Eyeglasses-powered, contact lens-like platform with high powertransfer efficiency

Young-Joon Kim1& Jimin Maeng2 & Pedro P. Irazoqui2

# Springer Science+Business Media New York 2015

Abstract We present a contact lens-like platform that is wire-lessly powered by an external coil embedded in eyeglasses viamagnetic resonance coupling at 13.56 MHz. The platform iscomposed of a transparent parylene film as a host substrate, anembedded spiral inductor as a power receiving coil, and metalinterconnects for additional electronics. A multilayer thin-filmparylene packaging process is used to meet the form factor ofa contact lens. A 36 μm-thick metal plating technique isemployed on a parylene film to enhance the quality factor(Q) of the receiving coil (Q=27.3 at 13.56 MHz). The powertransfer method and techniques to compensate for coil mis-alignment are demonstrated on a pig eye, achieving a powertransfer efficiency of 17.5 % at a 20-mm powering distance.The effect of tissue on the coil and the power transfer efficien-cy is examined. The high power transfer efficiency along withthe wearable prototype demonstrated herein make promisingprogress toward smart contact lens in ocular diagnostics.

Keywords Contact lens . Magnetic resonance coupling .

Ocular implants . Power transfer efficiency .Wearable .

Wireless power transfer

1 Introduction

Over the past few years, a contact lens-like smart sensor hasbeen receiving spotlight for its minimal-invasiveness and end-user friendliness (Farandos et al. 2015). Studies on contactlens sensors are especially promising since its scenarios canbe extended to monitoring symptoms in various diagnosticcriteria, including intraocular pressure for glaucoma (Leonardiet al. 2009; Mansouri et al. 2012), glucose level for diabetes(Yao et al. 2011), and lactate level for metabolic diseases(Thomas et al. 2012). Development of these contact lens sen-sors poses unique challenges. For example, the device has tosatisfy the physical parameter constraints of a contact lens,which includes the contact lens diameter (typically 12–14 mm, (McNamara et al. 1999)) and the base curve radius(BCR, typically 8–10 mm, (Farandos et al. 2015)). The mate-rial has to be biocompatible, transparent, and flexible to beeligible for ocular implant (Meng and Sheybani 2014). Also,the device requires a reliable power source embedded withinthe given device form factors to support the related electron-ics. Despite substantial miniaturization in recent years, thebattery is still considered large compared to othermicrofabricated components of the sensor (Meng andSheybani 2014). In biomedical applications, the limited life-time of batteries and the risk of malfunction or leakage poseadditional risks to the end user (Si et al. 2007). To overcomesuch restrictions for ocular sensing technology, researchgroups have integrated wireless power transfer technologyinto their contact lens electronics; however, most groups uti-lize high frequency electromagnetic (EM) radiation, whichmay have negative health effects (Leonardi et al. 2009;Pandey et al. 2010). EM radiation also suffers from low powertransfer efficiency (PTE) due to the small form factor and poorantenna performance. Some groups employ lower frequencymagnetic coupling for ocular device power transfer, but are

Young-Joon Kim and Jimin Maeng contributed equally to this work.

* Jimin Maengjmaeng@purdue.edu

1 Center for Implantable Devices, School of Electrical and ComputerEngineering, Purdue University, West Lafayette, IN 47907, USA

2 Center for Implantable Devices, Weldon School of BiomedicalEngineering, Purdue University, West Lafayette, IN 47907, USA

Biomed Microdevices (2015) 17:75 DOI 10.1007/s10544-015-9979-0

not able to fabricate coils appropriate for a contact lens (Zhaoet al. 2012). Others use solar energy as a power source (Chenet al. 2011), but scarce power availability impedes systemperformance and usability. Therefore, development of an effi-cient and reliable wireless powering scheme is called for tomake the smart contact lens technology more viable.

