Process Chain for Fabrication of External Maxillofacial ... · The process chain for the...
Transcript of Process Chain for Fabrication of External Maxillofacial ... · The process chain for the...
2017Vol. 2 No. 3: 27
Research Article
DOI: 10.4172/2472-1654.100067
iMedPub Journalshttp://www.imedpub.com
1© Under License of Creative Commons Attribution 3.0 License | This article is available in: http://healthcare-communications.imedpub.com/archive.php
Journal of Healthcare Communications ISSN 2472-1654
Nneile Nkholise
Central University of Technology, Bloemfontein, South Africa
Corresponding author: Nneile Nkholise
Central University of Technology, Bloemfontein, South Africa.
Tel: +234 8074965343
Citation: Nkholise N. Process Chain for Fabrication of External Maxillofacial Prosthesis Using Additive Manufacturing. J Healthc Commun. 2017, 2:3.
Process Chain for Fabrication of External Maxillofacial Prosthesis Using Additive
Manufacturing
Received: April 20, 2017; Accepted: May 05, 2017; Published: May 12, 2017
AbstractExternal maxillofacial prostheses are presently fabricated through traditional manufacturing techniques in South Africa. The limited number of technologists with the necessary skill as well as the lengthy time it takes to produce a single prosthesis results in a significant backlog in the provision of these prostheses at especially state run hospitals. Although AM has been used before as secondary process to manufacture these prostheses, no proper study has been done to determine which AM process is best suited to this application. An improved process chain to manufacture external maxillofacial prostheses can be developed using AM. These prostheses can be manufactured at reduced time and without the necessary skill required to first sculpt a pattern in wax. An investigation into determining which AM process is best suited to manufacture external maxillofacial prostheses will lead to more accurate and cost effective prosthetics.
Keywords: Maxillofacial; Healthcare; Ablative surgery; Surgical techniques; Medical field
IntroductionThe maxillofacial prosthetics specialty is relatively small compared to other healthcare professions and technological developments are often denied to them, as commercial companies cannot recoup the investment required from such a small market [1]. Therefore, many of their practices are inherently labor intensive, requiring inordinate amounts of skill and training to become competent. Faced with increasing case numbers and a reduced number of new prosthetists entering the profession, there is urgency to update techniques to improve efficiency. The need to embrace technological development is clear. The profession has a long and well-documented history of adopting, adapting, and improving on existing technologies found in industry.
Bibb et al. [2] proves the importance of maxillofacial prosthesis in treating patients suffering from facial deformity, congenital, traumatic, or ablative surgery. In Bibb’s study, the technology covers treatment, planning (methods and procedure) and rehabilitation of the patient thereafter.
Vast improvements in the medical field, has resulted in improvements in surgical techniques which has significantly contributed to the cancer survival rate [2]. These improvements have resulted in ever-increasing patient numbers requiring
maxillofacial prosthesis. A shortage of skilled technologists that are able to produce these prostheses leads to a significant shortfall in the delivery of this much needed service to the community.
With the increasing number of patients requiring a maxillofacial prosthesis, there is a great need to enhance the efficiency of maxillofacial prostheses manufacturing [3]. The lack of maxillofacial prosthesis accessibility poses a large problem in many countries especially Africa and South America. Approximately 4,259 external prostheses are produced by 50 hospitals in the United Kingdom (UK) annually [2]. This number includes other work typically undertaken by maxillofacial prostheses manufacturers such as breast, nipple, hand and finger prostheses. Whilst the UK enjoys a comprehensive maxillofacial prosthetics service through their National Health Service, other nations that rely on healthcare insurance have lower levels of provision. However, even taking that into account, it would be reasonable to anticipate similar levels of activity throughout countries in Western Europe, North America, Japan, Australia and New Zealand. Using the UK figures, a crude estimate based on population sizes would suggest that more than 64,000 facial prostheses are made each year in the wealthiest nations.
2017Vol. 2 No. 3: 27
2 This article is available in: http://healthcare-communications.imedpub.com/archive.php
ARCHIVOS DE MEDICINAISSN 1698-9465
Journal of Healthcare Communications ISSN 2472-1654
The increasing demand for facial prostheses throughout the developing world has led researchers to explore the potential benefits of using computer-aided technologies, such as AM for the manufacturing of prostheses. Additive Manufacturing is a technology that has the potential to successfully change certain spheres in the field of medical science [4]. The opportunities of this technology could be the much-needed answer to the challenging woes in maxillofacial prosthesis manufacturing procedures, such as lack of man-skill to meet the high demands of prostheses manufacturing.
