CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Procedía Engineering 181 (2017) 36 - 43

Procedía Engineering

www.elsevier.com/locate/procedia

10th International Conference Interdisciplinarity in Engineering, INTER-ENG 2016

Classification of the Elderly Foot Types Based on Plantar Footprints

Mariana Costeaa, Bogdan Sarghiea, Aura Mihaia'*, Elena Rezusb

aGheorghe Asachi Technical University of Iasi, 67 Dimitrie Mangeron Blvd., Iasi 70050, Romania bGrigore T. Popa University of Medicine and Pharmacy,16 Universitatii, Iasi 700115, Romania

Abstract

The article presents the methodology and the results of the studies conducted in order to develop a rational classification technique of the elderly foot typologies, based on parameters derived from the plantar footprint. The plantar footprints were taken from 67 women, aged between 52-84 years old with RSscan pressure plate and the associated system, Footscan 7 Gait, 2nd Generation, 0,5 Gait Scientific System. To classify the foot typology, Chippaux-Simark Index and Hallux-Valgus Angle have been used. According to the obtained results, the subjects were divided into five categories, subjects with Normal Foot, High Arched Foot, Flat Foot, Hallux-Valgus Foot and Hallux Varus Foot. The authors have developed a method of determining the middle area of a plantar footprint for a more accurate measurement, being able to register it in the same way, for all subjects. Using statistical analysis, differences between age groups have been found, demonstrating the necessity of modelling and designing customized shoe lasts, footwear and prophylactic components according to the age of the subject. A good fit of the shoe on the elderly foot involves modifying the shoe last, but this can only be achieved under certain standards. Having biomechanical data of the foot and a clear classification of it, the customization of the product is simplified, reducing the number of tests and increasing the footwear technological process efficiency. © 2017PublishedbyElsevierLtd. Thisisanopenaccess article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/4.0/).

Peer-review under responsibility of the organizing committee of INTER-ENG 2016

Keywords:elderly foot typology; plantar footprint; foot parameters; Chippaux-Simarkindex; Hallux-Valgus angle.

1. Introduction

A good fit of the shoe on the foot involves customizing the last, but this can only be achieved under certain standards. Having biomechanical data of the foot simplifies the customization of the product, reduces the number of

* Corresponding author. E-mail address: amihai@tex.tuiasi.ro

1877-7058 © 2017 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.org/licenses/by-nc-nd/4.0/).

Peer-review under responsibility of the organizing committee of INTER-ENG 2016

doi:10.1016/j.proeng.2017.02.360

test and increases the footwear technological process efficiency. Producing customized footwear requires knowledge of both foot dimensions and characteristics [1^6]. The identification of similar characteristics and their classification by groups of subjects has a high impact on shoe last's design [7^10], soles and insoles design as well as in the treatment and prevention of foot disorders [11^18]. Foot type it is a general term widely used, to describe a number of architectural features of the foot in order to obtain clues of its dynamic functioning [19]. Foot dynamics is perceived as being linked to a variety of musculoskeletal symptoms, including personal injury from running and plantar pain, especially on the heel [20]. The biomechanical studies have progressed by comparing the presence, absence or size of the foot typological parameters between groups with or without pathologies under investigation [21]. Despite the widespread use of these parameters, it was recognized that the objective and quantitative analysis of foot typologies remains elusive [22, 23]. The absence of an absolute dimension of the foot type has resulted in considerable variation in choosing the measurements to determine the foot typology [24, 25]. This is in opposition to the suggestion that in order to identify the relations between foot typology and pathology it is necessary to use a unique classification system to allow accurate recognition in each situation [26].

2. Method

Foot pressures were taken on barefoot. For statistical measurements, subjects were asked to stand with both feet on the pressure plate, in an equilibrium position. For dynamic measurements, subjects were asked to walk normally on the plate, marking a measurement for the right foot and one for the left foot. The procedure was repeated three times for each subject analysed. Unlike traditional methods of taking plantar footprints, a dynamic footprint is obtained, using this system, the software automatically links images in all phases of walking. Images were exported to scale 1:1 and 5 dimensions were measured [27, 28] to determine the main parameters of the footprint. This dynamic plantar footprint examination is a rapid, non-invasive, simple method that can quantify the foot configuration.

