research article development of a robotic assembly for...

Research ArticleDevelopment of a Robotic Assembly for Analyzingthe Instantaneous Axis of Rotation of the Foot Ankle Complex

Kelly N Salb Daniel M Wido Thomas E Stewart and Denis J DiAngelo

BioRobotics Laboratory Department of Orthopaedic Surgery and Biomedical EngineeringThe University of Tennessee Health Science Center 956 Court Avenue Suite E226 Memphis TN 38163 USA

Correspondence should be addressed to Denis J DiAngelo ddiangelouthscedu

Received 14 September 2015 Accepted 16 February 2016

Academic Editor Kiros Karamanidis

Copyright copy 2016 Kelly N Salb et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Ankle instantaneous axis of rotation (IAR) measurements represent a more complete parameter for characterizing joint motionHowever few studies have implemented this measurement to study normal injured or pathological foot ankle biomechanicsA novel testing protocol was developed to simulate aspects of in vivo foot ankle mechanics during mid-stance gait in a humancadaveric specimen A lower leg was mounted in a robotic testing platform with the tibia upright and foot flat on the baseplateAxial tibia loads (ATLs) were controlled as a function of a vertical ground reaction force (vGRF) set at half body weight (356N)and a 50 vGRF (178N) Achilles tendon load Two specimens were repetitively loaded over 10 degrees of dorsiflexion and 20degrees of plantar flexion Platform axes were controlled within 2microns and 0008 degrees resulting in ATLmeasurements withinplusmn2N of target conditions Mean ATLs and IAR values were not significantly different between cycles of motion but IAR valueswere significantly different between dorsiflexion and plantar flexion A linear regression analysis showed no significant differencesbetween slopes of plantar flexion paths The customized robotic platform and advanced testing protocol produced repeatableand accurate measurements of the IAR useful for assessing foot ankle biomechanics under different loading scenarios and footconditions

1 Introduction

The kinematic and structural properties of human joints maybe affected by diseases injuries or surgical alterations In thecase of the ankle joint any disease injury or elected surgerylike osteoarthritis or ankle arthroplasty will impact both themotion behavior of the joint (how it moves) and its structuralstiffness properties (how it rotates under amuscle load) [1ndash6]In vivo studies [5 7ndash11] and computational models on anklebiomechanics [12ndash15] typically analyze the range of motion(ROM) or joint stiffness properties Joint motion can also bedescribed by rolling and sliding of articular surfaces duringmotion representative of a moving axis of rotation withdependency on loading scenarios [12] The two-dimensionalinstantaneous axis of rotation (IAR) represents a moreadvanced parameter for characterizing joint kinematics yetit has been void frommost ankle biomechanics researchTheability to detect shifts in IARmay help in defining injury typeandor the impact of injury on foot ankle mechanics as well

as the effects of surgical procedures and implant and orthoticdesign

Some protocols assume that the ankle complex behaveslike a hinge joint with a single axis of rotation whereas otherstudies suggested that a fixed axis of rotation with articularcongruence may be an incorrect kinematic description forankle joint motion [8 12ndash17] Few biomechanical testingplatforms offer a physiologic loading environment [18ndash21]some only investigate one instance of gait [18 22] and othersapply loads or force a kinematic profile estimated by a singlespecimen [18 20 23]

To date few studies have been conducted which addressa two-dimensional instantaneous axis of rotation (IAR) orthree-dimensional instantaneous helical axis (IHA) analysisof the ankle joint [7 9 24] IAR data provide additionalparameters for characterizing tibiotalar joint motion wheremeasurements are a direct representation of the effects ofjoint articular geometry and soft tissue structures Three-dimensional gait studies have demonstrated minimal motion

Hindawi Publishing CorporationApplied Bionics and BiomechanicsVolume 2016 Article ID 5985137 9 pageshttpdxdoiorg10115520165985137

2 Applied Bionics and Biomechanics

out of the sagittal plane during stance phase gait reducingthe need for analyses of ankle jointmotion to two dimensions(ie IAR) Quantifying the IAR of the ankle joint during gaitmay have the potential to advance the understanding of thebiomechanical properties of the foot and ankle includingarch formation ankle arthroplasty design and surgical tech-nique and effects of orthotics and footwear

Current gait simulators and biomechanical testing proto-cols either are unable to analyze the IAR chose not to includethis parameter as part of the analysis [4 16 18ndash23 25ndash29] orare limited to high errors in the calculation of the IAR due tomethodology [7 24 29] There is a need for a biomechanicaltesting platform and protocol that will provide simulationof controlled tibia and Achilles tendon (AT) loads withoutconstraining the foot ankle kinematic profile

The purpose of this work was to establish a 2D testingmethod to determine the location of the IAR of the anklejoint under simulated stance phase conditions in a humancadaveric model Additional tests were done to confirmthe accuracy and repeatability of the IAR measurementOnce the rotational axis is known (describing where theankle motion occurs) a more in-depth analysis of the anklejointrsquos mechanics can be undertaken (such as convertingthe external joint moment into its internal components ofa muscle force times a moment arm) a level of in vitroanalysis currently not possible with current testing methodsFurther analysis of the path of the ankle jointrsquos rotational axisoffers new insight into the effects that disease and surgicalalterations have on the ankle kinematics and mechanics

2 Materials and Methods

21 Specimen Preparation A matched pair (male age of37 and body weight of 712N) of human below knee lowerextremity specimens was procured from Restore Life USA(Johnson City TN) Specimens were frozen at minus20∘C Beforepreparation the feet were thawed in a refrigeration systemfor two days Medial and lateral radiographs were used toverify the absence of anatomical abnormalities or surgerySoft tissue andmuscle were resected to expose approximately100mm of the proximal shafts of the tibia and fibula A singleone inch 6 wood screw was then placed across the proximalends of the tibia and fibula to stabilize the bones and addadditional fixation for the potting material After the AT wasexposed and soft tissue was excised to allow approximately150mmofAT length for clamping a cable puller was attachedand secured with a U-bolt to increase clamping powerThe tibial shafts were then mounted concentrically into acylindrical pot using an alignment frame to position the tibiain a neutral vertical orientation Low-melting-point bismuthalloy (Rotometals Inc San Leandro CA) was used to fix thetibia and fibula together and create a mounting fixture forattachment to the robotic testing platform (RTP) The finalprepared specimen is shown in Figure 1

