You are on page 1of 29

Manuscript Details

Manuscript number ADDMA_2017_168

Title Computer aided reverse engineering system in the design and


manufacturing of the insole shoe orthotics for diabetes patient

Article type Research Paper

Abstract
Design and manufacturing strategies of custom insole shoe orthotics for diabetes patients are presented in
this paper. The computer aided reverse engineering system (CARESystem) was applied here for minimizing a
file error generated by scanning of the foot patient. The digital data imaging and communication of medicine
(DICOM) images were used for designing of complex insole orthotics. The DICOM images obtained by
HandySCANN 700TM were then converted into Stereo Lithography (STL) file (triangle based data) format. The
color mapping generated by VX element and Geomagic software yielded an accurate and smooth STL file data
of the patient's foot, and provided in the 3D solid model of orthotics shoes (insole and outsole) using curve
base surface modelling method (CBS Model). The error of an image found on the stage of 3D mesh STL file
was then edited and smoothed, prior to the 3D prototype products being processed by 3D printers. The
prototype of insole shoe orthotics was fabricated by an adaptive manufacturing machine of 3D printer
EDEN350. The comparative analysis of the insole and outsole dimensional data provided the average size of a
deviation of less than 1 mm (0.0135 - 0.0209 mm), therefore it is very promising method to deliver the custom
shoes orthotics with improved size dimensional accuracy and comfort.

Keywords Customized shoe insole; Diabetes; DICOM; STL file; CARESystem

Corresponding Author Athanasius P. Bayuseno

Corresponding Author's Diponegor University


Institution

Order of Authors Paulus Wisnu Anggoro, Baju Bawono, M.M Avelina, J Jamari, Athanasius
P. Bayuseno

Submission Files Included in this PDF


File Name [File Type]
Cover Paper (JME).doc [Cover Letter]

Manuskrip JME(final).doc [Manuscript File]

CAPTION FIGURES.docx [Figure]

Tabel (CErese).doc [Table]

HIGLIGHTS.doc [Highlights]

To view all the submission files, including those not included in the PDF, click on the manuscript title on your
EVISE Homepage, then click 'Download zip file'.
Research article- submitted for publication in journal of Additive Manufacturing

Computer aided reverse engineering system in the design and

manufacturing of the insole shoe orthotics for diabetes patient

P.W. Anggoro1, Baju Bawono2, M.M Avelina3, J.Jamari4, A.P. Bayuseno5*


1,2,3Department of Industrial Engineering, Faculty of Industrial Technology, University of Atma

Jaya Yogyakarta, Jl. Babarsari 44, Yogyakarta 55281, Indonesia.


1,2,4,5Department of Mechanical Engineering, University of Diponegoro,

Jl. Prof. Soedarto, SH., Tembalang, Semarang 50275, Indonesia.

Email : 1 pauluswisnuanggoro@ymail.com (+62-857 294 99 575)


5apbayuseno@gmail.com (+62-813-9012-6407)

____________________________________________________________
*Corresponding author at the Department of Mechanical Engineering, Diponegoro University,

Tembalang Campus, Semarang 50275, Indonesia. Tel: +62-24-7460059; Fax: +62-24-7460059.

E-mail address: apbayuseno@gmail.com (A.P. Bayuseno)


1. Submission declaration

We confirm that the article submitted to this journal is original and has not been published

previously; that it is not under consideration for publication elsewhere. Its publication is also

approved by all authors and by the responsible authorities tacitly or explicitly where the work

was carried out. If accepted, it will not be published elsewhere in the same form, in English or in

any other language, including electronically without the written consent of the copyright-holder.

All the authors mutually agreed that it should be submitted to this journal.
1 Computer aided reverse engineering system in the design and

2 manufacturing of the insole shoe orthotics for diabetes patient

3
4 P.W. Anggoro1, Baju Bawono2, M.M Avelina3, J.Jamari4, A.P. Bayuseno5
5 1,2,3Department of Industrial Engineering, Faculty of Industrial Technology, University of

6 Atma Jaya Yogyakarta, Jl. Babarsari 44, Yogyakarta 55281, Indonesia.


7 1,2,4,5Department of Mechanical Engineering, University of Diponegoro,

8 Jl. Prof. Soedarto, SH., Tembalang, Semarang 50275, Indonesia.


9 Email : 1 pauluswisnuanggoro@ymail.com (+62-857 294 99 575)
10 5apbayuseno@gmail.com (+62-813-9012-6407)

11
12 ABSTRACT- Design and manufacturing strategies of custom insole shoe orthotics for

13 diabetes patients are presented in this paper. The computer aided reverse engineering system

14 (CARESystem) was applied here for minimizing a file error generated by scanning of the foot

15 patient. The digital data imaging and communication of medicine (DICOM) images were

16 used for designing of complex insole orthotics. The DICOM images obtained by

17 HandySCANN 700TM were then converted into Stereo Lithography (STL) file (triangle based

18 data) format. The color mapping generated by VX element and Geomagic software yielded

19 an accurate and smooth STL file data of the patient's foot, and provided in the 3D solid model

20 of orthotics shoes (insole and outsole) using curve base surface modelling method (CBS

21 Model). The error of an image found on the stage of 3D mesh STL file was then edited and

22 smoothed, prior to the 3D prototype products being processed by 3D printers. The prototype

23 of insole shoe orthotics was fabricated by an adaptive manufacturing machine of 3D printer

24 EDEN350. The comparative analysis of the insole and outsole dimensional data provided the

25 average size of a deviation of less than 1 mm (0.0135 - 0.0209 mm), therefore it is very

26 promising method to deliver the custom shoes orthotics with improved size dimensional

27 accuracy and comfort.