2.1. Equipment

In order to obtain the plantar footprints, RSscan pressure plate [29], figure 1 and the associated system, Footscan 7 Gait, 2nd Generation, 0,5 Gait Scientific System, produced by RSscan International were used.

2.2. Subjects

To determine the foot types, the plantar footprints were taken from 67 women, aged between 52-84 years old and weighing 45-70 kg. The subjects were divided in three categories, group 1: 52-59 years old, group 2: 60-64 years old and group 3: 65-84 years old. All subjects read and signed an informed consent before testing.

Fig. 1. Taking plantar footprints with RSscan System

3. Results

Each plantar footprint has been manually processed by constructing the reference lines [30] (figure 2): Longitudinal axis (UV) that passes through the heel centre, C point and the middle of second toe; Heel line, AB, perpendicular on longitudinal axis, UV; Toe line, FG, which connects the metatarsal phalangeal joints, I and V; The line of minimum width of plantar footprint, ED; The line from first metatarsal joint, tangent to first toe.

In most of the published studies concerning foot typology, the segment ED is measured differently from person to person, in the middle of plantar footprint, in the narrowest area. To have control of this measurement and to be able to register it the same for all subjects, the authors have establish the followings:

Point A (extreme interior point of the heel) will join point H (the intersection of toe line with longitudinal axis) to obtain the AH segment;

Point F (extreme point of first metatarsal phalangeal joint) will join point B

(extreme exterior point of the heel) to obtain the FB segment;

The segment ED is drawn parallel to FG, where AH and FB intersect.

Legend:

UV - foot length AB - heel width

DE - distance between the intersection ofBF with AH FG - toe width

P - angle of I toe deviation, Hallux-Valgus Angle (HVA)

Fig. 2. Plantar footprint - reference lines

Classification of different types of foot using the Footprint Angles and Chippaux-Smirak Index varies by the parameters analyzed, indicating the dependence to foot plantar print and the wide variety of foot morphology. It has been proposed a basic classification technique which allows for rational classification of foot types [31]. Comparisons between initial classification and classification using individual parameters showed significant discrepancies between the parameters studied [32]. Clinical measurements of foot structure have been developed to quantify certain aspects of foot geometry for diagnosing various diseases. Such type of measurement is the width of median-longitudinal arch. This can be achieved using the following parameter: Chippaux-Smirak Index (CSI) - the ratio between minimum width of midpoint and the widest width of toes [33]. Based on the measured parameters for each subject separately and applying the above calculus relationships the subjects were classified, considering the following criteria: based on Chippaux-Simark (CSI) Index; based on angle of I toe deviation, Hallux-Valgus Angle (HVA)

3.1. Foot types based on Chippaux-Simark Index (CSI)

Following the analysis of Chippaux-Simark Index, the results show that 22% of the subjects have high arched foot, 11% have a tendency of flattening, 11% have flat foot and the rest of 56% have normal and intermediary foot, table 1.

Table 1. Chippaux -Simark index

Foot type Index values The entire group Group 1, 52-59 years old Group 2, 60-64 years old Group 3, 65-84 years old

Left Right Left Right Left Right Left Right

High arched [0,20) 15 19 10 11 3 5 2 3

Normal [20,30) 13 15 5 8 6 4 2 3

Intermediary [30,40) 25 17 9 6 8 7 8 4

Low arch [40,45) 7 11 3 3 0 2 4 6

Flat >45 7 5 2 1 2 1 3 3

High arched Normal Intermediary Low arch Flat

Fig. 3. Foot type based on CSI index for the entire group, in %

The Chippaux-Simarklndex analysis highlights that there is a higher tendency of high arch foot in case of group 1, 52-59 years old, a flattened foot in case of group 3, 65-84 years old, figure 3.