22 Robotic Testing Platform An existing custom designedmultiaxis testing platform was utilized to simulate aspectsof stance phase gait mechanics under displacement controland force feedback (Figure 2) [30] Two linear actuators a

Mounting fixture

U-bolt

Tendon cable puller

Fixture reference

point

Figure 1 A prepared below the knee lower extremity specimenTissue was resected from the proximal shafts of the tibia fibulaand AT A cable puller was attached to the AT for tendon loadapplication Bismuth alloy material was used to create a mountingfixture for the proximal shafts of the tibia and fibula for rigid fixationto the testing platformThe fixture reference point was located at thecenter of the upper mounting fixture

Parker Hannifin Corp (Cleveland OH) 406XR series linearball screw actuator and an Exlar (Chanhassen MN) GSX-30 linear roller screw actuator were aligned in 119909-axis and 119911-axis respectively Control specifications of platform actuatorswere 2 microns in 119883 and 031 microns in 119885 The rotarymotors and drive units of the original test RTPwere upgradedwith Harmonic Drive units (Peabody MA model FHA-25C-160-US250) having improved resolutions of 0008 degreesPrevious position control pilot trials demonstrated that eachindividual axis was controllable to within tolerance of plusmn2times the resolution of each feedback sensor corresponding toplusmn00091 degrees plusmn4microns in 119911 direction and plusmn06micronsin 119909 direction for sagittal plane movement

Two six-axis load cells (JR3 Inc Woodland CA models100M40 and 45E15S) were attached to the robotic testingplatform assembly and measured the three orthogonal forcesandmoments applied to the tibia via the gimbal assembly andforces transferred through the foot to the ground at the baseof an 119883-119884 table Load cell resolutions and capacities were800 plusmn 044N in 119911-axis 400 plusmn 009N in 119909-axis and 119910-axisand 40 plusmn 002Nm about all axes respectively The load cellrated accuracies were 10 of full scale with linearity errorof 01 full scale The robot testing system actuator fixturesand load cells were significantly greater in stiffness thanthose of human tissue such that any deflection under testloading was negligible (ie less than 50 microns for 1000Nload) All reported deformation represented the compliancyof specimen tissue

Potted specimens were clamped securely in a mountingblock and rigidly connected to the testing platform A sagittalplane was established for each specimen by bisecting thesecond metatarsal and AT [11 13 17] with a vertical axisaligned with the tibia that matched the 119883-119885 plane of thetesting platform A static AT force vector (119865

119886) was applied

via a cable-pulley system designed to minimize friction to

Applied Bionics and Biomechanics 3

Rotation

X

Y

W

Z

Pulley

Achilles load

table

Pulley

Base load cell

Mounting block

Upper loadcell

X-Y

about Y

Translation in Z

Translation in X

Figure 2Diagramofmultiaxis robotic system configured for the ankle study [30]The testing platform consisted of an upper gimbal assemblywith two translational and one rotational axes along with a 6-axis load cell a mounting block to rigidly affix the specimen to the RTP a pulleysystem to apply a static load to the AT and an 119883-119884 table with a second 6-axis load cell to aid in the mounting procedure The coordinatesystem is shown by 119909-axis 119910-axis and 119911-axis Translational and rotational vectors of the controlled axes of the RTP are also shown

within four degrees of the vertical tibia axis while the heelwas raised approximately 50mm and the 119883-119884 table wasunlocked (Figure 3) This feature allowed for unconstrainedarch formation while transferring loads from the tibia tothe calcaneus and maintaining foot support throughout thegait simulation [31 32] The heel was then returned to theneutral position and an axial tibia force (119865

119905) was applied to

the specimen to meet the desired vertical ground reactionforce (vGRF) condition within a prescribed plusmn2N toleranceAfter the desired loading condition was met the 119883-119884 tablewas locked and the rotation of the specimen was started

23 Test Protocol Custom software (Adept Inc San JoseCA) was written to simulate the two-dimensional biome-chanics of the ankle joint during the early stance phase ofwalking gait A rigid body analysis was used to describe thedynamic loading characteristics of the ankle as a function ofan Achilles load and the external joint reaction forces (119865

119909 119865119911)

and bending moment (119872) The relationship between the ATforce (119865

119886) the tibia force and the vertical ground reaction

force (ie vGRF) minus the weight of the foot is shown inFigure 4 Using this relationship the AT was statically loadedas a percentage of the vGRF acting on the foot and ankle

set at half body weight (356N) and a 50 vGRF (178N)AT load was applied based upon values of physiologic peakcontractive tension of the AT [9 11] These conditions wereselected within the 800N capacity of the upper load cell toachieve reproducible subinjurious loading conditions

With the potted specimen mounted and aligned to thetesting platform (Figure 5) the fixture reference point locatedon the upper mounting fixture (Figure 5(a)) was transferredto the null tool tip (NTT) of the robot located at the centerof the gimbal assembly From there a tool tip transformationwas applied to reposition the robotrsquos rotational tool tip axis tothe initial prescribed ankle rotational axis (pARA) (located atthe center of the talus) which was measured off of a scaledmedial radiograph of the foot processed in ImageJ (NIH)

The tibia started in a neutral orientation (0 degrees)and was rotated about the initial pARA (Figure 5(a)) Aftereach rotation the position of the pARA was changed asmall amount (usually between 50 and 100 microns) alongeither direction of the rotated 119909-axis and 119911-axis so as toreduce the forces acting along those axes and the associatedincremental moment contribution (Δ119872pARA) By reducingthese off-axis forces to 2N or less the only load responsiblefor creating the joint rotation was the applied moment (ie