28 Keywords: Customized shoe insole; Diabetes; DICOM; STL file; CARESystem

29

30

1
31 1. Introduction

32 The human foot frequently suffers plantar pressure distribution during performing

33 physical activities such as standing, playing, and walking. This condition may lead to injury

34 risk, which can be minimized through wearing the personal footwear with optimum fit and

35 comfort. Additionally, the human foot has very complex geometry [1]. Therefore, people

36 may have different foot shapes, of which some can suffer foot deformities caused by

37 diabetes. This disease has the highest potential to damage of the organs foot [2, 3, 4]. It

38 was described by WHO (World Health Organization) in 2009 that, at least 289 million people

39 experienced diabetes and this number may increase to 438 million people in 2030. Likewise,

40 they may suffer foot deformities such as pronation, Metatarsalgia, flat feet, neuroma, plantar

41 fasciitis, arch pain [5-8]. Moreover, these deformities make often certain pain and discomfort

42 in foot when people wearing shoes or sandal. The common complication sickness of suffering

43 diabetic is due to the foot ulceration [5]. Many factors may promote the foot ulceration, as

44 results of elevated temperature/humidity within a shoe, an increased in-shoe plantar pressure

45 (pressure under the foot) [2,3] and an increased shear movement between the foot and the

46 shoe. If the diabetic foot is untreated, it can cause infection and ultimately the patient must

47 undergo total or partial foot amputation [1].

48 Further, the foot ulceration of a diabetic patient may be healed through the use of

49 custom-made foot orthoses [1], whilst the good fitted-footwear model can help prevent

50 ulcerations and amputations. Thus far, the foot orthotics can be implemented for pain relief,

51 increasing heel cushion, correction of flexible deformity, increasing foot stability and/or

52 prevention of skin breakdowns [2]. Accordingly, the custom-shoe orthotics are considered as

53 the promising method for foot pain treatment and lowering limb pathology, hence providing

54 the advantages of individualized prescription for designing and requiring the personalized

55 footwear with good fit and comfort [3].

2
56 Correspondingly, much attention has been paid on development of custom foot

57 orthotics in the manufacture of shoes and sandals [3]. However, the difficulty in the design

58 and fabrication of the customized foot orthotics has been usually found due to variations in

59 orthotic device design and fabrication. Here, different aspects of design, such as insole

60 geometry [4], outsole profile/stiffness [9-10] and shoe lacing contributing to each of the foot

61 ulceration [11] must be based on prescription and non-prescription foot orthoses.

62 Furthermore, the danger of trauma due to ulceration can be minimized by designing

63 appropriate footwear for specific foot shape and biomechanical foot functioning, and this is

64 established on an individualized prescription of the diabetic patient [12-13]. Conversely, the

65 invention process for custom orthotics products are frequently made manually in a handicraft

66 industry, where technicians with high skill and having experience are needed [6,15,17]. This

67 approach leads time-consuming process, while the product with a less accuracy and worst

68 surface finish may be obtained and therefore affect significantly the discomfort in foot of the

69 patient [4].

70 The best practice method for obtaining the optimal design of custom foot orthotics is

71 to use computer aided design (CAD), which spend less time time in design and reduce in

72 manufacturing cost [9]. With the availability of CAD packages, reverse engineering (RE) has

73 now emerged as a practical method to obtain an object without data engineering drawings,

74 documentation, as well as computer models [10,11]. The use of 3D (three-dimensional) CAD

75 makes RE shorten the design time and minimize the costs involved during the process of

76 manufacturing [12]. In this case, the obtained input data of the scanning process contains the

77 detail and accurate model. Additionally, RE was successfully adopted for fabrication of the

78 insole shoe orthotics made by Eva-rubber for normal people's feet and subsequently

79 manufactured by subtractive manufacturing machines [Rolland model MDX 40R]. This

80 process provided the best fitted-insole product [16], which was quite similar to the surface

3
81 contour of the human foot obtained from the process of scanning. However, the typical error

82 on the surface quality of the insole shoe orthotics could be found due to differences in the

83 size of the physical model and the prototype or the lost data of the inaccurate scanning results

84 or CAD output.

85 In contrast to the subtractive manufacturing process, rapid prototyping (RP) is a layer-

86 by-layer manufacturing technology commonly employed for medical model [1, 14, 19, 20].

87 In this case, the RP model of human anatomy can be generated from high resolution digital

88 images such as computed tomography (CT), magnetic resonance imaging (MRI) and any

89 other medical scanning processes [21]. The acquired medical data set from the medical

90 scanner is subsequently segmented into 2-Dimensional (2D) pictures. These pictures are

91 usually kept in digital imagery and communication in medicine (DICOM) file format

92 (voxels), which can be acquired in 3D-biomedical computer aided design (Bio-CAD) model.

93 This Bio-CAD model can be subsequently exported into stereo lithography (STL) file for

94 fabrication of RP medical model [22]. Nevertheless, the STL file format has some

95 dimensional error, specially in the RE model of human figure. The subsequent improvement

96 of quality in STL file of 3D imaging data can be performed.

