search

BubblyYi / Coronary-Artery-Tracking-via-3D-CNN-Classification

The PyTorch re-implement of a 3D CNN Tracker to extract coronary artery centerlines with state-of-the-art (SOTA) performance. (paper: 'Coronary artery centerline extraction in cardiac CT angiography using a CNN-based orientation classifier')
MIT License
187 stars 40 forks source link

How to Make a label #22

Open wxr521314 opened 1 year ago

wxr521314 commented 1 year ago

Hello, I would like to ask how to make the label, is the label converted into txt? There should be several txt files, what is the content of this txt file? Looking forward to your reply

francescopisu commented 1 year ago

The paper states that the 3 main vessels (RCA, LAD and LCX) as well well some minor branches must be annotated, for each CCTA scan. The annotations, in the form of .txt files, must contain IJK coordinates of points along the vessels centerlines, as well as the lumen radius at each point. Let's suppose that you choose to annotate the RCA, then you'd create a vessel_RCA.txt file having as many rows as the number of points along the centerline, and four columns, three for the X Y and Z IJK coordinates, and the fourth for the lumen radius.