X

Z

Figure 3 Specimen mounted in the RTP This image showsclearance between the heel and 119883-119884 table as the AT load is appliedallowing unconstrained arch formation

Δ119872pARA sim 0) This new location was considered as the trueARA (Figure 5(b)) and represented a ldquopure ankle momentrdquoloading condition

Once the pure moment condition was established theshear and compressive forces (1198651015840

119909

and 1198651015840119911

in Figure 6) wereadded to maintain the axial tibial force and vGRFs Fur-thermore by applying these forces along the new rotationalaxes that pass through the ankle rotational axis the appliedpure ankle moment condition was not altered The processwas repeated every 05 degrees until the full 10 degrees ofdorsiflexion or 20 degrees of plantar flexion were achieved Aunique feature of the robotic testing platform was the abilityto play back the motion and loading conditions in real timeThroughout the testing sequence specimens were kept moistat regular intervals with a 09 saline mist

24 Data Analysis Force and positional data were sampledat 20Hz A modified set of equations originally derived byCrisco III et al [33] were applied to the NTT rotational andpositional data for each increment of rotation to calculate theIAR of the ankle in the sagittal plane Unlike previous studiesthat used positional data from camera targets or radiographsto calculate the IAR [7 24] this study used the positional datafrom the robot to determine the IAR values In doing so theoverall error in the IAR calculations of the entire testing sys-tem (including the testing frame actuator gimbal assemblymounting fixtures and load cell) was within plusmn0050mm forloads up to 1000NThe path of the IARwas plotted back ontothe specimen and expressed relative to an ldquoankle coordinateframerdquo thatwas defined by the bisector of the tibia andhighestpoint on the tibial mortise in the initial neutral vertical ori-entation (Figure 7) The same process used to determine thelocation of the pARA and fixture reference point (Figure 2)was used to locate the ankle coordinate frame

The test protocol cycled each specimen five times fortissue conditioning and five times in dorsiflexion and plantar

Table 1 Average axial tibia loading conditions

SpecimenAxial tibia load (N)

(Mean plusmn standard deviation)Dorsiflexion Plantar flexion

1 5165 plusmn 10 5060 plusmn 15

2 5195 plusmn 10 5135 plusmn 05

flexion to demonstrate repeatability The final three cycleswere plotted and analyzed statistically using GraphPad Prism60 software (La Jolla CA) Mean axial tibia forces and mean119883 IAR values and 119885 IAR values were compared separatelyfor dorsiflexion and plantar flexion using One-Way ANOVAwith Holm-Sıdak post hoc test for multiple comparisons(120572 = 005) to show variability between cycles The mean119883 IAR values and 119885 IAR values for all three cycles werealso compared between dorsiflexion and plantar flexionAdditionally the slopes of the plantar flexion IAR paths werecompared between cycles using a linear regression analysis inGraphPad Prism 60 software Results are presented as meansplusmn standard deviations A 119875 value of less than 005 signifiedsignificant differences between comparisons

3 Results

The mean axial tibia loads throughout full rotation for eachspecimen are shown in Table 1 No statistical differences werefound between cycles of motion (119875 gt 005 for all compari-sons)Themaximum tibial force error was less than 1 of thetargeted value during dorsiflexion and less than 3 duringplantar flexion

Mean IAR values for specimen 1 and specimen 2 are pro-vided in Tables 2 and 3 respectively No statistical differenceswere found in mean119883 IAR values and 119885 IAR values betweencycles of dorsiflexion or plantar flexion for either specimen(119875 gt 005 for all comparisons) When comparing mean IARvalues for all three cycles between dorsiflexion and plantarflexion statistically significant values were found in themean119883 IAR values for both specimens (119875 lt 00001) and in themean 119885 IAR values in specimen 2 (119875 lt 00001)

The final 3 cycles of IAR pathways are plotted relativeto the ankle coordinate frame for specimen 1 (Figure 8) andspecimen 2 (Figure 9) along with mean IAR values and stan-dard deviationsThe IAR represents a moving axis of rotationdue to the joint articular geometry and surrounding softtissue structures Dorsiflexion paths show little movement inthe talus remaining within a 45mm range (ie 2mm in119883 and 4mm in 119885) In plantar flexion the first degrees ofmotion started up within the articular surface of the talus Asrotation continued the IAR path moved downward towardsthe middle of the talus The plantar flexion IAR path had amaximum range of approximately 21mm (ie 20mm in 119885and 6mm in 119883) A linear regression analysis was performedon the plantar flexion paths The slopes of the IAR pathsfor dorsiflexion testing are listed in Table 4 No statisticaldifferences were found between the slopes for specimen 1(119875 = 07823) or specimen 2 (119875 = 00826)


Calcaneus

Talus

Tibia

vGRF

Z

X

Ft

Fa

Wf

Ft = Fa + vGRF minus Wf

Figure 4 Force analysis of the foot and ankle used to set the parameters of loading where 119865119905

represented the axial tibia load 119865119886

was the ATload vGRF was the vertical ground reaction force and119882

119891

was the weight of the foot and tibia

Table 2 Mean IAR values for specimen 1

CycleXIAR (mm)

(Mean plusmn standard deviation)ZIAR (mm)

(Mean plusmn standard deviation)Dorsiflexion Plantar flexion Dorsiflexion Plantar flexion

1 minus295 plusmn 010 100 plusmn 130 minus900 plusmn 050 minus885 plusmn 520



Mean minus290 plusmn 005lowast

085 plusmn 030lowast

minus920 plusmn 020 minus895 plusmn 030

lowastStatistically significant between dorsiflexion and plantar flexion mean IAR values