97 Further improvement of quality of imaging data in STL file for RP medical models

98 could be attained by the accurate DICOM images [19]. The images can be converted into an

99 STL file (triangle based data) format, which is subsequently used for the manufacturing of

100 medical models using RP technologies. Since inaccuracy and rough surface of the RP model

101 was often generated due to the erroneous in STL file, a fast Fourier transform (FFT) method

102 was proposed to improve the surface contour data points, which can generate a surface and

103 STL file simultaneous by the delay triangulation method. This novel combination method

104 provided the generated STL file in accuracy and smoothness, compared to the initial STL file.

105 It is proposed in term of the present study that the STL files of the RP insole model may be

4
106 changed into the triangulation angle by this algorithm for trimming and hole-filling

107 operations.

108 Accordingly, advanced design methodologies in term of computer aided reverse

109 engineering system (CARESystem) have been proposed to ease the acquisition of design

110 knowledge and creative ideas through reverse innovative design (RID). RID can cover a wide

111 range of 3D digital design usages, such as the 3D digitization, the 3D CAD and CAID

112 (computer-aided industrial design), reverse engineering (RE), CAE (computer-aided

113 engineering) analysis and RP [1,6,14-17]. Recent studies have demonstrated the capability of

114 RID on the design of innovative footwear product [1, 6, 14-15, and 17] using the 3D image of

115 a patients foot and subsequent use this image for manufacturing shoes orthotics. Here RID

116 can develop and deliver a product with family design variation, [1] and allows adjustment of

117 a personalized shoe product. This systematic approach offers the benefits of obtaining

118 optimum results of the existing design and reducing the error rate in manufacturing products.

119 Nevertheless, there is only a limited systematic study based the RID methodology in

120 designing the customized shoes for the diabetic patient group. Thus, the design and improved

121 manufacturing process for the insole shoe orthotics based on the RE model of diabetic

122 patients is very crucial [13]. The design of specific diabetic footwear for increased comfort

123 requires automated systems as CARESystem.

124 The present study was undertaken to apply the beneficial innovative method of

125 CARESystem to produce a prototype insole and outsole fitted the foot shape of the diabetic

126 patient. The stages of this research were initiated by the foot scanning with HandySCANN

127 700TM, and STL files were verified with VX Element. Later, the PowerSHAPE 2016

128 software received a fix 3D CAD foot model with an error of less than 1 mm. The prototype

129 orthotic models were manufactured by the 3D printer machine EDEN350V. These important

5
130 steps were expected to be a standard method for constructing the custom insole orthotics with

131 lower production cost and improved quality product.

132

133 2. Materials and methods

134 2.1 Physical model of 3D digitizing and mesh processing

135 Two women participants of age around 50-75 years were selected for this research.

136 Selection of both patients was based on consideration of the long history suffering diabetes

137 for more or less 10 years and resulted in a sickness when the patient is wearing normal

138 sandals or shoes. Similar to the previous research [3,15,17], the RID was adopted here for the

139 design process of the insole shoe orthotic, which involves an integrated digital design

140 methodology incorporating digitizing, modeling with shape parameters and product

141 definition, analysis for CAE-based product optimization and RP models [3]. The 3D data

142 acquisition process was developed by physical model and the phase of mesh processing using

143 a non-contact the 3D scanners of HandySCAN 700TM equipped by Vx element software.

144 This scanner tool produced an output data in the form of STL file, which is insensitive to

145 vibration, and does not stick to the object. The technology uses white light or laser making

146 data processing faster in the ranges of 5-10 minutes, and the utilization of lasers can produce

147 accuracy levels of upward to 5 micrometers. The 3D scanner is equipped with two cameras

148 and seven types of lasers, which can retrieve the data objects form and reach a small gap of

149 image.

150 Furthermore, the HandySCAN 700TM was chosen because it has seven intersecting laser

151 beams, which enable to speed up shooting with high precision of about 0.030 mm (0.0012

152 in). This instrument delivers a single function laser that can be employed to arrive at the

153 infantry of a gap on the object of a complicated and complex surface contour. It is also

6
154 equipped with a laser beam for the x-y value with an accuracy of up to 30 microns, and z of 5

155 microns.

156 Correspondingly, the 3D CAD data of RE foot were initially verified prior to be

157 processed further. An indispensable component of the 3D mesh foot was marked from the

158 ball of foot area to the heel (the component that is in direct touch with the insole). The

159 verification was identified by a color mapping (Fig.1). This verification of the object enabled

160 the distribution of the size difference between the data of the scanning results in the form of

161 the 3D mesh and results of RE to be evident in the color symbol.

162 Furthermore, the color mapping was analyzed by comparing the 3D CAD surface with 3D

163 mesh STL with the same orientation. Comparative analysis of feature on VX Element 5 and

164 PowerSHAPE 2016 was also performed in the present work. This feature took the sample

165 points on the two images being compared and then calculated for the gap between the dots

166 and displayed in shades of red to blue. Correspondingly, the software of Vx Element 5

167 provided the 3D scanning to capture data directly (real time) from 3D scanner of

168 HandySCAN 700TM which was displayed on the computer screen. This software consists of

169 modules element Vx, Vx Inspect, and Vx models. Vx Element 5 can retrieve data and

170 perform 3D scanning process to check whether the mesh processing of obtaining 3D solid

171 models was fitted to the foot.