3.2. Foot types based on Hallux-Valgus Angle (HVA)

The hallux-valgus angle, HVA, is used to detect the presence on the foot of Hallux-Valgus (interior deviation of first toe) and Hallux-Varus (exterior deviation of first toe). Following the results of P angle, the results show that 28% of women have Hallux-Valgus condition, 17% have the imminent risk to develop it, 30% have normal foot and 25% have Hallux-Varus condition, an inward deviation of the first toe, table 2.

Table 2. HVA- Hallux-Valgus Angle

Foot type

Group 1, 52-59 years Group 2, Group 3,

The entire group old 60-64 years old 65-84 years old

Left Right Left Right Left Right Left Right

Index values

Hallux-Valgus Outward movement Normal Hallux-Varus

[10, 15) [0, 10) <0

5 7 11

Hallux-Valgus

Outward movement

Normal

Hallux-Varus

Fig. 4. Foot type based on HVA index for the entire group, in %

As the results show, the prevalence of Hallux-Valgus is higher for the third group, in case of women of 65 to 84 years old, figure 4. By joining the previous results, each group of subjects will be divided in 5 categories, respectively: Flat Foot, High Arched Foot, Hallux-Valgus Foot, Hallux-Varus Foot, Normal Foot.

3.3. Statistical analysis of foot indexes

A statistical analysis has been performed, using the values obtained by measuring plantar footprints. In table below, the results of a descriptive statistics with average, standard deviation, minimum, maximum and median indexes can be observed.

Table 3. Statistic indexes for plantar footprint parameters

Parameters Foot Average Standard deviation Minimum Maximum Medial

Heel width left 53.87 3.61 46.00 64.00 54.00

right 53.90 4.34 43.00 63.00 54.00

Minimum width of the foot left 25.90 13.03 0.00 45.00 28.00

right 24.36 13.42 0.00 47.00 25.00

Toe width left 89.13 6.56 70.00 102.00 89.00

right 88.67 5.96 75.00 98.00 88.00

HVA index left 6.94 12.32 -20.00 35.00 7.00

right 5.84 11.58 -18.00 39.00 5.00

CSI index left 28.71 14.37 0.00 49.43 31.58

right 27.24 14.71 0.00 49.47 29.87

By studying the values of standard deviation, higher values are observed in case of minimum width of footprint, HVA and CSI.

A Student T-test was conducted in order to find the statistical significance of foot indexes for left and right foot. The initial null hypothesis was: there are no difference between left and right foot and the mean values [34]. This test can be used to determine if two sets of data are significantly different [34], in case of an number of women subjects, with df=2x(n-l) degree of freedom, an accepted probability of P=65% and two-tailed test p-value = 0.05. In table 4 are presented Levene's and T tests for left and right foot CSI, in case of group 1 compared to group 2, group 1 compared to group 3 and group 2 compared to group 3. The two-tailed testp-valuein case of both foot, when comparing group 1 to group 3 is lower than 0.05, showing that the initial hypothesis is no longer valid.

Table 4. Independent Samples Test of Chippaux-Simark, comparing age groups

Levene's Test for Equality of Variances

t-test for Equality ofMeans

PC-tailed)

Mean Difference

Std. Error Difference

95% Confidence Interval of the Difference

CSI-left-comparing

Equal variances assumed

0.110 -0.959 46

0.343 -4.16171

4.34162

-12.90093

4.57752

Levene's Test for Equality of Variances

t-test for Equality ofMeans

p (2- Mean

Std. Error

95% Confidence Interval of the Difference

tailed) Difference Difference Lower Upper

group lto group Equal variances 2 not assumed

-0.999 43.523 0.323 -4.16171 4.16649 -12.56133 4.23791

CSI-right- Equal variances

comparing assumed

group lto group„ , 2 Equal variances

not assumed

0.124 0.727 -1.059 46

0.295 -4.60091 4.34306 -13.34304 4.14122

-1.066 39.461 0.293 -4.60091 4.31588 -13.32735 4.12553

CSI-left-comparing group 1 to group 3

Equal variances assumed

Equal variances not assumed

3.406 0.071 -2.340 46

0.024 -9.91434 4.23684 -18.44266 -1.38602

-2.481 44.975 0.017 -9.91434 3.99649 -17.96380 -1.86487

CSI-right-comparing group 1 to group 3

Equal variances ^ ^ assumed '