4 Discussion

A novel dynamic robotic testing platform and protocol weredeveloped and used to investigate the biomechanical behav-ior of the foot and ankle under simulated loading conditionsthat are representative of mid-stance gait The first objectiveof this study was to validate the accuracy of a loading pro-tocol The protocol simulates a static Achilles load duringearly stance phase of walking with decreased in vivo loadingconditions The role of soft tissue in this protocol wasaccounted for by preconditioning of the cadaver feet bysubjecting them to several load and movement cycles beforerelevant data were captured The simulation accounted for20 degrees of plantar flexion and 10 degrees of dorsiflexionwhereas the total functional range of walking motion of theankle during stance phase is approximately 15 degrees [4 5 1134] Force errors were typically controlled within prescribedplusmn2N tolerances throughout testingTheATLswere limited bythe upper loading capacity of the gimbal load cell such thatloading conditions were set at half body weight (356N) and

the Achilles tendon force was set to half the vGRF (178N)Nevertheless the ground reaction forces produced during thesimulation were in agreement with a percentage of recordedin vivo forces [1 3 5 6 16 34] Likewise for the simulatedloading profiles the applied tibial force remained within 3of the targeted value

The second objective of the current study was to addressintraspecimen variability between cycles of motion Stabilityof the ankle joint is determined by three main factors artic-ular congruity ligamentous structures and ankle position[8 16] All have been shown to have high variability betweenspecimens [35] Baxter showed a difference in biomechanicalproperties of sprinters versus nonsprinters suggesting thatphysical health plays a major role in mechanics of the ankle[7] Therefore the goal was to demonstrate a repeatablemeasurement of the IAR in the sagittal plane for eachspecimen This loading protocol showed variability in themean IAR measurements between cycles of motion (asmeasured by the standard deviation) of less than 1mm in bothspecimens For the specimens tested the talus dimensions


ETT defined the initial pARA

Null Tool

Null tool tip (NTT)

X

Upper load cell

Plantar flexion Dorsiflexion

Z

X

Z

Fz

Fx

(a)

True ARA

+X

pARA after force reduction

+Z

(ΔZ)

(ΔX)

ΔMpARA

Fz

Fx

(b)

Figure 5 Kinematic and load control strategy (a) Kinematic coordinate frames of the foot and ankle defined by the NTT and extendedtool tip (ETT) of the RTP The ETT (orange dot) represents the initial prescribed IAR of the foot located at the center of the talus and was anegative 119885 offset from the NTT The coordinate frame of the load cell was initially located at the center of the load cell and transformed tothe origin of the moving joint center that is the ETT (b) An angular input of ankle rotation was given every 05 degrees during which theloading protocol reduced the ankle moment due to off-axis forces by minimizing the distance (Δ119883 Δ119885) between the pARA (upper) and thetrue ARA (lower) thus creating a ldquopure momentrdquo condition


CycleXIAR (mm)



1 minus430 plusmn 065 minus025 plusmn 135 minus640 plusmn 070 minus355 plusmn 495




minus010 plusmn 015lowast




Table 4 Mean values of IAR slopes for specimen 1 and specimen 2during plantar flexion

CycleIAR slope values

(Mean plusmn standard deviation)Specimen 1 Specimen 2

1 minus40 plusmn 02 minus35 plusmn 02



Mean minus40 plusmn 01 minus36 plusmn 03

were approximately 60mm in width and 38mm in heightdemonstrating small relative variability on a clinical scaleWhile dorsiflexion paths remained within a 45mm rangeplantar flexion paths showed more movement Howeverplantar flexion slopes were not significantly different between

cycles of motion in specimen 1 or specimen 2 demonstratinga repeatable measurement of the IAR path

Current radiographic practice for defining anatomicallocations on a radiograph is plusmn05mm at best using advancedsoftware programs like Medical Metrics Inc 2012 and ismore than double that amount when processed from stan-dard radiographs In this research a similar level of accuracywas introduced when selecting the initially prescribed anklerotation axis (pARA) However all subsequent measures ofthe change in the location of the rotational axis (ie the pathof the IAR) were dependent on the resolution of the robotrsquoscontrolled axes which was 0001mm Collectively the accu-racy of the measured IAR calculated was plusmn50 microns Thiswas a significant improvement over current IAR measure-ments of the ankle joint where high errors in the calculationof the IAR were reported due to test methodology [7 24]Similar errors in determining the location of the IAR havebeen observed in other joint motion studies For example


+X

True ARA

+Z

F998400z

F998400x

Figure 6 Kinematic and load control strategy for full load appli-cation simulation Once the balanced position for pure momentloading was achieved (establishing the true ARA) the additionalshear and compressive loads (1198651015840

119909

1198651015840

119911

) were introduced to maintainthe axial tibial force and vGRFs By passing these forces through therotational axis of the ankle the moment value was unchanged TheNTT data remained aligned with the rotated tibia axes Achilles loadis not shown

X

Z

minus30

minus25

minus20

minus15

minus10

minus5

minus5

minus10

minus15 5 10 15

Figure 7 The ldquoankle coordinate framerdquo designated to plot the IARon individual specimen It was defined as the intersection of bisectorof the tibia and the highest point on the tibial mortise

in spinal motion studies calculations of the IAR locationswere greater than 20mm for rotations less than 1 degree andapproximately 3 to 4mm for rotations up to 5 degrees [30]

Some limitations are present in this study Only twospecimens were studied However the goal of this study wasto demonstrate minimal variability in the IAR measurement

200

250 750 1250

Cycle 3 dorsiflexionCycle 3 plantar flexionMean IAR dorsiflexionMean IAR plantar flexion

Cycle 1 dorsiflexionCycle 1 plantar flexionCycle 2 dorsiflexionCycle 2 plantar flexion

Dorsiflexion

Plantar flexionminus1250 minus750 minus250

minus300

minus800

minus1300

minus1800

minus2300

ZIAR (mm)

XIAR (mm)

(minus290 plusmn 005 minus920 plusmn 020)

(085 plusmn 030 minus895 plusmn 030)

Figure 8 IAR paths of specimen 1 during dorsiflexion and plantarflexion plotted onmedial radiographs Mean IAR values for all threecycles (circles) with standard deviations (too small to see) are shownDashed arrows denote the direction of the IAR through rotation