172 Additionally, the custom insoles required the high accuracy of mold, plumbing and

173 machining tools. Comparative analysis of using VX Element 5 can verify the output 3D

174 scanning. If the average deviation of less than 1 mm or 50% of the deviation is under 1 mm

175 then the data can be accepted for further processing. This means that the data of STL file are

176 fixed and could be used for the subsequent CBS (curve-based modeling) process to produce

177 insole and outsole shoe orthotic. Results of color mapping in VX Element 5 can be presented

178 in Table 1.

7
179 2.2 Modeling from 3D mesh foot to 3D solid of insole and outsole shoes orthotics

180 The 3D solid model of the foot patient in the VX Element 5 was then required for the

181 establishing process of a wireframe curve insole starting from the stage of processing oblique

182 to the generation of 3D surface editing for an existing curve. There are three modeling

183 strategies of RE widely known [1,3], namely automatic surface modeling, featured-based

184 solid modeling and CBS modeling. The CBS modeling was selected in this work because it

185 could reconstruct the 3D model of the OIS (orthotic insole shoe) with highly precise and

186 accurate result. This scheme has been successfully adopted on manufacturing insole shoe

187 orthotics for people with normal and foot deformities [6, 15-17]. Here, the CBS modeling

188 and smart features in PowerSHAPE 2016 yielded the 3D surface form using the created

189 curves and provided the reconstructed surface with high accuracy.

190 Furthermore, stages of CBS Modeling in PowerSHAPE 2016 included: (a) mesh

191 importing and pre-processes, (b) rewiring, (c) repoint, built and verification of 3d surface to a

192 solid foot with a solid doctor, (d) a 3D solid model of the foot from the mesh, (e ) oblique

193 processing, (f) foot wireframe, (g) support wire, (h) repoint wireframe curve, i) wire

194 reconstruction, (j) generating surface, (k) surface curve editing, and (l) 3D solid iso diabetes.

195 Results of the 3D solid model of OIS for the diabetic patient show with a variety of width

196 tolerances. A new optimal design of OIS with 2:00 mm width tolerance of 3D solid was set-

197 up according to the previous OIS made by RP model [15, 17]. Similarly, the outsole section

198 was performed, but wide tolerances between the insole and outsole was of about 0.75 mm.

199 2.3 Manufacturing process of the RP Model

200 A pair RP of OIS models was fabricated by an adaptive manufacturing technology with

201 3D printers Object EDEN 350 V, which contains the engine of 3D printers by Stratasys

202 released with a tray of 350 x 350 x 200 mm. This machine uses PolyJet technology that

8
203 allows the material expenditure with a thick layer up to 16 microns. It has an accuracy rate of

204 0.1 mm to 0.3 mm depending on the geometry, orientation, and size of the product.

205 Furthermore, two kinds of materials and two support of each 3.6 kg were labelled by

206 VeroBlue material, VeroWhitePlus, VeroBlack, TangoGray, TangoBlack, TangoBlackPlus,

207 FullCure720 Transparent, and FullCure Support. Material prototype of the customized insole

208 orthotics was assigned as OIS VeroWhitePlus RGD835. The plastic material is one of the

209 materials suggested by Stratatys in the operation of Eden 350V 3D print. The material

210 properties of the polymer are presented as follow: tensile strength of 50-65 MPa, elongation

211 at break of 10-25%, modulus of elasticity of 2000-3000 MPa, flexural Strength of 75-110

212 MPa, flexural modulus of 2200 to 3200 MPa, HDT (heat deflection temperature), OC @ 1.82

213 MPa 45-50 C, water absorption of 1.1-1.5%, Shore hardness of 83Scale-86 D, Rockwell

214 hardness Scale M of 73-76, 1:17 to 1:18 polymerized density (g/cm3), 0:23 to 0:26% ash

215 content (VeroGray, VeroWhitePlus), 0:01 to 0:02% (VeroBlackPlus) [18].

216

217 3. Results

218 The multiple images of 3D foot models resulted from the output scanning, verification of

219 the VX Element and prototype OIS products are given in Figures 2-6. The form of 3D mesh

220 foot (Fig.2) was subsequently justified for insole shoe and outsole orthotic models. The VX

221 software-Element 5 directly connected to the computer and Handyscan 700TM could

222 greatly simplify and speed to optimize 3D solid model of RE (Fig.3). Table 1 presents

223 results of the standard deviation for both pairs of legs diabetic patients with an average

224 deviation less than 1 mm. Output file of RE was then converted into a pair of insole and

225 outsole pair using methods CBS modeling[1] with a variation of width tolerance. The optimal

226 width tolerance of the previous research [15,17] employed in this study provided a good fit

9
227 with the patient's feet on the second insole, of which assembly with wide legs of 2.00 mm

228 tolerance can be shown in Figure 4.

229 Moreover, the RP models of the insole, outsole and the foot of a patient could be formed

230 in terms of all solid 3D images of insole, outsole and foot, and subsequently verified by the

231 open source system software [Autodesk Netfabb 2017]. The use of software made significant

232 improvements and optimizing STL file before being put into the 3D printer machine. The

233 need for this optimizing STL file was mainly directed to the failure reduction, when the file

234 was inserted into the machine tool (EDEN 350V). Consequently, the manufacturing products

235 of RP models can be shown in Figures 5 and 6.