Additive Manufacturing (AM)AM is a group of technologies that use layer-by-layer additive method to make 3D solid objects from a Computer Aided Design (CAD) model. The CAD model is a 3D digital object designed in any CAD (Figure 1), that can be fabricated in AM machines. The CAD model which constitutes the 3D part geometry is sliced at a thickness permissible for an AM machine to build [5,6].
AM is an amalgamation of many technologies using different principles, materials, and applications. In this study focus is on AM technologies commonly used for manufacturing of models for use in fabrication of external maxillofacial prostheses.
Currently AM is by far the technology that has the most potential to simplify the process of prosthesis manufacturing. The opportunities of this technology have been exploited and the results have been positive in comparison to previous technologies that are time-consuming, costly and required great amount of skill.
Application of Additive Manufacturing in Maxillofacial Prosthesis FabricationAs the number of patients requiring soft tissue prosthesis increases, it becomes evident that traditional prosthesis manufacturing methods are costly and thus new methods of maxillofacial prosthesis fabrication need to be employed in the ever-growing field of maxillofacial rehabilitation using soft tissue prosthesis. The introduction of AM in the process chain of fabricating maxillofacial prosthesis has had a lot of benefits in the manufacturing of the prosthesis. The benefits of that are documented in the various research studies that were done to prove the success of using AM in fabricating maxillofacial prosthesis.
AM is a group of technologies that use layer-by-layer additive method to make 3D solid objects from a Computer Aided Design (CAD) model. The CAD model is a 3D digital object designed in any CAD software such as SolidWorks® [7] that can be fabricated in AM machines. The CAD model which constitutes the 3D part geometry is sliced at a thickness permissible for an AM machine to build.
AM is an amalgamation of many technologies using different principles, materials, and applications. In this study focus is on AM technologies commonly used for manufacturing of models for use in fabrication of external maxillofacial prostheses.
Currently AM is by far the technology that has the most potential
to simplify the process of prosthesis manufacturing. The opportunities of this technology have been exploited and the results have been positive in comparison to previous technologies that are time-consuming, costly and required great amount of skill.
The manufacturing process entails many different AM processes. The main difference between these various processes being the method of material build-up to produce 3D models and the type of material utilised in the machines [8]. Chart 1 shows technologies that have already proved themselves in the medical industry.
The many AM technologies that have proven themselves over the years have resulted in the rise of 3D printers entering the market over the past ten years. With the many machines entering the market each year, selecting the right AM process and machine for fabrication of moulds to create external maxillofacial prosthesis in silicone, that replicates the exact features of the facial anatomy becomes a challenge. Therefore, it is important that thorough investigation is done on determining the best AM process by considering the accuracy yielded by different AM processes and machines selected for each process, and the surface roughness of each printed model, in relation to the desired roughness of the facial skin.
Liquid Based
Solid Based
Powder Based
Material Extrusion
Material Jetting
Vat photo polymerisation
Multi Jet Modelling (MJM)
Thermoplastics Finite Deposition Modelling (FDM)
Photo polymers & Wax
Steriolithography (SLA)
AM Processes
Digital Light Processing
Liquid photo polymer, composites
Liquid photo polymer
Material
Paper, plastic, metal laminates, ceramics
Laminated Object Manufacturing (LOM)
Sheet Laminating
Binder Jetting
Powder Bed Fusion
PolyJet Resin, ABS
Electron Beam Melting (EBM)
Selective Laser Sintering (SLS)
Direct Metal Laser Sintering (DMLS)
Selective Heat Sintering (SHS)
3D Printing
Prometal
Titanium powder
Paper, plastic, metal glass, ceramic composites
Stainless steel, cobalt, chrome, nickel alloy
Thermoplastic powder
Plastics, Nylon, Polyamide
Metal powders
Method Technology
The intestinal microbiome in healthy individuals and patients [22].