200

100 1100 2100

Plantar flexion

Dorsiflexion

minus2900 minus1900 minus900minus300

minus800

minus1300

minus1800



ZIAR (mm)

XIAR (mm)


between cycles of motion in individual specimen Addition-ally the proximal fusion of the tibia and fibula removedthe physiologic joint function where the bones may undergotranslations and rotations relative to each other This hasbeen observed primarily when the knee experiences externaland internal rotation [36] Our protocol accounts for tibialsagittal plane motion only During force application an out-of-plane load was produced as a mode of lateral stabilizationof the joint However a negligible moment build-up wasobserved throughout motion (05Nm) and did not restrict


flexion Additionally the current study does not incorporatea dynamic Achilles tendon force profile representative of invivo loading conditions and it does not account for anyof the other major plantar flexors extrinsic dorsiflexors orintrinsic muscles of the foot However these muscles arepredominantly active during heel strike and heel rise to toe offgait [8 34]The force applied via the Achilles tendon has beenshown to have the greatest role in the biomechanical behaviorof the ankle joint during stance phase when the foot is flat[8 34] Therefore it is the only tendon force accounted for inthis study no other intrinsic muscles were simulated Duringnormal gait the tibia is driven frommaximumplantar flexionto dorsiflexion with a continuous load on the joint Thisprotocol is limited in that the dorsiflexion and plantar flexionpaths are generated independently Furthermore plantarflexion is driven in a direction uncharacteristic of the targetin vivo stance phase motion However the Achilles tendonload is applied to the specimen in such a manner to simulatethe in vivo loading scenario where the joint is plantar flexedThe independent movement from the neutral orientationbetween dorsiflexion and plantar flexion may explain thedisconnection between the initial points of IAR paths indorsiflexion and plantar flexion Another limitation is thatour study was two-dimensional whereas the ankle joint canmove in three dimensions [1 16 29] While the talus andcalcaneus were free to translate or rotate throughout motionthe IAR directly reflected the motion of the tibia defined bythe robot tool tipwhich did not deflect from the sagittal planeMotion during stance phase gait is typically limited to thesagittal plane with the exception of heel strike and heel rise[11 16 37]Therefore the IAR analysis could be limited to twodimensions Increments of rotation could have been reducedto provide a more thorough and accurate IAR path Howeverthis was a very good compromise to describe the path within5 of the total rotation

5 Conclusions

This work has provided description of a novel loading proto-col developed to support in vitro cadaveric testing of the footand ankle complex The protocol provided a repeatable mea-sure of the two-dimensional analysis of the IAR of the anklejoint using tightly controlled loads in conjunction with apassive AT forceThe low force error tolerance and consistentaxial force values demonstrate the RTPrsquos ability to accuratelysimulate forces The results of the study are one of a handfulto address the instantaneous axis of rotation of the ankle jointand provide the most accurate measurement to dateThe IARmeasurement was repeatable within one millimeter smallerthan what is feasibly measurable in a clinical setting Becausethe IAR measurement is a direct representation of soft tissuestructures and articular geometry future in vitro studiesmay yield great insight into the biomechanical propertiesof the foot and ankle within the sagittal plane includingarch formation and effects of orthotics and footwear onankle kinematics Furthermore much like in knee jointmechanics where an injury to the ligaments within thejoint can cause a shift in the IAR path of the knee similarrelationships may occur between the structures of the ankle

and its IAR properties Detection of a deviation in the IARpath can be associated with a specific injury mechanismand serve as a way to improve the corrective capacity ofany treatment As new information describing the dynamicloading characteristics of the foot and ankle is obtained fromin vivo gait studies (stair climbing orthotics and footwear)this novel RTP and test protocol can readily simulate theseconditions Current development includes incorporating adynamic Achilles tendon load a higher capacity load cellto accommodate forces comparable to in vivo loads and anoptoelectronic motion tracking system to study the interac-tion of the bones of the arch of the foot throughout anklerotation Future use of the testing protocol and platform canbe directed towards the study of other muscle groups on loadand motion behavior of the foot ankle complex

Competing Interests

The authors declare that they have no competing interests

Acknowledgments

The authors would like to acknowledge Brian Kelly PhDand Richard Kasser PhD for their medical and technicalcontributions to the project

References

[1] C Bishop G Paul and D Thewlis ldquoDefining standards formodeling the biomechanics of the foot and ankle a systematicreviewrdquo Journal of Foot and Ankle Research vol 4 supplement1 p O9 2011

[2] N Blitz Contemporary Controversies in Foot and Ankle SurgeryAn Issue of Clinics in Podiatric Medicine and Surgery ElsevierPhiladelphia Pa USA 1st edition 2012

[3] J Hamill and K M Knutzen Biomechanical Basis of HumanMovement LippincottWilliamsampWilkinsNewYorkNYUSA2006

[4] S E Jarrell J R Owen J S Wayne and R S Adelaar ldquoBiome-chanical comparison of screw versus platescrew construct fortalonavicular fusionrdquo Foot amp Ankle International vol 30 no 2pp 150ndash156 2009

[5] C Kirtley Clinical Gait Analysis Theory and Practice ElsevierNew York NY USA 2006

[6] M Nordin and V H Frankel Basic Biomechanics of the Muscu-loskeletal System Lippincott Williams amp Wilkins PhiladelphiaPa USA 3rd edition 2001

[7] J R Baxter T A Novack H vanWerkhoven D R Pennell andS J Piazza ldquoAnkle joint mechanics and foot proportions differbetween human sprinters and non-sprintersrdquo Proceedings of theRoyal Society B Biological Sciences vol 279 no 1735 pp 2018ndash2024 2012

[8] A S Kelikian and S Sarrafian Sarrafianrsquos Anatomy of theFoot and Ankle Descriptive Topographic Functional LippincottWilliams ampWilkins New York NY USA 3rd edition 2011

[9] C N Maganaris V Baltzopoulos and A J Sargeant ldquoChangesin Achilles tendon moment arm from rest to maximum iso-metric plantarflexion in vivo observations in manrdquo Journal ofPhysiology vol 510 no 3 pp 977ndash985 1998

