Scripts for cardiac segmentation and registration applied to pulmonary hypertension.
Matlab 2016 or later.
Target images in NIfTI format.
3D cardiac atlases.
Run the script with the following arguments:
RunPAll('targets','3Datlas','PHatlas','3Dshapes',1,0,4)
Where targets are the images to be segmented in NIfTI format; 3Datlas are the healthy volunteer atlases; and PHatlas are the PH patient atlases.
The numerical arguments are: Landmarks (1=yes, 0=no); segment and derive meshes (1) or derive meshes from exisiting segmentations (0); and number of cores for parallel processing.
The general workflow is decribed below which was developed in a cohort of healthy adults with high-resoltuion cardiac MR imaging. This process was extended to pulmonary hypertension by the annotation of a set of disease specifc atlases with labels that inlcude the right ventricular cavity and freewall.
Detailed methods and validation are described in our paper:
Bai W, Shi W, de Marvao A, Dawes TJ, O'Regan DP, Cook SA, Rueckert D. A bi-ventricular cardiac atlas built from 1000+ high resolution MR images of healthy subjects and an analysis of shape and motion. Med Image Anal. 2015;26:133-145.
The mean template shape for pulmonary hypertension is available here: