Feasibility of Easy to Use and Inexpensive Three-Dimensional Printed Educational Model of Temporal Bone: Practiced Without Drilling

Abstract

OBJECTIVE

Three-dimensional (3D) printed temporal bone model draws great attention as a promising alternative for conventional cadaveric model in education of otologic surgery. However, its high price and requirement for specialized tools hinder widespread use. We devised a simple educational model based on lattice structure to overcome these problems and compared it with a commercial model.

METHODS

We converted high-resolution temporal bone computed tomography images into stereolithography format, and printed it using the G005 3D printing system from CUBICON^©. In this process, the part to be drilled out was made of lattice structure. We evaluated the model by a questionnaire prepared in advance, and compared the results with those of a commercial model.

RESULTS

We created an educational 3D printed temporal bone lattice model one-tenth the cost of commercial temporal bone. Our model reproduced the important structures of the temporal bone, produced less dust, and had similar strength and grinding sensation compared to the commercial model. The surface texture and reproducibility were comparable to the commercial model. Although most of structures were remodeled more elaborately in the commercial model than our model, our model demonstrated significant potential as a cost-effective educational tool for medical students and residents.

CONCLUSION

3D printed temporal bone lattice model has potential for widespread use due to low cost and easy accessibility. Further improvements in the fine structures of the temporal bone are necessary to enhance its utility as an educational model.

Introduction

The temporal bone is one of the most complex structures in the human body. ¹ Temporal bone surgery is also exceptionally challenging because critical structures – such as facial nerve, carotid artery, labyrinth, ossicles, temporal lobe dura, and sigmoid sinus – are compacted in a small space with intricate anatomical relationships.^2,3 Consequently, deliberate practice through cadaveric dissection has traditionally been considered essential for becoming an otologic surgeon.^1,4 However, acquiring cadavers becoming increasingly difficult, and significant cost are involved in maintaining them.^5–7 In addition, specialized tools such as drilling systems and surgical instruments are required for temporal bone dissection.

Due to these challenges, researches on alternative temporal bone dissection methods for medical student and otolaryngology residents are rapidly increasing. ⁶ Virtual reality (VR) training systems and three-dimensional (3D) printed temporal bone models are promising solutions to these problems.^2,8 VR training systems have demonstrated excellent training efficacy comparable to traditional methods. ⁹ 3D printed temporal bone models are also advantageous because it is relatively cheap, easy to manage, and free from ethical concerns.^6,10 These technologies allow inexperienced otologists to improve their surgical skills without complications associated with obtaining cadavers.

However, the actual utilization rate of these tools remains low. The cost of VR and 3D printed model is still burdensome for individual trainees, and opportunities to use these tools with appropriate equipment are limited. Thus, no tool has yet gained overwhelming advantage in terms of cost and accessibility over cadaveric models. ¹¹ To address these issues, a different surgical practice model that is significantly cheaper and more accessible than existing ones is needed.

To solve these problems, we devised a new educational temporal bone model based on a lattice system. The model is inexpensive and doesn’t require extensive equipment such as microscope, drilling devices, and suction irrigation systems. Here, we introduce our model, and discuss its advantages and what needs to be improved.

Materials and methods

Model design

The primary aim of this study is to create easily practicable temporal bone educational model using 3D printing techniques. We aimed to make tissues in surgical field of temporal bone model very soft so that they can be explored by manual grinding with basic tools. This model intends to make it easier and cheaper for trainees to learn temporal bone surgery at home without need for drilling.

To create the model, we acquired high-resolution temporal bone computed tomography (HR-TBCT) images without contrast agent from an anonymized healthy patient. We used Sensation 16 CT scanner (Siemens Healthineers^©, Germany), and followed the scanning protocol of the radiology department at Seoul National University Hospital (scan parameters: 120 kVp, slice thickness: 0.75 mm). HR-TBCT images were initially saved as Digital Imaging and Communications in Medicine (DICOM) format. We converted them into stereolithography (STL) format using Mimics and 3-Matic software (Materialise NV^©, Belgium).

Then, we divided the model into two parts – ‘to be removed’ and ‘to be remained’. The part that is usually drilled out in typical otologic canal wall down mastoidectomy was designated as the ‘to be removed’ part. // To achieve this, we performed canal wall down mastoidectomy on a standard 3D printed temporal bone model. Then we scanned the modified model and set ‘removed’ part into a lattice structure. An otologic specialist reviewed and modified the STL file during this process. Because the lattice structure is much more fragile than the other parts, it can be easily crushed out by simple tools.

Although the resolution of TBCT in this study is very high – the thickness of each section is within 0.75 mm – it is not sufficient to express structure of the ossicles. An ossicle model should be able to show all ossicles (malleus, incus, and stapes), and their substructures clearly. To solve this problem, we manually modified the assembled STL file, shaping the images to resemble real ossicles as closely as possible.

Equipment and materials

We printed temporal bone model using stereolithography method with G005 printing system (CUBICON^©, http://www.3dcubicon.com/, Republic of Korea). Photocurable resin was used as the plastic filament material. The models were printed to actual size with dimensions of 52 mm (width) × 90 mm (height) × 48 mm (depth) (Figure 1).

Figure 1.

3D printed lattice model of temporal bone. Some parts of this model are printed with a lattice structure. Trainees can learn stereoscopic anatomy and surgical procedures of the temporal bone by manually crushing the lattice structure.

Dissection and evaluation of the 3D printed lattice model

On September 18, 2019, twenty-four trainees participated in a temporal bone dissection training program at Seoul National University Hospital using 3D printed lattice models. All participants were third- or fourth-year residents. Lower-year residents and attendants already qualified as otolaryngology specialists were excluded. The trainees dissected the models using only basic surgical tools – freer elevators, curettes, and picks – without drilling devices (Figure 2). They also dissected temporal bone cadaver specimens using drilling devices. Each subject practiced one 3D printed lattice model and one cadaver specimen. Immediately after completing all practices, they evaluated 3D printed lattice models using a questionnaire we prepared (Supplementary Figure).

Figure 2.

Stepwise procedures of temporal bone dissection using 3D printed lattice model. Trainees can practice each step of mastoidectomy. (a) Beginning mastoidectomy around McEwen's triangle. (b) Opening aditus ad antrum. (c) Performing posterior tympanotomy. (d) Completing canal wall down mastoidectomy.

Questionnaire

The questionnaire comprises 30 questions about 3D printed lattice model, focusing on physical properties, structural resemblance, and usefulness in learning. In the “Physical properties” section, we evaluate dust production, grinding sensation, strength, surface texture, and reproducibility. This section assesses only the physical characteristics independently of subjective evaluation. ‘Dust’ refers to the amount of dust produced, ‘Grinding sensation’ to the heaviness of grinding, and ‘Strength’ to the hardness of the model. ‘Surface texture’ represents the similarity of surface patterns to the original cadaver, and ‘Reproducibility’ measures how closely the overall structure and small organs are represented at high resolution compared to the original cadaveric specimen. An optimal balance is desired for dust, grinding sensation, and strength, while higher scores are preferable for surface texture and reproducibility. Consequently, the cadaver specimen is scored 5 for dust, grinding sensation, and strength, and 10 for surface texture and reproducibility. For analysis, we classified the first 3 questions as group A and the last 2 as group B.

In the “Structural resemblance” section, we estimated the similarities of 17 structures to cadaver specimens. Higher scores indicate closer resemblance to the cadaver specimen. In the “Usefulness in learning” section, we evaluate subjective usefulness in surgical learning. Higher scores are more desirable in this section.

Dissection and evaluation of the commercial temporal bone model

We used commercial temporal bone model, ENT Model in ANATDEL solution (MEDICAL IP^©, https://medicalip.com/, Republic of Korea), as secondary material to cadaver specimens in temporal bone dissection training program. On October 7, 2020, twenty-eight trainees participated in this program. All included subjects were third- or fourth-year residents. Lower-year residents or attendants already qualified with otolaryngology specialty were excluded. All of them dissected both commercial model and cadaver specimens in the same way using drilling devices. They evaluated commercial model using a questionnaire we prepared (Supplementary Figure). No one had attended previous temporal bone dissection training program. Therefore, no subjects participated in both the evaluation of the 3D printed lattice model and the commercial model.

Statistical analysis

The results are presented as mean ± standard deviation and standard error. The questionnaire results were analyzed utilizing one-sample t-test. We set the criteria for statistical significance at p-value of less than 0.050. All statistical analyses were performed using R version 4.3.3 software (The R Foundation for Statistical computing, Vienna, Austria).

Compliance with reporting guidelines

The reporting of this study conforms to the CONSORT 2010 statement. ¹²

Results

Physical properties

In all criteria, commercial model is closer to a cadaver specimen than the 3D printed lattice model (Figure 3, Supplementary Table 1). Scores of group A questionnaires are closer to 5 when they are similar to cadavers. In the categories of dust and strength, commercial model shows nearly no difference from a cadaver specimen, while 3D printed lattice model has lower scores. The grinding sensation shows nearly no difference between commercial model and 3D printed lattice model. Scores for group B questionnaires are closer to 10 when they are similar to cadavers. In the categories of surface texture and reproducibility, the commercial model shows considerable differences from the cadaver specimen, but the 3D printed lattice model has even larger differences.

Figure 3.

Comparison of questionnaire results on physical properties of cadaver specimen, commercial model, and 3D printed lattice model. In all criteria, commercial model is more similar to the cadaver specimen than 3D printed lattice model (a) Scores of group A questions – dust, strength, and grinding sensation (b) Scores of group B questions – surface texture and reproducibility.

Structural resemblance

In all structures, scores in the commercial model are higher than those for the 3D printed lattice model (Figure 4, Supplementary Table 2). Most structures showed highly significant differences between the commercial model and 3D printed lattice model. Only external auditory canal showed no significant difference, and the middle fossa dura and sigmoid sinus had p-values over 0.010. However, these structures were located outside the lattice field, and their importance in practice is not very high. Structural resemblance scores for the commercial model showed not much variance between all structures (range: 5.42 ∼ 6.77). In the other hand, those for 3D printed lattice model had a wider distribution (range: 2.26 ∼ 5.74).

Figure 4.

Comparison of questionnaire results on structural resemblance of commercial model and 3D printed lattice model to cadaver specimen. In all structures, scores of commercial models are higher than those of 3D printed lattice model. Score distribution is more even in commercial model than 3D printed lattice model. M: malleus, I: incus, S: stapes, C: cochlea, V: vestibule, LSCC: lateral semicircular canal, SSCC: superior semicircular canal, PSCC: posterior semicircular canal, TM: tympanic membrane, EAC: external auditory canal, A: attic, FR: facial recess, FN: facial nerve, MFD: middle fossa dura, SS: sigmoid sinus, CA: carotid artery, VN: vestibular nerve.

Usefulness in learning

In all questions, trainees rated the commercial model higher than 3D printed lattice model (Figure 5, Supplementary Table 3). Trainees responded that the commercial model is most useful in learning how to use drill. However, they gave low scores for the reproducibility of soft tissue in the commercial model. For the 3D printed lattice model, trainees gave high scores for learning the stereoscopic structure, but low scores for practicing physical properties.

Figure 5.

Comparison of questionnaire results on usefulness of commercial model and 3D printed lattice model to cadaver specimen. In all criteria, trainees find the commercial model more useful than the 3D printed lattice model.

Discussion

In recent years, there has been a notable surge in interest surrounding 3D printing technologies, with increasing applications in the medical field. ¹³ Within the otologic field, much of the research has concentrated on the development of educational models for the temporal bone. Technical barriers to temporal bone surgery and difficulties in obtaining cadaveric specimens have made demands for temporal bone educational models significant. ¹¹ However, the actual utilization of existing 3D printed educational temporal bone models remains not so active due to relatively high cost and low accessibility.

To overcome these problems, we approached the issue in a different direction. Instead of aiming to create a model that closely resembles cadaveric specimen, we aimed to produce a model which can be practiced more easily for inexperienced otologists. We recognized that the main issue with existing models is that they can only be practiced in a highly organized facility with specialized equipment like microscopes, drilling systems, and suction irrigation systems. Therefore, our goal is to design models that could be practiced at home without any need for specialized equipment. As a result of our efforts, we developed a 3D printed lattice model. We designed some part of the model as a thin lattice structure that could be drilled out. The lattice structure was intentionally made very thin so that they could be easily crushed using simple tools. With this model, trainees can practice temporal bone dissection using basic instruments such as a freer, curette, and pick elevator.

After model production, we compared it with a commercially available model. Since these two models serve different purposes, a direct comparison may be somewhat inappropriate. 3D printed lattice model is developed to make the cheap and easy method for the education of otologic surgery so that trainees can practice at home individually. In contrast, commercial model is made to make the most similar substitute for cadaver. However, it is necessary to compare this model to similar model to grasp its properties comprehensively because there is no comparison target for 3D printed lattice model.

When considering “Physical properties” such as strength, surface texture, reproducibility, dust, and grinding sensation, the commercial model appeared to be more similar to a cadaveric specimen. 3D printed lattice model's “Structural resemblance” did not surpass that of the commercial model in any criteria. And, in terms of “Usefulness”, the commercial model received higher scores from the trainees across all aspects. In summary, our model did not catch up existing commercial models in terms of physical characteristics, structure, and learning effectiveness.