Chart 1
2017Vol. 2 No. 3: 27
3© Under License of Creative Commons Attribution 3.0 License
ARCHIVOS DE MEDICINAISSN 1698-9465
Journal of Healthcare Communications ISSN 2472-1654
MethodsThe process chain for the fabrication of external maxillofacial prosthesis (Chart 2), shows techniques on data acquisition using either laser scanning or CT/MRI reconstruction, data manipulation in different 3D CAD software that have proven to have accurate capabilities of creating 3D models from the data acquired of the patient’s facial anatomy, and printing 3D models in AM printers which serves as moulds.
Printing models in AM printer can be achieved by many AM printers currently in the market, but having many processes and machines that are proving the significant impact of AM in medical sector. It has been proven in many cases that AM is one of the most important technologies in the medical sector. However, the full potential of AM cannot be realised without performing analysis of the output quality of products manufactured using various AM processes. Hence quality characteristics of the fabricated part such as accuracy, surface finish, repeatability, and resolution of the geometrical features of the parts produced must be thoroughly assessed to demonstrate the capabilities and limitations of AM processes. Two quality characteristics experiments were done; being dimensional accuracy test and surface roughness test.
Dimensional accuracyAccuracy of a final built model in comparison to the 3D digital model is very important in determining AM machine accuracy, and the capabilities of the machine to build models that will not show significant deformities when subjected to heat and pressure increase.
Surface roughnessSurface roughness is important when dealing with AM printed models, because the printing process involves stepping of layers of materials of a specific thickness that results in difference in surface roughness of models printed using different AM machines. External maxillofacial prostheses must have a certain amount of surface roughness that resembles some roughness of human skin caused by facial contours, lines and pores, as human skin is not perfectly smooth. Surface roughness testing determines the surface roughness of models built using different AM technologies selected for this study, machines and materials used.
To determine dimensional accuracy and surface roughness of AM machines, four processes were selected in the study, those being Stereolithography (SLA), Laser Sintering (LS), Polyjet, and Fused Deposition Modelling (FDM). In each process, step pyramid (Table 1) was designed in SolidWorks® software. SolidWorks® is a CAD platform for drawing and designing models in 2D orientation and 3D orientation (Figure 2). The step pyramid 3D part model was fabricated using AM processes selected for the study and the machines selected in each process (Tables 2 and 3).
Dimensional accuracy testIn each selected process, 3 models were produced from selected machines. The models where scanned using a Kreon Solano with Ace scanning arm (Figure 3), the laser scanner uses reverse engineering techniques of building a model in the digital window by capturing data points during the scanning process. The digital model created from scanning process was used as reference model.
The models were backfilled with yellow-stone plaster, to determine whether plaster back-filling causes changes in model dimensions. Post plaster backfilling, the models were scanned using the laser scanning equipment, and data was generated of the observed number of points at each deviation range.
2.9.1. Data Acquisition
2.9.2. Data Segmentation
2.9.3. Defect Reconstruction
2.9.4. Model Building
Reverse engineering
CT/MRI reconstruction
Laser imaging CT imaging
MRI imaging
Range views 2D Segmentation
3D region growing in MIMICS® [11] 3D registration in Geomagic
3D digital model STL (in Geomagic® [10])
Defect reconstruction in Geomagic® FreeForm
Defect reconstruction in 3-matic [12]
3D digital model STL (in MIMICS®)
3D model slicing
Model building in AM printer
AM detailed process.Chart 2
Model Name Angular step hollow pyramid, with circular, hollow, perpendicular, and rectangular features
Dimensions
Height: 18 mmRadius: 30 mmAngular step height (mm): 0, 3, 6, 9, 12, 15, 18Angular step angles: 20⁰, 30⁰, 40⁰, 50⁰, 60⁰, 70⁰, 80⁰, 90⁰
3D CAD model Include Figure 1 (CAD design of step pyramid in SolidWorks®)
2D SolidWorks drawing
Include Figure 2 (2D drawing of step pyramid in SolidWorks®)
Table 1: Step pyramid dimensions and drawings.
2017Vol. 2 No. 3: 27
4 This article is available in: http://healthcare-communications.imedpub.com/archive.php
ARCHIVOS DE MEDICINAISSN 1698-9465
Journal of Healthcare Communications ISSN 2472-1654
Post backfilling, medical grade silicone was cast in the hollow step region of the models. The silicone could cure, post-curing, the silicone moulds were scanned, to determine observed number of points at each deviation range.