236 The RP models of OIS product were then verified for the precise dimensional size and

237 suitability of fit between the physical model and the 3D CAD models using a measuring

238 instrument vernier caliper with 0:01 mm tolerance. The measurements of RP models were

239 performed by determination of area dimensions in the CAD drawings (the measured point)

240 given in Figure 7. The results of dimensional verification of RP models in CAD drawings and

241 prototype products of OIS show an average size difference of less than 0.5mm (Tables 2 and

242 3). Also, there is the relatively small deviation in dimensional size for prototype having

243 deviated by more than 1mm. This means between the CAD drawings of scanning results

244 and color mapping of a prototype are in good fitting of a dimensional model. Therefore,

245 the results of the prototype can be confirmed in accordance with existing CAD

246 drawings.

247

248

249

250

251

10
252 4. Discussion

253 The present study has demonstrated the excellent method in the design of OIS product

254 through the 3D mesh foot obtained from the 3D scanning process. The process was carried

255 out by the infrared laser beam of HandySCANN 700TM on the foot surface required at least

256 ten minutes for a pair of legs. The results of the 3D scanning process are in form of 3D

257 mesh STL files. The 3D mesh STL or triangle mesh is a set of the custom surface of

258 dots that form a small triangle. This small triangle can be linked to form a custom

259 geometry of an object. Data triangle mesh is more comfortable for RE because the

260 shaped surface can be outfitted with the foot in term of cloud points not dots. This

261 procedure allows the RE to be quicker because it does not need to be converted into

262 dots point cloud of STL. The 3D scanning process was performed in the present

263 research to get data insoles of the base up to the ankle area. Foot area is an important

264 part because it is a direct contact with the footwear. Arches and the part of foot surface

265 were required to form the 3D insole orthotics, which can be obtained from the 3D

266 scanning process. The results of the 3D mesh STL file could be in conformity with the

267 original shape of the foot. Correspondingly, the level of accuracy below the tolerance

268 limit of 1 mm could be achieved in the present study. This important file can be

269 employed in the subsequent machining with CNC machine for an insole or outsole

270 material such as EVA rubber foam and polyurethane.

271 Further, the CAD technology has some limitations for image processing with organic

272 contour shape of the foot such as diabetic patients with high-risk scale. This condition

273 resulted in frequent errors when doing image editing foot using the solid feature. The CAD

274 software cannot extract the surface of the foot 3D mesh perfectly with the features of shrink

275 wrap. Nevertheless, the benefits of CAD can be explored for feature comparison analysis

11
276 between the 3D mesh and 3D CAD insole. The results of this analysis can allow researchers

277 to take a decision whether the RE process can be continued or not.

278 Additionally, the function of Vx Element 5 can simplify the analysis data of RE with

279 PowerSHAPE2016. The 3D mesh can be changed by automatic surfacing clicking on surface

280 features into the 3D surface of the mesh so that it can be read and analyzed in PowerSHAPE

281 2016. The benefit of Vx Element 5 to process the RE model of patients with foot deformities

282 has been also reported previously [6, 8,15-17]. In the present research, the process of

283 extracting the 3D surface of the mesh to solid by this software yielded an average deviation in

284 less than 1 mm. Therefore the program enables to enter the optimal surface mesh data for the

285 manufacturing process.

286 Based on the color mapping, it can be seen that the results of RE are green, which indicates

287 that there is a small deviation. It appears that the improvement in the patient's leg 1 produces

288 an average deviation of 0.1395833 to 0.0937500 mm with an error range -0.3756 to 0.6413 in

289 left foot and right foot of -0.3707 mm to 0.5926 mm (Table 1). The color mapping for patient

290 2 indicates an average deviation of 0.1326389 up to 0.1451389 mm with an error range left

291 foot -1.6146 to 0.6066 mm and the right foot of - 0.7644 to 0.5640 mm (Table 1). The results

292 of RE are appropriate and can be further processed in an RP machine using 3D printer

293 EDEN350V to obtain optimal new OIS products. The mean deviation image RE results in

294 Table 1 gives the figure a deviation of less than 1 mm. This indicates that the level of

295 accuracy of the image described in this paper is high and automatic surfacing results are in

296 accordance with the scanning form of physical models.

297 Once 3D models of the fixed foot are obtained, the insole and outsole shoe orthotics can

298 designed based on CBS modeling [1, 6, 14, 15 and 17]. The 3D model of the new insole and

299 outsole shoe orthotics can be successfully produced by CBS modeling. Also the CAE

300 analysis with a width tolerance of 0.75 mm frequently provides the optimal design of insole

12
301 product for diabetic patients [6, 17]. In this study, the optimal design was then constructed

302 using the 3D print EDEN350V yielding the prototype foot, insole and outsole of special

303 customized OIS.

304 The verification process of size dimensions in the image STL file, 3D CAD models and

305 prototype RP models were conducted similarly to the previous RP medical models [21]. The

306 outcomes of this study equally demonstrate the value of the dimensional error of less than 1

307 mm, while the mean error in the initial STL models was reported in the range of 0.38mm and

308 0.16 mm [21]. In this work, notwithstanding, the mean error of the STL smooth model shows

309 a tolerant at values of 0.201 mm and 0.268 mm for patient number 2 and 0.254 mm and 0.214

310 mm for patient number 1. This indicates that the application CARESystem on the design of

311 insole shoe orthotics can deliver a good custom orthotics similar to the medical models [21].