[10] R Semple G S Murley J Woodburn and D E TurnerldquoTibialis posterior in health and disease a review of structure


and function with specific reference to electromyographic stud-iesrdquo Journal of Foot and Ankle Research vol 2 no 1 article 242009

[11] D AWinter Biomechanics andMotor Control of HumanMove-ment JohnWiley amp Sons Hoboken NJ USA 4th edition 2009

[12] A Leardini J J OrsquoConnor F Catani and S Giannini ldquoAgeometric model of the human ankle jointrdquo Journal of Biome-chanics vol 32 no 6 pp 585ndash591 1999

[13] A Leardini J J OrsquoConnor F Catani and S Giannini ldquoKine-matics of the human ankle complex in passive flexion a singledegree of freedom systemrdquo Journal of Biomechanics vol 32 no2 pp 111ndash118 1999

[14] A Leardini ldquoGeometry and mechanics of the human anklecomplex and ankle prosthesis designrdquo Clinical Biomechanicsvol 16 no 8 pp 706ndash709 2001

[15] A Leardini J J OrsquoConnor F Catani and S Giannini ldquoMobilityof the human ankle and the design of total ankle replacementrdquoClinical Orthopaedics and Related Research vol 424 pp 39ndash462004

[16] M D Castro ldquoAnkle biomechanicsrdquo Foot and Ankle Clinics vol7 no 4 pp 679ndash693 2002

[17] G J Sammarco A H Burstein andV H Frankel ldquoBiomechan-ics of the ankle a kinematic studyrdquo Orthopedic Clinics of NorthAmerica vol 4 no 1 pp 75ndash96 1973

[18] J Burg K Peeters T Natsakis G Dereymaeker J V Sloten andI Jonkers ldquoIn vitro analysis of muscle activity illustrates medio-lateral decoupling of hind and mid foot bone motionrdquo Gait ampPosture vol 38 no 1 pp 56ndash61 2013

[19] K-J Kim H B Kitaoka Z-P Luo et al ldquoIn vitro simulationof the stance phase in human gaitrdquo Journal of MusculoskeletalResearch vol 5 no 2 pp 113ndash121 2001

[20] A Suckel O Muller P Langenstein T Herberts P Reize andNWulker ldquoChopartrsquos joint load during gait In vitro study of 10cadaver specimen in a dynamic modelrdquo Gait and Posture vol27 no 2 pp 216ndash222 2008

[21] K Watanabe H B Kitaoka T Fujii et al ldquoPosterior tibial ten-don dysfunction and flatfoot analysis with simulated walkingrdquoGait amp Posture vol 37 no 2 pp 264ndash268 2013

[22] R Bahr F Pena J Shine W D Lew and L Engebretsen ldquoLiga-ment force and joint motion in the intact ankle a cadavericstudyrdquo Knee Surgery Sports Traumatology Arthroscopy vol 6no 2 pp 115ndash121 1998

[23] N A Sharkey and A J Hamel ldquoA dynamic cadaver model ofthe stance phase of gait performance characteristics and kineticvalidationrdquo Clinical Biomechanics vol 13 no 6 pp 420ndash4331998

[24] T Sumiya Y Suzuki T Kasahara andH Ogata ldquoInstantaneouscenters of rotation in dorsiplantar flexion movements of post-erior-type plastic ankle-foot orthosesrdquo Journal of RehabilitationResearch and Development vol 34 no 3 pp 279ndash285 1997

[25] C Hurschler J Emmerich and NWulker ldquoIn vitro simulationof stance phase gait Part I Model verificationrdquo Foot amp AnkleInternational vol 24 no 8 pp 614ndash622 2003

[26] C J Nester AM Liu EWard et al ldquoIn vitro study of foot kine-matics using a dynamic walking cadaver modelrdquo Journal of Bio-mechanics vol 40 no 9 pp 1927ndash1937 2007

[27] L D Noble Jr RW Colbrunn D-G Lee A J VanDen Bogertand B L Davis ldquoDesign and validation of a general purposerobotic testing system formusculoskeletal applicationsrdquo Journalof Biomechanical Engineering vol 132 no 2 pp 1ndash12 2010

[28] E C Whittaker P M Aubin and W R Ledoux ldquoFoot bonekinematics as measured in a cadaveric robotic gait simulatorrdquoGait amp Posture vol 33 no 4 pp 645ndash650 2011

[29] C J Nester ldquoLessons from dynamic cadaver and invasive bonepin studies do we know how the foot really moves during gaitrdquoJournal of Foot and Ankle Research vol 2 article 18 2009

[30] B P Kelly andD J DiAngelo ldquoAmultiaxis programmable robotfor the study ofmultibody spine biomechanics using a real-timetrajectory path modification force and displacement controlstrategyrdquo Journal of Medical Devices Transactions of the ASMEvol 7 no 3 Article ID 034502 2013

[31] D B Thordarson H Schmotzer J Chon and J PetersldquoDynamic support of the human longitudinal arch a biomech-anical evaluationrdquo Clinical Orthopaedics and Related Researchno 316 pp 165ndash172 1995

[32] A Aquino and C Payne ldquoFunction of the plantar fasciardquo TheFoot vol 9 no 2 pp 73ndash78 1999

[33] J J Crisco III X Chen M M Panjabi and S W WolfeldquoOptimal marker placement for calculating the instantaneouscenter of rotationrdquo Journal of Biomechanics vol 27 no 9 pp1183ndash1187 1994

[34] MM Rodgers ldquoDynamic biomechanics of the normal foot andankle during walking and runningrdquo Physical Therapy vol 68no 12 pp 1822ndash1832 1988

[35] R P Kleipool and L Blankevoort ldquoThe relation between geom-etry and function of the ankle joint complex a biomechanicalreviewrdquoKnee Surgery Sports Traumatology Arthroscopy vol 18no 5 pp 618ndash627 2010