In this study, we propose a parylene-based contact lensplatform, wirelessly powered by an external coil embeddedin eyeglasses via magnetic resonance coupling (Kurs et al.2007) at 13.56 MHz. Magnetic resonance coupling is chosenas the powering method since it can provide a substantiallyhigher PTE than a high-frequency EM radiation method. Anear-theoretically-maximum PTE can be achieved at aneyeglass-to-eye scenario under variation in distance or orien-tation. A theoretical background as well as the design methodfor the magnetic resonance coupling is described. Due to itsbiocompatibility, transparency, flexibility, and thickness con-trollability, parylene serves as the host substrate (Ha et al.2012; Salvatore et al. 2014; Maeng et al. 2015). Once thedevice components are implemented on the substrate,parylene also serves as an effective sealing material. A multi-layer thin-film parylene packaging process and a thick metalplating technique are utilized to implement an embeddedhigh-Q receive coil to achieve a high PTE. The variation inthe coil Q under device bending is assessed to confirm thefeasibility to be used as a contact lens with a curvature. Theeffect of coil misalignment and the compensation techniquesare provided. Lastly, the effect of EM field on human tissue isexamined for safety consideration.

2 Experimental design and methods

2.1 Concept

This report details a highly efficient wireless power transfer(WPT) platform for smart ocular diagnostics embedded into acontact lens (Fig. 1). A contact lens-like device ismicrofabricated on a parylene substrate and wirelesslypowered by an external coil embedded in eyeglasses(Fig. 1a–c). A high-Q thin film coil is integrated in theparylene platform for high-efficiency wireless power transfer.As shown in Fig. 1d, the physical parameters of the devicematch those of a contact lens. The inner diameter of the con-tact lens coil is larger than the human pupil to avoid interferingwith vision. Since eyeglasses are very natural in everyday life,the external transmit coil can be attached or embedded oneyeglasses. The power is delivered wirelessly from the exter-nal coil to the eye-worn parylene platform by 13.56-MHzmagnetic resonance coupling (MRC), a means for deliveringa near-distance wireless power with high efficiency but withsmall impact on the tissue. This report describes the 1) designand fabrication of a high-Q eye-wearable coil on a parylene

platform, 2) design, evaluation, and optimization of wirelesspower transfer into the eye-worn platform, and 3) a safetyconsideration with an EM field analysis.

2.2 Wireless power transfer theory and design

In this work, wireless power transfer via magnetic resonancecoupling is implemented by using a resonator-coupledbandpass filter (BPF) design technique in microwave engi-neering. For a single transmitter to a single receiver MRCdesign, the wireless power transfer scheme is identical to aresonator-coupled two-pole BPF model. The advantage ofusing this model is that the PTE can be predicted by calculat-ing the insertion loss of the system. The characteristics of thebandpass filter are obtained by deriving the impedance matrix(1) and its correspondingmatrices (2,3) (Cameron et al. 2007).

Z ¼ Rþ sI þ jM ð1Þ

The coupling matrix M is described in (2) where subscriptsS and L denote source and load, respectively. The unloaded Q(Qi) of the resonator is included in the calculation by (3) whereFBW denotes the fractional bandwidth of the passband. Thediagonal entries in R and I matrices indicate the loads andresonators in the network, respectively.

M ¼0 MS1 0 0

MS1 −M 11 M 12 00 M 12 −M 22 M 2L

0 0 M 2L 0

2664

3775 ð2Þ

Mii ¼ − jQi FBW

ð3Þ

R ¼1 0 0 00 0 0 00 0 0 00 0 0 1

2664

3775; sI ¼

0 0 0 00 s 0 00 0 s 00 0 0 0

2664

3775 ð4Þ

s ¼ j

FBW

w

w0−w0

w

� �ð5Þ

Once the impedance matrix is derived, the insertion loss iscalculated by (6), which can be made identical to the PTE ofthe system.