Surface roughness testSurface roughness of each model was measured using the stylus technique. The technique uses a roughness testing device (Mitutoyo SJ-210), the testing device has a stylus which is transversed across the work-piece in a vertical motion (Figure 4) [9]. The motion is converted into an electrical signal, using a transducer. When the signal intensifies, it creates digital information which results in the display values of arithmetic roughness; Ra, Rq, Rz.
ResultsDimensional accuracyThe data obtained from dimensional accuracy testing returned large range of data. The data collected establishes foundation for
determining the output quality of models produced by selected AM machines. However, the data can only give partial judgement of comparing each AM machine’s ability to produce models as
CAD design of step pyramid in SolidWorks®.Figure 1
Machine Type Process Description
EOS P380 LS printing
P380 LS printer is one of the AM printer range manufactured by EOS [13]. It has the ability of processing thermoplastics, without the use of supports which makes printing economical.
Objet500 Connex
PolyJet printing
Objet500 Connex is PolyJet printing technology developed by Stratassys [14]. The printer processes photopolymer using UV light to create models of a large range of materials such as Tango family. The machine builds models with supports.
UP Mini FDM printing
UP Mini is a low-cost printer developed by 3D printing systems [15]. It processes only ABS material by extrusion, to build models with supports.
Fortus 250mc FDM printing
Fortus 250mc is an FDM printing technology developed by Stratasys [16]. The machine has the ability of printing models of fine layers using range ABSplus thermoplastics.
SLA® 350 SLA printing
The SLA®350 by 3D Systems is a thin layer printing machine, producing fine model surfaces in resin, using a laser printing head [17,18].
Table 2: Description of AM processes and machines.
2D drawing of step pyramid in SolidWorks®.Figure 2
Touch probe scanned 3 digital models.Figure 3
Process Machine type Build capacity Build material Layer thickness
LS EOS P380 500 × 380 × 380 mm Nylon PA2200 0.06 mm
SLA SLA® 350 350 × 350 × 400 mm Epoxy 5190 0.05 mm
FDMLow-cost UP Mini 120 × 120 ×
120 mm ABS 0.15 mm
FDM Expensive Fortus 250 mc 254 × 254 ×
305 mm ABSplus–P430 0.17 mm
PolyJet Objet 500 Connex
490 × 390 × 200 mm Digital ABS 0.015 mm
Table 3: AM machines and materials selected for the study.
2017Vol. 2 No. 3: 27
5© Under License of Creative Commons Attribution 3.0 License
ARCHIVOS DE MEDICINAISSN 1698-9465
Journal of Healthcare Communications ISSN 2472-1654
accurate as the CAD model. It is therefore challenging to simply make objective interpretation of the results and conclude which is better. It is important for application of hypothesis testing to better interpret data collected and can provide objective conclusion of results obtained.
Hypothesis testing is a process of using statistics to prove the probability of a certain hypothesis to be true or false. To prove such, there needs to be absolute knowledge derived from examining the entire data collected (termed population).
To conduct hypothesis testing of data that was collected, we considered some statistics terms important for the process. Those being:
- Null hypothesis (H0) is the hypothesis statement that needs to be made, which needs to be proved to being true or false.
- Alternate hypothesis (H1) is the alternate statement made of the hypothesis, which is contrary to the null hypothesis.
- Significance level (α) is the percentage risk of concluding that there is a difference that will exist between population comparisons. It represents how far out of the null hypothesis should a line be drawn to determine accepted values and rejected values. The significance level is usually between 1–5%
- Test statistic is a variable used to determine how close a specific result falls in the hypothesis being tested, by telling whether the null hypothesis is true or false. The Z-score (Z) is often used as the test statistic.
- Z-score (Z): The Z value is determined by using the Z-table wherein the significance level selected is used to determine the test statistic.
