Abstract
Neuroimaging technologies such as Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) collect three-dimensional data (3D) that is typically viewed on two-dimensional (2D) screens. Actual 3D models, however, allow interaction with real objects such as implantable electrode grids, potentially improving patient specific neurosurgical planning and personalized clinical education. Desktop 3D printers can now produce relatively inexpensive, good quality prints. We describe our process for reliably generating life-sized 3D brain prints from MRIs and 3D skull prints from CTs. We have integrated a standardized, primarily open-source process for 3D printing brains and skulls. We describe how to convert clinical neuroimaging Digital Imaging and Communications in Medicine (DICOM) images to stereolithography (STL) files, a common 3D object file format that can be sent to 3D printing services. We additionally share how to convert these STL files to machine instruction gcode files, for reliable in-house printing on desktop, open-source 3D printers. We have successfully printed over 19 patient brain hemispheres from 7 patients on two different open-source desktop 3D printers. Each brain hemisphere costs approximately $3–4 in consumable plastic filament as described, and the total process takes 14–17 hours, almost all of which is unsupervised (preprocessing = 4–6 hr; printing = 9–11 hr, post-processing = <30 min). Printing a matching portion of a skull costs $1–5 in consumable plastic filament and takes less than 14 hr, in total. We have developed a streamlined, cost-effective process for 3D printing brain and skull models. We surveyed healthcare providers and patients who confirmed that rapid-prototype patient specific 3D models may help interdisciplinary surgical planning and patient education. The methods we describe can be applied for other clinical, research, and educational purposes.
Highlights
Neuroimaging technologies such as Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) have become indispensable tools for the diagnosis and treatment of central nervous system disease
We have found that acrylonitrile butadiene styrene (ABS) prints may be slightly easier to clean, and one can consider using an acetone vapor bath with ABS prints to smooth the surface; we tend to print in polylactic acid (PLA) for the reasons described previously
Whereas several groups have presented the possible functions of 3D printed objects, we have presented more detail on a step-by-step process of how to go from clinical Digital Imaging and Communications in Medicine (DICOM) images to final 3D object, and provided data by which to select desktop 3D printing parameters as well as how to modify them for a given desktop 3D printer, environment, or need
Summary
Neuroimaging technologies such as Magnetic Resonance Imaging (MRI) and Computed Tomography (CT) have become indispensable tools for the diagnosis and treatment of central nervous system disease While both MRIs and CTs collect three-dimensional data (3D), PLOS ONE | DOI:10.1371/journal.pone.0136198. The progression of open source platforms has increased the capabilities and popularity of desktop 3D printers These 3D printers are less expensive than industrial 3D printers with the most affordable and common type currently utilizing fused deposition modeling (FDM) technology, known as fused filament fabrication (FFF). This additive process prints the object one layer at a time by extruding heated plastic through a nozzle, while moving around a build plate. We contextualize the possible utility of such a process through surveys with healthcare providers and patients in an epilepsy clinic
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
