**3. Results**

The final 3D printed model of the rat brain (Figure 6) is composed of the eight smoothed parts, all parts of which fit together perfectly. The scaled measures of each part are shown in Table 2. In addition, an external container was made to accommodate the eight regions.

**Figure 6.** Final 3D model (disassembled) of the rat template.

This container was designed using the external contour of the brain in the template. It was then enlarged a little (4.2:1) to facilitate the accommodation of the other parts. In addition, the cover was divided in two to make it possible to open, as shown in Figure 7. Figure 6 shows the whole-brain model disassembled, all eight parts and the container that encloses them. Figure 7 shows that all of the parts fits perfectly together.

**Table 2.** Measurements on X, Y, and Z axes of each part, total assembly without container and total assembly with container.


**Figure 7.** Final 3D model of the rat template.

In terms of material for the printing process, poly(lactic acid) (PLA) was chosen. This material is a thermoplastic polymer that is made from renewable resources. It melts at 160 degrees Celsius and it is less rigid than Acrylonitrile Butadiene Styrene (ABS), another commonly used material in 3D printing. This is important for the 3D model because it makes the printed parts more elastic and shock-resistant.

#### **4. Discussion**

Rapid prototyping has grown beyond its initial use in industrial sectors, e.g., the automobile industry, and today can be regarded as one of the most promising techniques for use in medical imaging. Although medical applications are relatively recent, their enormous potential has already been demonstrated in several studies [25–27]. Rapid prototyping objects are useful for training surgeons, as they allow surgical procedures to be simulated in a realistic manner. Even though medical research has already benefitted from rapid prototyping [28,29] and efforts have been made in the development of artificial organs and tissues [30,31], the traditional approach of teaching anatomy remains the same. This approach focuses mainly on MRI, CT and dissections of reals animals, without taking into consideration any variations and pathological changes. Currently, there is a well-known rat brain model called "C29 Rat Brain Comparative Anatomy", which aims for educative point of view, but is not detailed enough for scientific purposes.

The 3D printed brain that is presented in this paper could serve as the medium for bringing anatomical variations to pre-clinical studies in order to improve the understanding of anatomy [7] while preserving animal lives. In addition, by using other materials to adapt transparency or rigidity, certain aspects can be emphasized for the trainee and the student.

In the scientific domain, adding the cranium of the subject to the 3D model, we envision applications of this approach in designing and probing brain implant prototypes, as recording chambers, electric and/or optogenetic stimulation holders, and designing coils for transcranial magnetic or direct current stimulation.

#### **5. Conclusions**

In this paper, a step-by-step, end-to-end easily replicable methodology to generate a 3D printed model of a rat brain has been presented. It is based on an MRI template of the rat brain co-localized with the Paxinos and Watson rat brain atlas, combined with 3D CAD tools and 3D prototyping technologies to obtain a tangible 3D printed atlas of the rat brain. In addition, each 3D printed region is designed to be easily attached or detached from the other parts. This model could have a big impact on the way biologists and neuroscientists teach in the future. It makes more "tangible" the understanding of the relevant areas of the rat brain for teaching, while reduces the use of experimental animals for these purposes. In addition, the methodology can be extrapolated to different regions of the brain or even to other body regions.

To obtain the necessary images to generate the 3D parts, a digital atlas, which was correlated with a stereotaxic anatomical atlas, was used. With the participation of an experienced biologist, 466 structures were extracted and clustered into eight different parts, which correspond to the eight pieces of the printed rat brain that is presented in this paper. These parts are the Neocortex, the Archicortex, the Paleocortex, the Basal Ganglia, the Basal Telencephalon, the Diencephalon, the Mesencephalon, and the Metencephalons. Each part has joins to fasten together the different parts. In addition, a housing formed by three additional pieces was created. This in turn performs the function of storing the rest of the pieces and serves as a spatial orientation for the assembly of the brain. In further research, a complete rat brain with 38 sub-regions will be developed.

This work presents a methodology that could be expanded to almost any field of clinical and pre-clinical research, and moreover it avoids the need of dissect animals to teach anatomy. In addition, all parts are available at http://dmoratal.webs.upv.es/research.html.

**Author Contributions:** Conceptualization, D.R.Q., S.C. and D.M.; Data curation, D.R.Q. and D.M.; Formal analysis, D.R.Q., J.F.-A., R.P.-F. and S.C.; Funding acquisition, J.A.G.-M. and D.M.; Investigation, D.R.Q., J.F.-A., S.C. and D.M.; Methodology, D.R.Q., R.P.-F., J.A.G.-M., S.C. and D.M.; Project administration, J.A.G.-M. and D.M.; Resources, J.A.G.-M., S.C. and D.M.; Software, D.R.Q., J.F.-A. and R.P.-F.; Supervision, J.A.G.-M., S.C. and D.M.; Validation, D.R.Q., S.C. and D.M.; Visualization, D.R.Q.; Writing—original draft, D.R.Q. and J.F.-A.; and Writing—review and editing, R.P.-F., J.A.G.-M., S.C. and D.M.

**Funding:** This work was supported in part by the Spanish Ministerio de Economía y Competitividad (MINECO) and FEDER funds under grants BFU2015-64380-C2-2-R (D.M.) and BFU2015-64380-C2-1-R and EU Horizon 2020 Program 668863-SyBil-AA grant (S.C.). S.C. acknowledges financial support from the Spanish State Research Agency, through the "Severo Ochoa" Programme for Centres of Excellence in R&D (ref. SEV- 2013-0317). D.R.Q. was supported by grant "Ayudas para la formación de personal investigador (FPI)" from the Vicerrectorado de Investigación, Innovación y Transferencia of the Universitat Politècnica de València.

**Acknowledgments:** The authors are grateful to Begoña Fernández for their technical support.

**Conflicts of Interest:** The authors declare no conflict of interest.