- Z- alternate score (Zα) is the alternate Z value which determined using the following equations:
Observed values - Expected valuesZ = Standard deviation
O - EZ = ó
Observed values (O): Number of points collected at each scanning point;
Expected values (E);
( )observed points for model A + observed point for model B × Total no. of observed point for model AE =
Total no. of observed points
( )A B 1O - O × nE =
n
( )n 2ii-1
x -xó =
n∑
xi: % of observed points
x ̅: Mean of total % of observed points
n: Number of points
Application in our studyGiven the terms and equations above, when applied in our study, the following hypothesis test was done:
H0=0 this is a hypothesis made that the models printed using the selected AM machines are the same as the CAD model and that there is no displacement at all. Therefore, we must prove whether the hypothesis is true or false
H1 ≠ 0 is the alternate hypothesis, meaning none of the models obtained will be the same dimensions as the CAD model
α=0.02 we only allowed a percentage error of 2% in our results
Z value–To determine the Z-value, given the significance level of 0.02 (0.01 for each side), a two-tier graph (Figure 5) is considered where the critical point is determined by using the Z-table (Figure 6).
To determine (Zα), equation 1.1, 1.1.1 and 1.1.2 are used to determine the value using data collected for the scanning process and comparing reference models with physical models before casting silicone and after casting silicone.
The results of (Zα) were computed using data from scanning models. Tables 4 and 5 gives summary of Zavalues that fall in the non-rejection region.
A computation of the γ-square test and γ-square statistic (denoted X2) which is the measure of how expectations compare to the results shows that our expectations were false. Our expectations which are described using the Null hypothesis wherein we expected that H0=0 for the result comparison of the reference model to models after plaster backfilling and after casting silicone is not true. Therefore, the Chi-square statistic is 0.
Although the Null hypothesis was rejected using Chi-square test, the Z-test proves that there are certain areas where the Null hypothesis is not rejected. Considering the significance level of 0.02 which is the percentage error that was assumed to exist in the results comparison; it is shown in Table 5, that the Null hypothesis is not rejected mostly at areas where there is less concentration of observed number of points.
One of the important observation that is made from the (Zα) values obtained and the conclusions made is that there are
Surface roughness testing procedure.Figure 4
2017Vol. 2 No. 3: 27
6 This article is available in: http://healthcare-communications.imedpub.com/archive.php
ARCHIVOS DE MEDICINAISSN 1698-9465
Journal of Healthcare Communications ISSN 2472-1654
significant changes that happened in the models post-casting, this change can be seen particularly in the population distribution in each process. In the reference model, the population distribution of observed number of points is where the deviation is between -0.05 and 0.05 mm. After backfilling, the population is slightly spread to deviation ranges between -1.0 and 1.0 mm. After casting, the population spread is between deviation ranges -2.0 and 2.0 mm [10-18].
Surface roughness The results of surface roughness arithmetic characteristics obtained from experiment are represented in Table 6, showing the average values of the characteristics determined (x ̅) and the standard deviation of the 3 values obtained at each angle (σ).
The results obtained for surface roughness of models produced using selected AM processes and machines show that Polyjet process produces smooth output products than all other processes. The SLA process has intermediate surface roughness (neither too rough nor too smooth). All the other processes and machines used, resulted in high surface roughness result.
The reason behind the high surface roughness in FDM process and LS process is the material building method. The extrusion method used in FDM processes results in step creation, the steps could be of different angles, depending on the input given to the machine during build orientation. The steps are more significant in the FDM Low-cost machine because of the high minimum layer thickness.
From the average surface roughness results obtained, the roughness grade can be computed from the standard roughness
Z-value graph.Figure 5
Standard normal probabilities table.Figure 6
Table 4: Za values for plaster backfilled moulds compared to reference model.
2017Vol. 2 No. 3: 27
7© Under License of Creative Commons Attribution 3.0 License
ARCHIVOS DE MEDICINAISSN 1698-9465
Journal of Healthcare Communications ISSN 2472-1654
Table 5: Za values silicone cast moulds compared to reference model.