312 The similarity lies in the average value of the dimensional error generated in. STL smooth

313 image model of the RP model. Accordingly, the results of this study can be applied in shoes

314 and sandal industries for producing a low cost of custom insole and outsole orthotics for

315 diabetic patients. The CARESystem strategy is very promising for design application of

316 all patients with foot deformities such as flat foot, high heel, syndrome Morton's

317 neuroma, and Metatarsalgia.

318

319 5. Conclusions

320 CARESystem was applied in this study providing the customized special orthotic

321 products that are specifically designed and manufactured for diabetics mellitus patients with

322 high risk. OIS product results have precision surface contours and dimensional fit of the foot

323 of patients, which can provide a good level of comfort and reduce pain in foot patients. Also

324 CARESystem in the future can be made for designing custom shoe orthotics for some

325 disabled patients foot (flat feet, high heel, Morton's neuroma syndrome, Metatarsalgia and

13
326 other foot deformities). This system can be expected to resolve the design and manufacturing

327 problems faced by sandal and shoe industries in an effort of development of customized

328 special OIS products.

329

330 Acknowledgements

331 The authors would like to gratefully thank to: Mr. Rieky L Liberman and Mrs. Stefani as

332 Managing Director and Head of Engineer PT. Trimitra Wisesa Abadi, Mr. Y Bambang

333 Nugroho as the managing director of PT. Hankook Delcam Indonesia, Mr. V. Ariyono, as

334 Chairman of the Department of Industrial Engineering and Mr. A.Tonny Yuniarto, as Head

335 Laboratory Production Process, Industrial Engineering Program, Faculty of Industrial

336 Technology, University of Atma Jaya Yogyakarta, and Tribology Laboratory of the

337 Department of Mechanical Engineering, University of Diponegoro in Semarang, who already

338 provided full support in the form of infrastructure CAD, CAM, CAE 3D HandySCAN and

339 3D Printer during the design and manufacturing process.

340
341
342 5. References

343 [1] J.A. Bernabu, M. Germani, M. Mandolini, M. Mengonia, C. Nester, S. Preece, R.

344 Raffaeli, CAD tools for designing shoe lasts for people with diabetes, Computer

345 Aided Design 45 (2013) 977-990.

346 [2] A.S. Salles, D.E. Gyi, The specification of personalized insoles using additive

347 manufacturing, Work 41(2012)1771-1774.

348 [3] X. Ye, H. Liu, C. Lei, C. Zhiyang, X. Pan, S. Zhang, Reverse innovative design - an

349 integrated product design methodology. Computer Aided Design 40 (2008) 812-820.

350 [4] T.X. Qiu, E.C. Teo, Y.B. Yan, W.Lei, Finite Element Modeling of a 3D Coupled

351 Foot-Boot Model. Journal of Medical Engineering and Physics 33 (2011)1228-1237.

14
352 [5] A.W.L. Yao. Application of 3D Scanning and Reverse Engineering Technique for

353 Quality Control for Quick Response Product. Int. J. Adv. Manuf. Technol. 26 (2005)

354 1284-1288.

355 [6] P.W. Anggoro, B. Bawono, P. Policharpus, J. Jamari, A.P. Bayuseno. Finite element

356 modeling of the insole shoe orthotic for foot deformities, International Conference on

357 Engineering Tribology and Applied Technology (ICETAT 2016), (2016), November

358 4-6.

359 [7] P.W. Anggoro, B. Bawono, A. Wicaksono, J. Jamari, A.P. Bayuseno, Reverse

360 innovative design from 3D mesh to 3D model insole shoe orthotic, Procedings of

361 International Conference on Engineering Tribology and Applied Technology

362 (ICETAT 2016), (2016), November 4-6

363 [8] https://www.mcad.com/3d-printing/3d-printing

364 materials/polyjetmaterials/PolyJet_Systems_and_Materials_Overview_MCAD ; 2017

365 [accessed 8.01.17].

366 [9] The American Orthopaedic Food& Ankle Society, Volume 30, Issue 1, Winter 2016.

367 [10] B. Vicenzino, Foot orthotics in the treatment of lower limb conditions: a

368 musculoskeletal physiotherapy perspective. Manual Therapy 9 (2004) 185-192.

369 [11] P.O. Mogeni, A/P. Vickeswari. Duraijajah, S. Gobee, Design and development of a

370 CAD/ CAM System for foot orthoses. ISBN: 978-0-9891305-4-7 2014 SDIWC:

371 91-99.

372 [12] Oancea, Ivan N.V, Pescaru R. Computer aided reverse engineering system for

373 costumized product. Annals of MTeM for 2013 & Proceedings of the 11th

374 International MTeM Conference 2013: 181-190

15
375 [13] R. Pescaru, G. Oancea, I. Badan. Parametric modelling of shoe lasts using reverse

376 engineering technique. Academic Journal of Manufacturing Engineering 10 (2012) 6-

377 11.

378 [14] J.D. Chapman, S. Preece, B. Braunstein, A. Hhne, C.J. Nester, P. Brueggemann, S.

379 Hutchins, Effect rocker shoe design features on forefoot plantar pressure in people

380 with and without diabetes. Journal of Critical Biomechanics 28 (2013) 679-688.

