Assuming that we already know the image of the centerline, we want the lumen to be displayed on a plane, which requires curved projection reformation(CPR).The figure below is a common coronary centerline display in the paper.
The center line is a representation of a series of points.
You can call the function cpr(img_name, center_line_name) directly, where img_name represents the path of the image and center_line_name represents the path of the centerline point. The former supports .nii.gz and .mha files. The latter only supports .npy files.
The center line is represented by a 3D image.
You need to first convert the centerline image into a series of sequential points. Here provides a minimum path method for continuous centerline extraction, which is called find_point_list(thin_label_name, start, end), where thin_label_name is the path of 3D image, start is the coordinates of the starting point of the center line and end is the coordinates of the ending point of the center line.This function will save the center line as an .npy file.