Mechanical occlusion of the Left atrial appendage (LAA) using a purpose-built device has emerged as an effective prophylactic treatment in patients with atrial fibrillation at risk of stroke and a contraindication for anticoagulation. A crucial step in procedural planning is the choice of the device size. This is currently based on the manual analysis of the “Device Landing Zone” from echocardiographic images.
We aimed to develop an algorithm for automated segmentation of the LAA landing zone from 3D echocardiographic images of the LAA.
MATERIAL AND METHODS:
In this experimental study, 2D axial images were derived from the 3D echo datasets. After image pre-processing, binary images were created using a thresholding method. A binary image matrix was then formed and scanned using 8-adgacency approach resulting in segmentation of the objects with a closed circumference within the image. Erosion/dilation techniques were then applied to remove small objects. A feature-based approach was then used to firstly detect the LAA region and secondly to identify the device landing zone.
A total of 22 datasets were used in this study. The algorithm produced up to 9 axial images as the proposed landing zone. The selected axial images were compared to the echocardiographic images. In 18 cases (81.8%), the algorithm successfully segmented the LAA and proposed the landing zone based on the defined features.
We have developed a simple and fast algorithm for semi-automated segmentation of the LAA landing zone. Further studies are needed to assess the accuracy of the proposed landing zones by this method.