However, 3D printed lattice model has a big merit in price. Making price of our model is appraised $35 for each by production company. It is only one-tenth of commercial 3D printed model. Considering its material cost is about $5 for each, we might be able to make it much cheaper than current price through mass production. And its affordability, accessibility, and compatibility with simple tools make 3D printed lattice model be easily utilized at home without the need for complex equipment.

In the process of acquiring surgical skills, a trainee may use multiple educational models in learning path. ⁶ Each model can fulfill its own distinct role, compensating for the weaknesses of other models. 3D printed lattice model can serve as an initial educational tool for understanding the anatomical structure of the temporal bone, ultimately contributing to the overall training process (Figure 6). If the 3D printed lattice model is utilized with VR and commercial 3D printed models, trainees could have plentiful training experiences before performing real surgery.

Figure 6.

The position of 3D printed lattice model in the overall learning path of otologic surgery. A 3D printed lattice model can serve as a useful and cost-effective practical subject for use at home and in the classroom.

To be a more effective educational model, 3D printed lattice model should be improved in several areas. First, higher quality CT images are required. Although the resolution of the HR-TBCT used in this study is significantly higher than that of standard CT images, further improvements are needed to obtain more detailed information regarding small structures like the ossicles. However, enhancing CT resolution is not a simple task, because it requires higher radiation, which can be harmful to patients. We believe that good solutions such as utilizing artificial intelligence techniques would be appear in close future. Second, resolution of the printer also needs to be improved. In the short term, one alternative solution could be to increase the size of the model. Since the actual surgical procedure is performed within the microscope's field of view, enlarging the model could be a viable approach that doesn't compromise the effectiveness of practice. However, in the long term, improving the printing technique itself would be a fundamental solution. Third, it is essential to assign a unique color to each part of the temporal bone. Differentiating various parts with distinct colors makes trainees learn temporal bone anatomy more easily and accurately. Although the technology colorizing each part already exists, it is too expensive to utilize.

This study possesses several limitations. The evaluation of this model relied solely on a questionnaire which was not validated and pilot-tested, which introduces the possibility of subjective bias from the trainees. Additionally, since the trainees using the commercial model and those using the 3D printed lattice model were different, the results might have been influenced by the inherent tendencies of each group. To overcome these limitations, future research which assesses trainees’ surgical performances using objective tools in randomized controlled trial format would be needed. Such research would provide a more comprehensive understanding of the comparative effectiveness of these different approaches.

Conclusion

We developed a novel 3D printed temporal bone educational model using lattice structure. Although this model did not surpass commercial model in most criteria, it still has strength in price and accessibility. With its enhanced resolution and color of fine structures, this model would have the potential to serve as a valuable educational model.

Supplemental Material

sj-docx-1-mde-10.1177_23821205241289500 - Supplemental material for Feasibility of Easy to Use and Inexpensive Three-Dimensional Printed Educational Model of Temporal Bone: Practiced Without Drilling

Supplemental material, sj-docx-1-mde-10.1177_23821205241289500 for Feasibility of Easy to Use and Inexpensive Three-Dimensional Printed Educational Model of Temporal Bone: Practiced Without Drilling by Sang-Youp Lee, Baren Jeong, Whal Lee and Moo Kyun Park in Journal of Medical Education and Curricular Development

Supplemental Material

sj-docx-2-mde-10.1177_23821205241289500 - Supplemental material for Feasibility of Easy to Use and Inexpensive Three-Dimensional Printed Educational Model of Temporal Bone: Practiced Without Drilling

Supplemental material, sj-docx-2-mde-10.1177_23821205241289500 for Feasibility of Easy to Use and Inexpensive Three-Dimensional Printed Educational Model of Temporal Bone: Practiced Without Drilling by Sang-Youp Lee, Baren Jeong, Whal Lee and Moo Kyun Park in Journal of Medical Education and Curricular Development

Supplemental Material

sj-doc-3-mde-10.1177_23821205241289500 - Supplemental material for Feasibility of Easy to Use and Inexpensive Three-Dimensional Printed Educational Model of Temporal Bone: Practiced Without Drilling

Supplemental material, sj-doc-3-mde-10.1177_23821205241289500 for Feasibility of Easy to Use and Inexpensive Three-Dimensional Printed Educational Model of Temporal Bone: Practiced Without Drilling by Sang-Youp Lee, Baren Jeong, Whal Lee and Moo Kyun Park in Journal of Medical Education and Curricular Development