[36] J Scott H Lee W Barsoum and A J Van Den Bogert ldquoTheeffect of tibiofemoral loading on proximal tibiofibular jointmotionrdquo Journal of Anatomy vol 211 no 5 pp 647ndash653 2007

[37] S Yamaguchi Y Tanaka S Kosugi Y Takakura T Sasho andS A Banks ldquoIn vivo kinematics of two-component total anklearthroplasty during non-weightbearing andweightbearing dor-siflexionplantarflexionrdquo Journal of Biomechanics vol 44 no 6pp 995ndash1000 2011

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



out of the sagittal plane during stance phase gait reducingthe need for analyses of ankle jointmotion to two dimensions(ie IAR) Quantifying the IAR of the ankle joint during gaitmay have the potential to advance the understanding of thebiomechanical properties of the foot and ankle includingarch formation ankle arthroplasty design and surgical tech-nique and effects of orthotics and footwear

Current gait simulators and biomechanical testing proto-cols either are unable to analyze the IAR chose not to includethis parameter as part of the analysis [4 16 18ndash23 25ndash29] orare limited to high errors in the calculation of the IAR due tomethodology [7 24 29] There is a need for a biomechanicaltesting platform and protocol that will provide simulationof controlled tibia and Achilles tendon (AT) loads withoutconstraining the foot ankle kinematic profile

The purpose of this work was to establish a 2D testingmethod to determine the location of the IAR of the anklejoint under simulated stance phase conditions in a humancadaveric model Additional tests were done to confirmthe accuracy and repeatability of the IAR measurementOnce the rotational axis is known (describing where theankle motion occurs) a more in-depth analysis of the anklejointrsquos mechanics can be undertaken (such as convertingthe external joint moment into its internal components ofa muscle force times a moment arm) a level of in vitroanalysis currently not possible with current testing methodsFurther analysis of the path of the ankle jointrsquos rotational axisoffers new insight into the effects that disease and surgicalalterations have on the ankle kinematics and mechanics

2 Materials and Methods

21 Specimen Preparation A matched pair (male age of37 and body weight of 712N) of human below knee lowerextremity specimens was procured from Restore Life USA(Johnson City TN) Specimens were frozen at minus20∘C Beforepreparation the feet were thawed in a refrigeration systemfor two days Medial and lateral radiographs were used toverify the absence of anatomical abnormalities or surgerySoft tissue andmuscle were resected to expose approximately100mm of the proximal shafts of the tibia and fibula A singleone inch 6 wood screw was then placed across the proximalends of the tibia and fibula to stabilize the bones and addadditional fixation for the potting material After the AT wasexposed and soft tissue was excised to allow approximately150mmofAT length for clamping a cable puller was attachedand secured with a U-bolt to increase clamping powerThe tibial shafts were then mounted concentrically into acylindrical pot using an alignment frame to position the tibiain a neutral vertical orientation Low-melting-point bismuthalloy (Rotometals Inc San Leandro CA) was used to fix thetibia and fibula together and create a mounting fixture forattachment to the robotic testing platform (RTP) The finalprepared specimen is shown in Figure 1

22 Robotic Testing Platform An existing custom designedmultiaxis testing platform was utilized to simulate aspectsof stance phase gait mechanics under displacement controland force feedback (Figure 2) [30] Two linear actuators a

Mounting fixture

U-bolt

Tendon cable puller

Fixture reference

point

Figure 1 A prepared below the knee lower extremity specimenTissue was resected from the proximal shafts of the tibia fibulaand AT A cable puller was attached to the AT for tendon loadapplication Bismuth alloy material was used to create a mountingfixture for the proximal shafts of the tibia and fibula for rigid fixationto the testing platformThe fixture reference point was located at thecenter of the upper mounting fixture

Parker Hannifin Corp (Cleveland OH) 406XR series linearball screw actuator and an Exlar (Chanhassen MN) GSX-30 linear roller screw actuator were aligned in 119909-axis and 119911-axis respectively Control specifications of platform actuatorswere 2 microns in 119883 and 031 microns in 119885 The rotarymotors and drive units of the original test RTPwere upgradedwith Harmonic Drive units (Peabody MA model FHA-25C-160-US250) having improved resolutions of 0008 degreesPrevious position control pilot trials demonstrated that eachindividual axis was controllable to within tolerance of plusmn2times the resolution of each feedback sensor corresponding toplusmn00091 degrees plusmn4microns in 119911 direction and plusmn06micronsin 119909 direction for sagittal plane movement

Two six-axis load cells (JR3 Inc Woodland CA models100M40 and 45E15S) were attached to the robotic testingplatform assembly and measured the three orthogonal forcesandmoments applied to the tibia via the gimbal assembly andforces transferred through the foot to the ground at the baseof an 119883-119884 table Load cell resolutions and capacities were800 plusmn 044N in 119911-axis 400 plusmn 009N in 119909-axis and 119910-axisand 40 plusmn 002Nm about all axes respectively The load cellrated accuracies were 10 of full scale with linearity errorof 01 full scale The robot testing system actuator fixturesand load cells were significantly greater in stiffness thanthose of human tissue such that any deflection under testloading was negligible (ie less than 50 microns for 1000Nload) All reported deformation represented the compliancyof specimen tissue

Potted specimens were clamped securely in a mountingblock and rigidly connected to the testing platform A sagittalplane was established for each specimen by bisecting thesecond metatarsal and AT [11 13 17] with a vertical axisaligned with the tibia that matched the 119883-119885 plane of thetesting platform A static AT force vector (119865

119886) was applied

via a cable-pulley system designed to minimize friction to


Rotation

X

Y

W

Z

Pulley

Achilles load

table

Pulley

Base load cell

Mounting block

Upper loadcell

X-Y

about Y

Translation in Z

Translation in X






119909 119865119911)








X

Z




119909

and 1198651015840119911







1 5165 plusmn 10 5060 plusmn 15

2 5195 plusmn 10 5135 plusmn 05


3 Results





Calcaneus

Talus

Tibia

vGRF

Z

X

Ft

Fa

Wf





119891



CycleXIAR (mm)