S21 ¼ 2Z−14;1 ð6Þ

To maximize efficiency, we seek the external couplingvalues of MS1 and M2L that maximizes /S21/. The couplingvalues are then effectively converted to lumped componentsas shown in Fig. 2a (Ha et al. 2014).

The electrical parameters of the receive coil on parylene,such as inductance (L) and quality factor, are mainly governedby the physical parameters of a contact lens. The design of thereceive coil is described in Section 2.3. On the other hand, the

75 Page 2 of 9 Biomed Microdevices (2015) 17:75

physical dimension of the external transmit coil has fewerrestrictions. We use a 45-mm-diameter AWG-14 wire for theexternal transmit coil, which fits in a typical circular shapedeyeglasses. The electrical parameters of the transmit coil aremeasured in advance as shown in Fig. 2b. Based on the pa-rameter values obtained, a theoretical calculation predicts aninsertion loss of −5.05 dB (equivalent to a PTE of 31.3 %)assuming a 20-mm powering distance (i.e., eyeglasses-to-eyedistance) (Fig. 2d).

The human eye is frequently in motion and the loca-tion of the external eyeglasses can change, which willvary the coupling coefficient and degrade PTE. As aneffort to achieve maximum efficiency throughout thevariation in coupling strength, the capacitance at thetransmit side is implemented as a programmable capac-itor bank (Fig. 2c) by using relay switches (TQ2-L-5 V,Panasonic). The discrepancy of the capacitance is lessthan 12 % from the target capacitance condition in thecurrent scenario. Although it is best to implement asimilar adaptive tuning technique in the receive side aswell, the physical dimension of a contact lens poses achallenge. Therefore, the efficiency compensation net-work is only implemented in the transmit side whilethe receive side has a fixed condition. Even though

the compensation happens only in the transmit side,calculation shows over 90 % of the theoretical maxi-mum efficiency can be recovered throughout thepowering distance of 5–30 mm (Fig. 2e).

2.3 Coil design

The receive coil embedded on the parylene platformshould be carefully designed to maximize the PTE,while meeting the size requirements of a contact lens.We design our coil as an octagonal spiral inductor(Fig. 1d) to achieve the required L and Q valuesassessed in advance. First, the outer diameter (do inFig. 1d) and the inner diameter (di in Fig. 1d) of thecoil is determined as 10 mm and 6 mm, respectively,accounting for the size of a typical commercial softcontact lens (14 mm) and the diameter of a humanpupil (2–6 mm for different luminance levels (Winnet al. 1994)). Then, the number of turns (N), coil width(W), and coil space (S) are determined as N=3, W=0.5 mm, S=0.25 mm to obtain a reasonable inductancevalue (>100 nH) required for our powering design. Giv-en these parameters, the way to maximize Q is to in-crease the metal thickness to minimize the series

Fig. 1 System design. (a) External coil is attached or embedded toeyeglasses. A thin and small coil is embedded on wearable paryleneplatform and receives power from eyeglasses. (b) Conceptdemonstration of eyeglasses-powered contact lens. Parylene platformsits on eye and powered by external coil attached to eyeglasses. (c) A

close up of (b). (d) Description of coil size onwearable parylene platform.The inner diameter (di) of the coil is larger than the human pupil (dp, 2–6 mm) so that vision is not impaired. The outer diameter (do) is smallerthan a typical size of a soft contact lens (14 mm)

Biomed Microdevices (2015) 17:75 Page 3 of 9 75

resistance of the coil. Simulation results show that theQ improves steadily as the metal thickness increases butbeyond a thickness of 30 μm, the increment becomesvery small while fabrication becomes difficult (Fig. 2f).Here, we employ a 36 μm-thick Cu electroplating. On aside note, it is confirmed by simulation that with anoctagonal-shape spiral design, Q increases by 10 %while L decreases by 10 % compared to a rectangulardesign.