381 [15] U. Luigi, G. Claudia, The role of footwear in the prevention of diabetic foot problem,

382 J. The Diabetic Foot. Medical and Surgical Management, Contemporary Diabetes,

383 DOI 10.1007/978-1-61779-0_26, Springer Science+Business Media.2012; 519-528.

384 [16] Z. Xia, Application of reverse engineering based on computer in product design.

385 International Journal of Multimedia and Ubiquitous Engineering 2014; Vol.9, No.5 :

386 343-354

387 [17] P.W. Anggoro, B. Bawono, A. Wijayanto, J. Jamari, A.P. Bayuseno. Parameter

388 optimatizion of strategies at CNC Milling Machines Rolland Modela MDX 40R CAM

389 against surface roughness made insole shoe orthotic Eva Rubber Foam. International

390 Journal of Mechatronic & Mechanical Engineering 06 (2016) 96-106.

391 [18] P.W. Anggoro, B. Bawono, A.A. Antony, J. Jamari, A.P. Bayuseno, Determining

392 optimal toolpath strategy in the manufacture of ortothic insole shoe made from eva

393 rubber for diabetes patients. Proceding of 1st International Joint Conference on

394 Science and Technologi, (IJST 2016), Oktober 12-13.

395 [19] E. Piperi, L. M. Galantucci, J. Kaani, E. Shehi and T. Spahiu, From 3D foot scans to

396 footwear designing and production, 6th International Conference of Texile 20

397 November 2014, Triana, Albania.

398 [20] Delcam Healthcare Solution, (2015)

16
399 [21] A. Manmadhachary, Y. Ravi Kumar, L. Krishnanand, Improve the accuracy, surface

400 smoothing and material adaption in STL file for RP medical models. Journal of

401 Manufacturing Processes 21 (2016) 4655.

402 [22] H.I.Othman, Role of computer aided design and computer aided manufacturing

403 technology in prosthetic implant restorations. Int. J. Dental Clin. 4 (2012) 22-34.

404 [23] O.H. Karatas, T. Ebubekir, Three-dimensional imaging techniques: a literature

405 review. Eur. J. Dent. 8 (2014) 132-40.

406 [24] Ay. Mustafa, T. Kubat, C. Delilbasi, B. Ekici, H.E. Yuzbasioglu, S. Hartomacioglu,

407 3DBio-CAD modeling of human mandible and fabrication by Rapid-prototyping

408 technology, Usak University Journal of Material Sciences 2 (2013)135-45.

409 [25] C.Y. Huang, L.J. Luo, P.Y. Lee, J.Y. Lai1, W.T. Wang, S.C. Lin, Efficient

410 segmentation algorithm for 3D bone models construction on medical images. J. Med.

411 Biol. Eng. 31 (2010) 375-86.

412 [26] J. Santolaria, R. Jimenez, M. Rada, F. Loscos, Error compensation method for

413 improving the accuracy of biomodels obtained from CBCT data. Med. Eng. Phys. 36

414 (2014) 397-404.

415 [27] D.V. Nguyen, N.V. Quang, L.H. Le, E.H.M. Lou, Validation of 3D surface

416 reconstructionof vertebrae and spinal columnusing 3D ultrasound data-a pilot study.

417 Med. Eng. Phys. 37 (2015) 239-44.

418 [28] H. Ameddah, M. Assas. Bio-CAD reverse engineering of free-form surfaces by planar

419 contours. Computer Aided Design Appl. 8 (2011) 37-42.

420 [29] W.E. Lorensen, H.E. Cline, Marching cubes: a high resolution 3D surface

421 construction algorithm. Comput. Graph 21 (1987)163-169.

17
422 [30] D. Wu, H. Tian, G. Hao, Z. Du, L. Sun, Design and realization of an interactive

423 medical images three dimension visualization system. 3rd International Conference on

424 Biomedical Engineering and Informatics 2010; 189193.

425 [31] D.J. Yoo. Three-dimensional human body model reconstruction and manufacturing

426 from CT medical Image Data using a heterogeneous implicit solid based approach.

427 Int. J. Precis. Eng. Manuf. 12 (2011) 293-301.

428 [32] C.S. Wang, W.H.A. Wang, M.C. Lin, STL rapid prototyping bio-CAD model for CT

429 medical image segmentation, Comput. Ind. 61(2010)187-97.

430 [33] C. Denga, H. Lin, Progressive and iterative approximation for least squares B-spline

431 curve and surface fitting. Comput. Aided Design 47 (2014) 32-44.

432 [34] J.D. Janisse, W. Coleman. Pedothric care of the diabetic risk category. In: The

433 diabetic foot. (2008) 529-546.

434 [35] Clinical Best Practice Guidelines. Assessment and Management of Foot Ulcers for

435 People with Diabetes, Second Edition. Registered Nurses Association of Ontario;

436 March 2013

437 [36] L.A. Lavery, D.G. Armstrong, R.P. Wunderlich, J. Tredwell, A.J. Boulton,

438 Predictive value of foot pressure assessment as part of a population-based diabetes

439 disease management program. Diabetes Care 26 (2003)1069-1073.

