subset of cross-sections is proposed as a possible solution for reducing these motion artifacts.