Key result
Fully automatic 3D aortic root and LVOT segmentation achieves a ~0.9 Dice similarity coefficient.
Why the study?
Analyzing aortic and LVOT anatomy is crucial for risk assessment and TAVI planning, but generating training data for fully automatic 3D segmentation requires efficient annotation methods.
Does a fully automatic 3D segmentation network trained on sparse annotations provide accurate measurements for TAVI planning?
Does a fully automatic 3D segmentation network trained on sparse annotations provide accurate measurements for TAVI planning?
A fully automatic 3D segmentation network trained on sparse annotations provides accurate and reproducible measurements of the aortic root and LVOT for TAVI planning.
May aid automated TAVI measurements; leaves open prospective clinical validation before practice change.
Analyzing the anatomy of the aorta and left ventricular outflow tract (LVOT) is crucial for risk assessment and planning of transcatheter aortic valve implantation (TAVI). We propose 2D cross-sectional annotation and point cloud-based surface reconstruction to train a fully automatic 3D segmentation network for the aortic root and the LVOT. Our sparse annotation scheme enables easy and fast training data generation for tubular structures like the aortic root. Based on this annotation concept, we trained a 3D segmentation model that achieves a Dice similarity coefficient (DSC) of 0.9 and an average surface distance (ASD) of 0.96 mm. In addition, we show that our fully automatic segmentation approach facilitates reproducible and quantifiable measurements for TAVI planning. Our approach achieves an aortic annulus maximum diameter difference between prediction and annotation of 0.45 mm (inter-observer variance: 0.25 mm).
No takes yet. Share an insight, caveat, or question.
Brosig et al. (2024) studied Aortic root and LVOT anatomy for TAVI planning. Fully automatic 3D segmentation network based on sparse annotations vs. Manual annotation (inter-observer variance) was evaluated on Dice similarity coefficient (DSC), average surface distance (ASD), and aortic annulus maximum diameter difference. A fully automatic 3D segmentation model for the aortic root and LVOT achieved a Dice similarity coefficient of 0.9, an average surface distance of 0.96 mm, and a diameter difference of 0.45 mm.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: