Open-source 4D statistical shape model of the heart for x-ray projection imaging

We present an open-source 4D (3D+t) statistical shape model of the heart developed as numerical phantom for cone-beam CT simulation. The training set consists of surface meshes from 20 ten-phase-CT angiography data sets extracted using automatic registration-based segmentation. Incorporating 90% of the training set variation, the model exhibits a generalization ability of 5.00 ± 0.93 mm and specificity of 7.30 ± 0.97 mm. The model is implemented in CONRAD, an open-source simulation and reconstruction framework. We provide all algorithms and high resolution projection data on the project's homepage. To the best of our knowledge, this is the first open-source 4D statistical shape model of the heart.