4 Discussion






Null Tool

Null tool tip (NTT)

X

Upper load cell


Z

X

Z

Fz

Fx

(a)

True ARA

+X


+Z

(ΔZ)

(ΔX)

ΔMpARA

Fz

Fx

(b)



CycleXIAR (mm)






















+X

True ARA

+Z

F998400z

F998400x


119909

1198651015840

119911


X

Z

minus30

minus25

minus20

minus15

minus10

minus5

minus5

minus10

minus15 5 10 15




200

250 750 1250



Dorsiflexion


minus300

minus800

minus1300

minus1800

minus2300

ZIAR (mm)

XIAR (mm)




200

100 1100 2100

Plantar flexion

Dorsiflexion


minus800

minus1300

minus1800



ZIAR (mm)

XIAR (mm)





5 Conclusions



Competing Interests


Acknowledgments


References










































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Rotation

X

Y

W

Z

Pulley

Achilles load

table

Pulley

Base load cell

Mounting block

Upper loadcell

X-Y

about Y

Translation in Z

Translation in X






119909 119865119911)








X

Z




119909

and 1198651015840119911







1 5165 plusmn 10 5060 plusmn 15

2 5195 plusmn 10 5135 plusmn 05


3 Results





Calcaneus

Talus

Tibia

vGRF

Z

X

Ft

Fa

Wf





119891



CycleXIAR (mm)










4 Discussion






Null Tool

Null tool tip (NTT)

X

Upper load cell


Z

X

Z

Fz

Fx

(a)

True ARA

+X


+Z

(ΔZ)

(ΔX)

ΔMpARA

Fz

Fx

(b)



CycleXIAR (mm)






















+X

True ARA

+Z

F998400z

F998400x


119909

1198651015840

119911


X

Z

minus30

minus25

minus20

minus15

minus10

minus5

minus5

minus10

minus15 5 10 15




200

250 750 1250



Dorsiflexion


minus300

minus800

minus1300

minus1800

minus2300

ZIAR (mm)

XIAR (mm)




200

100 1100 2100

Plantar flexion

Dorsiflexion


minus800

minus1300

minus1800



ZIAR (mm)

XIAR (mm)





5 Conclusions



Competing Interests


Acknowledgments


References










































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










X

Z




119909

and 1198651015840119911







1 5165 plusmn 10 5060 plusmn 15

2 5195 plusmn 10 5135 plusmn 05


3 Results





Calcaneus

Talus

Tibia

vGRF

Z

X

Ft

Fa

Wf





119891



CycleXIAR (mm)










4 Discussion






Null Tool

Null tool tip (NTT)

X

Upper load cell


Z

X

Z

Fz

Fx

(a)

True ARA

+X


+Z

(ΔZ)

(ΔX)

ΔMpARA

Fz

Fx

(b)



CycleXIAR (mm)






















+X

True ARA

+Z

F998400z

F998400x


119909

1198651015840

119911


X

Z

minus30

minus25

minus20

minus15

minus10

minus5

minus5

minus10

minus15 5 10 15




200

250 750 1250



Dorsiflexion


minus300

minus800

minus1300

minus1800

minus2300

ZIAR (mm)

XIAR (mm)




200

100 1100 2100

Plantar flexion

Dorsiflexion


minus800

minus1300

minus1800



ZIAR (mm)

XIAR (mm)





5 Conclusions



Competing Interests


Acknowledgments


References










































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Calcaneus

Talus

Tibia

vGRF

Z

X

Ft

Fa

Wf





119891



CycleXIAR (mm)










4 Discussion






Null Tool

Null tool tip (NTT)

X

Upper load cell


Z

X

Z

Fz

Fx

(a)

True ARA

+X


+Z

(ΔZ)

(ΔX)

ΔMpARA

Fz

Fx

(b)



CycleXIAR (mm)






















+X

True ARA

+Z

F998400z

F998400x


119909

1198651015840

119911


X

Z

minus30

minus25

minus20

minus15

minus10

minus5

minus5

minus10

minus15 5 10 15




200

250 750 1250



Dorsiflexion


minus300

minus800

minus1300

minus1800

minus2300

ZIAR (mm)

XIAR (mm)




200

100 1100 2100

Plantar flexion

Dorsiflexion


minus800

minus1300

minus1800



ZIAR (mm)

XIAR (mm)





5 Conclusions



Competing Interests


Acknowledgments


References










































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











Null Tool

Null tool tip (NTT)

X

Upper load cell


Z

X

Z

Fz

Fx

(a)

True ARA

+X


+Z

(ΔZ)

(ΔX)

ΔMpARA

Fz

Fx

(b)



CycleXIAR (mm)






















+X

True ARA

+Z

F998400z

F998400x


119909

1198651015840

119911


X

Z

minus30

minus25

minus20

minus15

minus10

minus5

minus5

minus10

minus15 5 10 15




200

250 750 1250



Dorsiflexion


minus300

minus800

minus1300

minus1800

minus2300

ZIAR (mm)

XIAR (mm)




200

100 1100 2100

Plantar flexion

Dorsiflexion


minus800

minus1300

minus1800



ZIAR (mm)

XIAR (mm)





5 Conclusions



Competing Interests


Acknowledgments


References










































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










+X

True ARA

+Z

F998400z

F998400x


119909

1198651015840

119911


X

Z

minus30

minus25

minus20

minus15

minus10

minus5

minus5

minus10

minus15 5 10 15




200

250 750 1250



Dorsiflexion


minus300

minus800

minus1300

minus1800

minus2300

ZIAR (mm)

XIAR (mm)




200

100 1100 2100

Plantar flexion

Dorsiflexion


minus800

minus1300

minus1800



ZIAR (mm)

XIAR (mm)





5 Conclusions



Competing Interests


Acknowledgments


References










































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











5 Conclusions



Competing Interests


Acknowledgments


References










































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation








































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









research article development of a robotic assembly for...

Documents