Equal variances not assumed

0.320 -2.812 46

0.007 -11.59038 4.12161 -19.88674 -3.29402

-2.922 43.243 0.006 -11.59038 3.96643 -19.58815 -3.59261

CSI-left-comparing group 2 to group 3

Equal variances assumed

Equal variances not assumed

0.008 0.928 -1.427 36

0.162 -5.75263 4.02997 -13.92578 2.42052

-1.427 35.739 0.162 -5.75263 4.02997 -13.92785 2.42259

CSI-right-comparing group 2 to group 3

Equal variances assumed

Equal variances not assumed

0.322 0.574 -1.600 36

0.118 -6.98947 4.36857 -15.84934 1.87039

-1.600 35.191 0.119 -6.98947 4.36857 -15.85642 1.87747

In Table 5 are presented Levene's and T tests for HVI left and right foot, in case of group 1 compared to group 2, group 1 compared to group 3 and group 2 compared to group 3. The Levene's test p-value in case of left foot, when comparing group 1 to group 3 and group 2 to group 3 is lower than 0.05, accepting that equal variances are not assumed, there are difference between those age groups.

Table 5. Independent Samples Test of Halux Valgus Angle, comparing age groups

Levene's Test for Equality of Variances

t-test for Equality ofMeans

p (2- Mean tailed) Difference

Std. Error Difference

95% Confidence Interval of the Difference

Equal variances

assumed °-104 °-748 °-865 46

HVA-left-comparing group lto group 2 Equal varianCes not assumed

HVA -riaht- Equal variances

nv^. iigm . 0.034 0.855 1.260 46

comparing group assumeu

lto group 2 Equal variances not assumed

0.391 2.79492 3.23088 -3.70850 9.29834

0.872 39.736 0.388 2.79492 3.20376 -3.68147 9.27131

0.214 3.80399 3.02019 -2.27534 9.88332

1.312 43.486 0.196 3.80399 2.89945 -2.04141 9.64940

HVA -left- Equal variances comparing assumed groupltogroupEqual

variances not assumed

HVA -right- Equal variances

comparing group assumed

lto group 3 . .

a r Equal variances

not assumed

7.020 0.011 -0.040 46

-0.037 29.790 0.971 -0.15245

Equal variances assumed

HVA -left-comparing group2togroupEqual

variances not assumed

3.061 0.087 0.246 46

0.231 30.895 5.978 0.020 -0.678 36

-0.678 31.797 0.502 -2.94737

HVA -right- Equal variances

comparing group assumed

2 to group 3 . .

b r Equal variances

not assumed

3.978 0.054 -0.733 36

-0.733 29.905

0.969 -0.15245 3.85782 -7.91784 7.61294

4.13835 -8.60658 8.30168

0.807 0.90926 3.69870 -6.53585 8.35436

0.819 0.90926 3.93009 -7.10732 8.92583

0.502 -2.94737 4.34422 -11.75786 5.86312

4.34422 -11.79848 5.90374

0.468 -2.89474 3.94854 -10.90274 5.11327

0.469 -2.89474 3.94854 -10.95980 5.17033

4. Conclusions

These studies have been conducted in order to develop a rational classification technique of foot typologies based on parameters derived from the plantar footprint. The authors have established a rule of measuring the middle area of a plantar footprint in order to have control of this measurement, and to be able to register it,in the same way for all subjects. In order to classify the foot typology, Chippaux-Simark Index and Hallux-Valgus Angle have been used. According to the obtained results, the subjects will be divided into five categories, subjects with Normal Foot, High Arched Foot, Flat Foot, Hallux-Valgus Foot and Hallux Varus Foot.

Using statistical analysis, Levene and T-test, differences between age groups have been found, demonstrating the necessity of modelling and designing customized shoe lasts, footwear and prophylactic components according to the age of the subject.