440 [37] D.G. Armstrong, E.J. Peters, K.A. Athanasiou, L.A. Lavery, Is there a critical level of

441 plantar foot pressure to identify patients at risk for neuropathic foot ulceration? The

442 Journal of Foot and Ankle Surgery 37(1998) 303-307.

443 [38] S. Bus, J.S. Ulbrecht, P.R. Cavanagh. Pressure relief and load redistribution by

444 custom-made insoles in diabetic patients with neuropathy and foot deformity. Clinical

445 Biomechanics 19 (2004) 629638.

18
446 [39] D. Brown, J.J. Wertsch, G.F. Harris, J. Klein, D. Janisse. Effect of rocker soles on

447 plantar pressures. Archives of Physical Medicine and Rehabilitation 85 (2004) 81-86.

448 [40] C. van Schie, J.S. Ulbrecht, M.B. Becker, P.R. Cavanagh, Design criteria for rigid

449 rocker shoes. Foot & Ankle International 21(2000) 833-844.

450 [41] P.R. Cavanagh, A.J.M. Boulton, P. Sheehan, J.S. Ulbrecht, G.M. Caputo, D.G.

451 Armstrong. Therapeutic footwear for people with diabetes. The Journal of the

452 American Medical Association (JAMA) 288 (2002)1231-1233.

453 [42] M. Germania, M. Mandolinia, M. Mengonia, C. Nesterb, R. Raffaeli. Tools for design

454 and validation of shoe lasts for diabetic patients. Footwear Science 4 (2012) 221-241.

455

19
FIGURES CAPTION

Figure 1 Ball of foot indicated by the blue line of the 3D mesh foot area

Figure 2 The 3D Mesh of STL file for foot: (a) patient no. 1; (b) patient no. 2

Figure 3 Output file of RE foot patient in STL file format

Figure 4 Top view and isometric 3D new design of asembly foot & insole of diabetic patient

Figure 5 Top views of prototype assembly RP model for diabetic patient

Figure 6 Prototype of RP model for insole shoe orthotics

Figure 7 Area of measurement point of foot, insole and outsole orthotics


Figure 1

Figure 2

Figure 3
Figure 4

Figure 5
Figure 6

Figure 7
Tables
Table 1 Color mapping analysis and verification of scanning from 3D mesh to 3D CAD
model of insole and outsole shoe orthotics for (a) patient 1; (b) patient 2
Table 2 Measured dimensions of CAD models and prototype models of OIS for patient 2
Table 3 Measured dimensions of CAD models and prototype models of OIS for patient 1

Table 1

Mean Standard
Sample Point Point Error range
Patient Foot deviation deviation
point calculates interpolated
(mm) (mm) (mm)
Left 263.005 37.857 225.148 0.1395833 -0.3756 to 0.6413 0.24792
1
Right 236.766 37.56 199.26 0.0937500 -0.3707 to 0.5926 0.18333
Left 373.552 38.773 334.779 0.1326389 -1.6146 to 0.6066 0.35417
2
Right 303.396 36.835 266.561 0.1451389 -0.7644 to 0.5640 0.28611

Table 2

Right Foot Left Foot

Point Prototype Prototype


CAD Differences in CAD Differences
measurement measurement
Measurement size (mm) measurement in size (mm)
(mm) (mm)
A 94.62 94.55 0.07 94.53 94.26 0.27
B 72.90 73.03 0.13 76.19 76.26 0.07
C 57.93 58.56 0.63 66.49 66.23 0.26
D 221.33 221.58 0.25 215.67 216.54 0.87
E 102.10 102.09 0.01 102.29 102.62 0.33
F 97.35 97.69 0.34 96.31 96.45 0.14
G 76.19 76.37 0.18 72.48 72.32 0.16
H 251.67 251.64 0.03 251.33 251.59 0.26
I 90.65 90.56 0.09 86.99 87.27 0.28
J 97.32 97.10 0.22 98.13 97.51 0.62
K 82.72 82.96 0.24 76.61 76.76 0.15
L 35.59 35.71 0.12 39.83 39.84 0.01
M 243.33 243.64 0.31 241.67 242.00 0.67
Average Error in .STL smooth model 0.201 mm (right foot) and 0.268 mm (left foot)
Table 3

Right Foot Left foot

Point Prototype Prototype


CAD Differences in CAD Differences in
measurement measurement
measurement size (mm) measurement size (mm)
(mm) (mm)

A 88.92 88.69 0.23 86.34 86.14 0.2


B 63.75 63.68 0.07 58.87 58.65 0.22
C 51.08 50.92 0.16 49.84 49.30 0.54
D 217.00 213.98 3.02 215.50 216.43 0.93
E 90.20 90.17 0.03 94.69 94.70 0.01
F 60.05 59.94 0.11 66.95 67.00 0.15
G 69.8 69.40 0.40 72.80 72.77 0.03
H 235.05 234.47 0.58 233.20 233.10 0.1
I 85.98 85.72 0.26 83.57 83.35 0.22
J 82.61 82.79 0.18 94.69 94.54 0.15
K 59.68 59.1 0.58 62.89 62.92 0.03
L 52.62 52.11 0.51 65.67 65.52 0.15
M 226.20 226.47 0.27 224.50 225.10 0.6
Average Error in .STL smooth model 0.254 mm (right foot) and 0.214 mm (left foot)
HIGLIGHTS

CERESystem provide a standard method for designing the insole shoe orthotics
Reverse innovative design (RID) yields the 3D CAD of orthotics
The ease of manufacturing for custom insole shoe orthotics with lower production cost
The systematic of CERESystem leads accuracy in dimension of products