Quantitative assessment of 3D printing model accuracy in delineating the normal heart anatomy based on in vitro phantom experiments

Shen-yuan Lee, Andrew Squlech, Zhonghua Sun


Although the diagnosis of heart disease has improved with the rapid development of scanning techniques such as computed tomography (CT), magnetic resonance imaging (MRI) and echocardiography, there are still limitations in diagnosing patients with congenital heart disease (CHD) due to its complex morphology.

The aim of this study is to use a preserved pig heart for conducting phantom experiments and creating a highly accurate 3D model using 3D printing technique.

Palatinate pig hearts were used in the phantom experiments to investigate the accuracy of the 3D printed model in comparison with the CT images and 3D segmentation files as well as the real object of the pig’s heart.

Eight comparisons and scatter plots were generated from six different datasets consisting of pig heart, 3D printed model, two standard tessellation language (STL) files and two CT images data. A strong correlation (r=0.99) was noted in each scatter plot while pig heart and 3D printed model averaging 0.21mm in difference.

This study has shown that the 3D model which was printed with a pig heart has high accuracy in replicating normal cardiac anatomy.
Full Text: PDF