The necessity for these kind of studies is found both in the subsequent analysis of foot anthropometrical parameters, in modelling and designing prophylactic footwear, in reducing the number of test of physical prototypes, therefore increasing the footwear technological process efficiency.

Acknowledgement

This work was supported by UEFSCDI Bucharest under the Partnership Programme project MOBILITY: Preventing gait deficiencies and improving biomechanical parameters for the elderly population by designing and developing customized footwear - code PN-II-II-PT-PCCA 2013-4, contract 122/2014.

References

[1] S. R. Asanka, R. S. Goonetilleke, C. P. Witana, Model based foot shape classification using 2D foot outlines, Computer-Aided Design 44 (2012) pp. 48-55, available at www.elsevier.com/locate/cad.

[2] R.E. Wunderlich, P.R. Cavanagh, External foot shape differences between males and females and among races. Abstract book, 23rd Ann. Mtg ofthe American Society of Biomechanics, ASB; 1999. pp. 68-9.

[3] I.V. Herghiligiu, A. Mihai, B. Sarghie, R. Souto Bizarro, C. Arias, Framework ofthe e-learning training program on corporate social responsibility, The 12thInternational Scientific Conference eLearning and Software for Education, Bucharest, 2016, 10.12753/2066-026X-16-255, pp. 526.

[4] B. Sarghie, A. Mihai, I.V. Herghiligiu, E-learning application for 3D modelling ofcustom shoe lasts using templates, The 12th International Scientific Conference eLearning and Software for Education, Bucharest, 2016, 10.12753/2066-026X-16-260, pp.553.

[5] D. Ionesi, L. Ciobanu, B. Sarghie, E-Learning Application for a better understanding ofshoes 3D modeling, 10th International Scientific Conference ELSE, ISSN 2360-2198, 2014, 10.12753/2066 026X 14 285, pp.196.

[6] D.C. Deselnicu, A.M. Vasilescu, A. Mihai, A.A. Purcarea, G. Militaru, New products development through customized design based on customers' needs. Part 2: Foot Pathology Manufacturing Parameters, The 9th International Conference Interdisciplinarity in Engineering, INTER-ENG 2015, Procedia Technology 22 (2016) 1059 - 1065, available at www.sciencedirect.com.

[7] M. Kouchi, An analysis of foot shape variation based on the medial axis offoot outline. Ergonomics 1995;38:1911-20.

[8] M. Mochimaru, M. Kouchi, M. Dohi, Analysis of 3D human foot forms using the Free Form Deformation method and its application in grading shoe lasts. Ergonomics 2000;43(9):1301-13.

[9] R.E. Wunderlich, P.R.Cavanagh, Gender differences in adult foot shape: implications for shoe design. Medicine & Science in Sports & Exercise 2001; 33(4):605-ll.

[10] M. Pajtina, A. Mihai, N. Bilalis, Finite element analysis for insole-sole prototypes, Proceedings of'The 4th International Conference on Advanced Materials and Systems", ICAMS 2012, Bucuresti, 2012, ISSN 2068-0783, pp. 359-364.

[11] T.J. Hwang, K. Lee, H.Y. Oh, J.H. Jeong, Derivation of template shoe-lasts for efficient fabrication ofcustom-ordered shoe-lasts. Computer-Aided Design 2005;37(12):1241-50.

[12] M.R. Hawes, D. Sovak, M. Miyashita, S.J. Kang, Y. Yoshihuku, S. Tanaka, Ethnic differences in forefoot shape and the determination of shoe comfort. Ergonomics 1994;37(l):187-96.

[13] B.M. Nigg, M.A. Nurse, D.J. Stefanyshyn, Shoe inserts and orthotics for sport and physical activities. Medicine & Science in Sports & Exercise 1999;31(7): S421-8.

[14] R.S. Goonetilleke, A. Luximon, K.L.Tsui, The quality of footwear fit: whatwe know, don't know and should know. In: Proceedings ofthe human factors and ergonomics society conference. San Diego (CA); 2000. pp. 515-8.

[15] M. Dohi, M. Mochimaru, M. Kouchi, Foot shape and shoe fitting comfort for elderly Japanese women. Japanese Journal ofErgonomics 2001;37(5):228-37.