2.4 Fabrication

The fabrication of the proposed contact lens platform is basedon a thin-film multilayer parylene packaging technology(Maeng et al. 2011), which allows conformal deposition andpatterning of successive thin (<5μm)metal/parylene layers on

a parylene host substrate (5–30 μm, typically) and their mul-tilayer interconnect through micro via-holes. In this work, athick (36μm)metal electroplating is uniquely combined to theparylene packaging process to achieve a high-Q coil that en-hances the power receiving capability. The final package isreleased from a rigid carrier such that a flexible electronics isachieved on a freestanding, transparent, and biocompatibleparylene film. This fabrication strategy allows the device tomeet the physical requirements of a contact lens in terms ofsize, thickness, transparency, biocompatibility, andbendability.

Figure 3 illustrates the contact lens platform fabricationprocess. First, a sacrificial photoresist (PR) layer is spin-coated and cured on a standard silicon wafer. On top of thePR layer, a 30-μm-thick parylene film is coated using a room-temperature parylene deposition system (PDS 2010, Specialty

Fig. 2 a Circuit diagram of the wireless power transfer betweeneyeglasses and parylene platform. b Measured electrical parameter ofthe external coil on eyeglasses. c The compensation network fordistance variation. d Calculated frequency response of the WPT system.

e Comparison of the calculated maximum achievable PTE and the PTEobtained by tuning the capacitance in transmit (TX) side. f Simulated Qfor receive coil embedded in parylene over different metal thickness (inair)

75 Page 4 of 9 Biomed Microdevices (2015) 17:75

Coating Systems, Inc.) (Fig. 3a). This parylene film serves as ahost substrate for the contact lens. Then, the first metal layer(Ti/Au, 1 μm) is deposited using a sputtering system (MAN-TIS Deposition Ltd.) and patterned by chemical wet etching(Fig. 3b). Another layer of parylene (2 μm) is deposited as aninsulator and via-holes are formed by an oxygen plasma reac-tive ion etching (RIE) process to allow interconnects to theupper metal layer (Fig. 3c). Next, a thick (36 μm) secondmetal layer is formed by a series of standard electroplatingprocess including seed metal deposition (Ti/Cu, 0.1 μm)(Fig. 3d), PR masking (Fig. 3e), Cu electroplating (Fig. 3f),and PR removal followed by seed metal etching (Fig. 3g).Here, we form a thick (40 μm) PR mask by spin-coating andcuring a positive photoresist (AZ 9260) twice at a low rotationspeed to allow the metal thickness higher than 30 μmwhich isrequired for obtaining a reasonable Q of the coil. Cu is chosenas the plating metal for its higher electrical conductivity thanthat of Au, which is also favorable for enhancing Q of the coil.To prevent Cu oxidation, we coat a thin (0.2 μm) Au layer ontop of the plated Cu. The parylene package is then releasedfrom the silicon carrier wafer by cutting around the edge anddissolving the sacrificial PR in an acetone solution (Fig. 3h).In this way, a contact lens-like platform with an embeddedpower receiving coil is achieved on a single, freestandingparylene film (Fig. 3i, left). It is important to note that otherelectronic components, such as sensors and chips, can be in-tegrated into this platform by surface mounting them on theextra pads formed by the second metal layer. Finally, a curva-ture is created on the platform to conform to the curvature ofan eyeball (Fig. 3i, right). We followed a heat molding process

described in (Li et al. 2006), and were able to achieve a BCRof 9 mm. In Fig. 3j–l, the fabricated contact lens platform isshown in view of different magnifications.

Although parylene serves as a biocompatible, transparent,and flexible substrate material, low oxygen permeability ofparylene film (Spivack 1970) could result in corneal discom-fort with prolonged use (Lingley et al. 2011). To mitigate theconcerns, it will be necessary to make diffusion windows onthe fabricated parylene package (i.e., Fig. 3i) and encase it in agas permeable hydrogel material (e.g., polyHEMA) using anencapsulation molding technique (Yao et al. 2012).

