The Future of Biomechanical Spine Research: Conception and Design of a Dynamic 3D Printed Cervical Myelography Phantom.

Background Three-dimensional (3D) printing is a growing practice in the medical community for patient care and trainee education as well as production of equipment and devices. The development of functional models to replicate physiologic systems of human tissue has also been explored, although to a lesser degree. Specifically, the design of 3D printed phantoms that possess comparable biomechanical properties to human cervical vertebrae is an underdeveloped area of spine research. In order to investigate the functional uses of cervical 3D printed models for replicating the complex physiologic and biomechanical properties of the human subaxial cervical spine, our institution has created a prototype that accurately reflects these properties and provides a novel method of assessing spinal canal dimensions using simulated myelography. To our knowledge, this is the first 3D printed phantom created to study these parameters. Materials and methods A de-identified cervical spine computed tomography imaging file was segmented using threshold modulation in 3D Slicer software. The subaxial vertebrae (C3-C7) of the scan were individualized by separating the facet joint spaces and uncovertebral joints within the software in order to create individual stereolithography (STL) files. Each individual vertebra was printed on an Ultimaker S5 dual-extrusion printer using white "tough" polylactic acid filament. A human cadaveric subaxial cervical spine was harvested to provide a control for our experiment. Both models were assessed and compared in flexion and extension dynamic motion grossly and fluoroscopically. The maximum angles of deformation on X-ray imaging were recorded using DICOM (Digital Imaging and Communications in Medicine) viewing software. In order to compare the ability to assess canal dimensions of the models using fluoroscopic imaging, a myelography simulation was designed. Results The cervical phantom demonstrated excellent ability to resist deformation in flexion and extension positions, attributed to the high quality of initial segmentation. The gross and fluoroscopic dynamic movement of the phantom was analogous to the cadaver model. The myelography simulator adequately demonstrated the canal dimensions in static and dynamic positions for both models. Pertinent anatomic landmarks were able to be effectively visualized for assessment of canal measurements for sagittal and transverse dimensions. Conclusions By utilizing the latest technologies in DICOM segmentation and 3D printing, our institution has created the first cervical myelography phantom for biomechanical evaluation and trainee instruction. By combining new technologies with anatomical knowledge, quality 3D printing shows great promise in becoming a standard player in the future of spinal biomechanical research.