Premium
A novel 3D printing method for accurate anatomy replication in patient‐specific phantoms
Author(s) -
Okkalidis Nikiforos
Publication year - 2018
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1002/mp.13154
Subject(s) - imaging phantom , dicom , hounsfield scale , contouring , 3d printed , voxel , pixel , computer science , nuclear medicine , replication (statistics) , biomedical engineering , medicine , artificial intelligence , computed tomography , computer graphics (images) , radiology , virology
Purpose A novel 3D printing method has been developed, which employs human CT images to construct patient specific phantoms by controlling the filament extrusion rate. Methods An in‐house software was developed comprising pixel‐by‐pixel (PbP) reading of the Hounsfield Units ( HU ) values in the original patient DICOM images and their sufficiently accurate 3D printed replication in the phantom produced. Results The PbP method was applied to two sets of anonymized patients’ CT chest and skull images. The respective patient specific phantoms were 3D printed and then CT scanned and compared to the actual patient images. The chest phantom images were also compared to those of another phantom created employing the older variable infill density method ( VID ). Conclusions The results clearly indicated a significant improvement both visually and in the phantom HU values achieved. In contrast to other methods published, its major advantages are: (a) no need for manual contouring and 3D modeling of a patient's organs, (b) wider density range, and (c) significantly better simulation of the organs’ HU .