3 Results and discussion

3.1 Coil characterization

The embedded coil on the parylene platform is characterizedin air (Fig. 4a) as well as on a pig eye (Fig. 4b), accounting foreyeglasses-to-eye powering scenario. A pig eye was chosendue to its similarity to a human eye (Leonardi et al. 2009). Anetwork analyzer (N5230C, Agilent) was used for S-parameter measurements to extract L and Q values.Figure 4c and d show the measured L and Q of this coil overa frequency span of 1–100 MHz, in comparison with theHFSS simulation results. At 13.56 MHz, the measured valuesare obtained as L=108.8 nH and Q=27.3 in air, and L=108.6nH and Q=24.9 on a pig eye. These Q values are comparableto those of state-of-the-art inductors with similar frequenciesof interest (Wu et al. 2013; Sun et al. 2014). The drop of Q by

Fig. 3 a–h Contact lens platformfabrication process. a Depositionof sacrificial PR and parylene-Cfilm on Si carrier wafer. bFormation of first metal layer. cDeposition of parylene insulatinglayer and formation of via-holes.d Deposition of seed metal layer.e Thick PR masking. fElectroplating of second metallayer. g PR strip and seed metaletching. h Package releasing fromcarrier wafer. i Photo images offabricated device before (left) andafter (right) heat molding. j–lPhoto images of fabricated deviceviewed in differentmagnifications

Biomed Microdevices (2015) 17:75 Page 5 of 9 75

9 % on a pig eye can be explained by the parasitic parallelcapacitance generated in the presence of a tissue under thecoil, and hence the decrease in the self-resonant frequency ofthe coil. This aspect is illustrated in Fig. 4d and agrees with thesimulation.

In addition, we perform a series of bending tests for theembedded coil on parylene to investigate its parametervariation (i.e., change in L and Q values) under devicebending (Fig. 5). Because such parameter variation willdirectly affect the overall PTE, it is important to charac-terize them prior to the final coil design. Figure 5a illus-trates the device bending and the bending radius, ρ. Thecoil is measured for different degrees of bending as ρ=10 mm (Fig. 5b), ρ=5 mm (Fig. 5c), ρ=2.5 mm (Fig. 5d).Here, various bending radii are tested to obtain informa-tion useful not only for this contact lens study but also forpotential applications with different bending require-ments. The measured L and Q values are plotted inFig. 5e and f in comparison to the case with no devicebending. As shown, the changes in L and Q are not sig-nificant (ΔL=2.3 %, ΔQ=0.9 % at 13.56 MHz) when thedevice bending is mild as ρ=10 mm. One would expect asimilar result on the curvature of an eye with a typicalBCR of 8–10 mm. On the other hand, the effect of bend-ing becomes non-trivial (ΔL=12.5 %, ΔQ=12.1 % at13.56 MHz) for a bending of ρ=2.5 mm, which oneshould take into his design consideration for other appli-cations where a harsher device bending is required.

3.2 Power transfer efficiency (PTE)

To evaluate wireless power transfer in the fabricated con-tact lens platform, PTE was measured with having thedevice in air and on a pig eye as shown in Fig. 6a andb, respectively. In both cases, the device was aligned tothe external transmit coil while the vertical distance waschanged from 5 to 30 mm. A compensation techniquedescribed in Section 2.2 was used in the measurementsfor varying distances to help recover the PTE degradationdue to the mismatch induced by the change in couplingcoefficient. A 2-W power source was generated using awaveform generator (33522A, Agilent) and a power am-plifier (ZHL-1-2 W-S+, Mini-Circuits). The power re-ceived at the contact lens platform was measured in a50-Ω loading condit ion using a signal analyzer(N9010A, Agilent). Figure 6d shows the measured PTE.The PTE is slightly lower in a pig eye than in air becauseelectrical characteristics of the coil degrade due to thetissue in proximity (as predicted in Fig. 4d). A PTE of17.5 % was achieved on a pig eye when powered by anexternal transmit coil located 20 mm away, which is thedistance between eyeglasses and an eye. For a 5-mmpowering distance, a PTE of 42.3 % was obtained. ThePTE measurement results obtained herein show signifi-cant improvement compared to the state-of-the-art WPTsystems in similar applications (Pandey et al. 2010;Lingley et al. 2011; Varel et al. 2014).