Process Angle Rα Rq Rz
Polyjet
x ̅ (µm) σ (µm) σ x ̅ (µm) σ10⁰ 2.134 0.043 2.476 0.027 7.912 0.50420⁰ 2.263 0.020 2.654 0.020 9.057 0.04630⁰ 2.052 0.011 2.464 0.006 8.455 0.01440⁰ 2.143 0.043 2.546 0.060 9.105 0.31850⁰ 2.062 0.045 2.512 0.053 9.409 0.05060⁰ 2.180 0.005 2.945 0.021 9.983 0.25670⁰ 2.092 0.586 2.658 0.725 10.283 2.32880⁰ 9.096 0.109 10.343 0.118 33.771 0.39290⁰ 2.276 0.007 2.777 0.010 10.015 0.190
SLA
10⁰ 11.787 0.204 12.670 0.472 55.064 0.37620⁰ 19.333 0.126 21.291 0.343 84.996 0.26630⁰ 14.623 0.097 15.198 0.084 59.047 0.10140⁰ 13.468 0.012 14.716 0.863 51.939 1.23850⁰ 7.350 0.198 8.586 0.030 32.701 2.43560⁰ 4.330 0.450 6.047 0.455 20.991 1.89470⁰ 3.378 0.500 4.184 0.071 16.852 2.25780⁰ 2.897 0.390 4.078 0.154 13.742 1.54490⁰ 2.473 0.263 3.172 0.028 12.947 1.293
LS
10⁰ 42.455 1.522 43.345 1.575 238.135 5.30020⁰ 50.686 2.377 52.182 1.972 260.505 3.22030⁰ 41.222 2.454 41.743 2.651 239.926 7.80540⁰ 25.021 1.973 25.932 2.307 146.617 13.18350⁰ 22.464 1.152 23.104 1.050 134.957 6.36660⁰ 21.504 2.457 22.119 2.768 132.121 3.28870⁰ 23.140 1.116 23.290 1.335 141.308 1.70380⁰ 14.417 0.443 15.337 0.327 86.085 3.21790⁰ 15.235 0.674 16.140 0.758 90.813 5.130
Table 6: Summary of surface roughness results.
2017Vol. 2 No. 3: 27
8 This article is available in: http://healthcare-communications.imedpub.com/archive.php
ARCHIVOS DE MEDICINAISSN 1698-9465
Journal of Healthcare Communications ISSN 2472-1654
Process Angle Rα Rq Rz
FDM Low Cost
10⁰ 37.855 0.602 33.928 0.614 235.375 12.23620⁰ 35.739 0.332 36.585 0.848 241.699 1.62830⁰ 29.951 0.149 31.165 0.863 234.023 3.35040⁰ 25.880 0.591 27.032 0.571 129.708 1.89050⁰ 28.626 2.631 29.716 1.510 181.618 25.43260⁰ 27.042 1.026 23.546 1.160 185.889 4.54870⁰ 21.030 1.622 22.507 1.965 146.681 16.51480⁰ 15.259 0.156 16.243 0.189 91.006 1.13390⁰ 16.033 0.514 16.784 0.622 107.970 8.550
FDM Expensive
10⁰ 20.042 0.530 20.984 0.583 119.083 3.36620⁰ 36.666 0.115 37.304 0.038 206.131 10.92330⁰ 30.070 2.481 31.296 2.934 177.624 12.23140⁰ 22.560 0.705 23.552 0.444 222.104 11.22350⁰ 25.026 2.747 26.102 2.831 215.766 10.58260⁰ 22.928 1.495 23.878 1.460 219.785 14.28070⁰ 21.567 0.679 22.626 0.420 205.716 10.68080⁰ 19.204 1.181 18.820 1.335 178.419 9.91690⁰ 18.258 1.122 19.006 1.203 134.969 7.235
Average Roughness values (Rα) µm
Roughness grade number
Roughness triangle symbols
50 N1225
12.5N11N10
6.33.21.6
N9N8N7
0.80.40.2
N6N5N4
0.010.05
0.025
N3N2N1
Table 7: Standard average roughness grade table.
Process Average Roughness (Rα) µm
Roughness grade number
Roughness triangle symbol
Polyjet 2.922 N8
SLA 8.850 N10
LS 28.46 N12
FDM Low Cost 26.38 N12
FDM Expensive 24.035 N11
Table 8: Average roughness grading for AM machines.
grade table (Table 7). Table 8 shows the roughness grading for each process, based on the average surface roughness data obtained in the study.
ConclusionThe results obtained, determining the accuracy of models produced using select0065d AM machines, show that AM machines have the capabilities of producing accurate models.