[16] R.S. Goonetilleke, A. Luximon, Designing for comfort: a footwear application. In: Proceedings ofcomputer-aided ergonomics & safety conference. Maui (Hawaii, Plenary Session); 2001.

[17] C.P. Witana, R.S. Goonetilleke, J. Feng, Dimensional differences for evaluating the quality offootwear fit. Ergonomics 2004;47(12):1301-17.

[18] M.S. Rathleff, R.G. Nielsen, O. Simonsen, C.G. Olesen, U.G. Kersting, Perspectives for clinical measures ofdynamic foot function— reference data and methodological considerations. Gait & Posture 2010;31(2):191-6.

[19] H. Lee, R. Grosse, R. Ranganath, A.Y. Ng, Convolutional deep beliefnetworks for scalable unsupervised learning ofhierarchical representation, International Conference on Machine Learning, 2009, pp. 77-85.

[20] S. Sivapalan, D. Chen, S. Denman, S. Sridharan, C. Fookes, Gait energy volumes and frontal gait recognition using depth images, IEEE International Joint Conference on Biometrics, 2011, pp. 501-506.

[21] S. Zheng, K. Huang, T. Tan, Evaluation framework on translation-invariant representation for cumulative foot pressure image, The 18th IEEE International Conference on Image Processing (ICIP), 2011, pp. 201-204.

[22] S. Zheng, K. Huang, T. Tan, D. Tao, A cascade fusion scheme for gait and cumulative foot pressure image recognition, Pattern Recognition, no. 45, 2012, pp. 3603-3610, available atwww.elsevier.com/locate/pr.

[23] P.F. Felzenszwalb, R.B. Girshick, D. McAllester, D. Ramanan, Object detection with discriminatively trained part-based models, IEEE Transactions on Pattern Analysis and Machine Intelligence, no. 32, 2010, pp. 1627-1645.

[24] I. Bouchrika, M. Nixon, Model-based feature extraction for gait analysis and recognition, International Conference on Computer Vision/Computer Graphics Collaboration Techniques, 2007, pp. 150-160.

[25] T. Zhang, X. Li, D. Tao, J. Yang, Multimodal biometrics using geometry preserving projections, Pattern Recognition, no. 41, 2008, pp. SOS-SIS.

[26] I. Mathieson, D. Upton, A. Birchenough, Comparation offootprint parameters calculated from static and dynamic footprints, The foot, no 9,1999, pp. 145-149.

[27] C. Shan, S. Gong, P.W. McOwan, Fusing gait and face cues for human genderrecognition, Neurocomputing, no. 71, 2008, pp. 1931-1938.

[28] M. Costea, A. M. Vasilescu, G. Hortal, A. Mihai, Plantar footprints analysis - case study (part 2), Leather and Footwear Journal, vol. 14, no.4, Certex Publishing House, 2014, ISSN 15834433, pp.243-250.

[29] http://www.rsscan.co.uk/, accessed at 03/10/2016.

[30] A. Mihai, M. Pastina, Classification offoot types, based on plantar footprint, Proceedings of'The 4th International Conference on Advanced Materials and Systems", ICAMS 2012, Bucuresti, 2012, ISSN 2068-0783, pp. 347-352.

[31] E. Billis, E. Katsakiori, C. Kapodistrias, E. Kapreli, Assessment offoot posture: Correlation between different clinical techniques, The Foot, no. 17, 2007, pp. 65-72, available atwww.sciencedirect.com.

[32] M.E. Nikolaidou, K.D. Boudolos, A footprint-based approach for the rational classification offoot types in young schoolchildren, The Foot, no. 16, 2006, pp. 82-90, available atwww.sciencedirect.com.

[33] M. O. Papuga, J. R Burke., The reliability ofthe associate platinum digital foot scanner in measuring previously developed footprint characteristics: a technical note, Journal ofManipulative and Physiological Therapeutics, vol. 34, no. 2, 2010, pp. 114-118, available at www.sciencedirect.com.

[34] https://en.wikipedia.org, accessed at 03/10/2016.