However, models produced using the low-cost AM machine did not produce the expected accuracy in the models as desired, this is considering the (Zα)values being rejected of the post-cast model in comparison to the reference model. Usually accuracy of models is attributed to a range of factors, such as build orientation especially in FDM processes where there is a filament extrusion; the orientation of the model on the build platform can affect the overall accuracy. Another factor is the minimum layer thickness that the machine can print in; the low-cost machine has the highest minimum layer thickness compared to all the other selected machines. The layer thickness affects the overall strength of the model because there is a high chance of the model being brittle.
SLA and FDM-expensive machine show consistency in population distribution after plaster backfilling and after silicone casting, this is considering the distribution of observed points recorded and the areas where the null hypothesis is not rejected. Therefore FDM and SLA processes are more accurate in our study.
Based on the surface roughness grading of each process and machine, we can conclude that SLA process yields better surface roughness for our study because the roughness resembles that of skin. However, FDM models can be chemically furnished to create a desirable surface roughness that replicates the porosity of skin, but further studies should be made to determine to what degree of surface furnishing needs to be done on models, to achieve the desired surface roughness that replicates that of the facial skin, so that when silicone is cast in the mould, it can result in a final prosthesis that looks like a natural facial feature.
From the results obtained and observations made, FDM proves to be a process that is favoured to produce moulds to be used for casting silicone for external prosthesis fabrication.
Future WorkThe rapid growth of the 3D printing industry has seen the
2017Vol. 2 No. 3: 27
9© Under License of Creative Commons Attribution 3.0 License
ARCHIVOS DE MEDICINAISSN 1698-9465
Journal of Healthcare Communications ISSN 2472-1654
introduction of low-cost printing machines that yield accurate models, it is therefore important to consider some of the low-cost 3D printing machines that are currently being used in the medical sector for prosthesis fabrication.
Although the study focused on geometrical accuracy of AM machines, it would be important to extend the study to consider performance analysis of AM machines, considering the cost and time because they are factors that determine which process can be favoured over the other.
Surface roughness of models is important in determining which process and machine is selected for producing moulds, however surface roughness of moulds can be enhanced using finishing techniques. Although, such finishing techniques are not discussed in the study, it would be good to determine the possible surface finishing techniques that can be used to enhance the roughness of models to replicate that of human skin.
2017Vol. 2 No. 3: 27
10 This article is available in: http://healthcare-communications.imedpub.com/archive.php
ARCHIVOS DE MEDICINAISSN 1698-9465
Journal of Healthcare Communications ISSN 2472-1654
References1 Eggbeer D (2008) The computer aided design and fabrication of
facial prostheses.
2 Bibb R, Eggbeer D, Evans P (2010) Rapid prototyping technologies in soft tissue facial prosthetics: current state of the art. Rapid Prototyping J 16: 130-137.
3 Eggbeer D, Bibb R, Evans P, Ji L (2012) Evaluation of direct and indirect additive manufacture of maxillofacial prostheses. Proc Inst Mech Eng H 226: 718-728.
4 http://www.intechopen.com/source/pdfs/20116/InTech-Medical_applications_of_rapid_prototyping_a_new_horizon.pdf
5 http://scholar.google.com/scholar?hl=en&btnG=Search&q=intitle:Additive+Manufacturing+:+Status+and+Opportunities#0\nhttp://scholar.google.com/scholar?hl=en&btnG=Search&q=intitle:Additive+manufacturing:+Status+and+opportunities#0
6 http://web.iitd.ac.in/~pmpandey/MEL120_html/RP_document.pdf.
7 http://www.solidworks.com/
8 Brennan J (2010) Production of anatomical models from CT scan data. Faculty of Technology, De Montfort University, UK.
9 http://3dprintingsystems.com/products/3d-printers/up-mini-3d-printer/
10 http://www.geomagic.com/en/
11 http://www.materialise.com/en/medical/software/mimics
12 http://www.materialise.com/en/software/3-matic
13 https://www.eos.info/en
14 http://www.fr-3d.com/documents/InTechOptimization_of_additive_manufacturing_processes_focused_on_3d_printing.pdf
15 http://3dprintingsystems.com/products/3d-printers/up-mini-3d-printer/
16 http://www.stratasys.com/3d-printers/technologies/fdm-technology
17 https://www.3dsystems.com/3d-printers
18 http://surfacefinishequipment.com/mitutoyo-sj-210.html