Fig. 4 Electrical parameters of coil measured a in air and b on pig eye. c Measurement of inductance compared with simulation. d Quality factor

75 Page 6 of 9 Biomed Microdevices (2015) 17:75

A voltage doubler and a light-emitting diode (LED) areintegrated into the contact lens platform to verify the energyharvesting scenario in the contact lens prototype (Fig. 6c).Figure 6e illustrates a schematic of this energy harvestingcircui t . Here , the LED (SML-P11VTT86, RohmSemiconductor) at the output load represents any power con-suming microchip. For a 50 mW input power at a 20 mmdistance, the LED illuminated in red (Fig. 6c), indicating suc-cessful power conversion and delivery to the load from anexternal power source (coil on eyeglasses).

3.3 SAR analysis

In order to ensure safety in practical use, we consider themaximum transmit power in a human body model. At thefrequency of interest (13.56 MHz), tissue heating is the majoreffect of absorption. This is usually measured by the specificabsorption rate (SAR) which should not exceed 2W/kg as perthe guidelines of International Commission on Non-IonizingRadiation Protection (ICNIRP). Figure 7 shows the resultinglocalized SAR distribution in a HFSS human body modelwhere the peak value (0.021 W/Kg) is much less than therecommended safety limit. The transmit coil is 12 mm awayfrom the eye, excited by a 2-W source. This result, combinedwith the PTE achieved in Section 3.2, indicates that over350 mW can be safely delivered to the device within a 20-

mm distance. This is a sufficient power to run anymicrosystem designed for an ocular implant (Chandrakasanet al. 2008).

4 Conclusions

In conclusion, we demonstrate a highly-efficient,parylene-based contact lens platform that can be wireless-ly powered by eyeglasses. A PTE of 17.5 % has beenachieved on a pig eye at a 20-mm powering distance, byvirtue of a magnetic resonance-based power transfer de-sign accounting for tissue effect. A compensation tech-nique for coil misalignment has been implemented andnear theoretical maximum PTE was achieved for distancevariation. The PTE achieved here is a significant improve-ment over the previously reported ocular implant deviceswith an EM antenna or two-coil based powering scheme.The SAR simulation results show that over 350 mW canbe safely delivered to the device within a-20 mm distance,which opens up a variety of applications. Parylene isshown to meet the requirements for a base material for acontact lens owing to its favorable material properties aswell as thin-film process compatibility. A high-Q (Q=27.3 at 13.5 MHz) thin-film receive coil has beenachieved on a parylene film and shown not to undergo a

Fig. 5 Device (coil) bending test. (a) Illustration of device bending andbending radius, ρ. (b)–(d) Photo images of measurement setup fordifferent degrees of device bending with (b) ρ=10 mm, (c) ρ=5 mm,

and (d) ρ=2.5 mm. (e) Measured inductance for different degrees ofdevice bending. (f) Measured quality factor for different degrees ofdevice bending

Biomed Microdevices (2015) 17:75 Page 7 of 9 75

significant performance variation under a moderate devicebending, which promises wear on an eye and wireless

powering. For future work, a study on sensor integrationand hydrogel encapsulation is ongoing.

Fig. 6 Measurement of power transfer to parylene contact lens platform.a PTE measurement setup in air. b PTE measurement setup on pig eye. cWireless powering demonstrated on pig eye with red LED illumination. dMeasured PTE results in air and on pig eye. In both cases, measurements

are performed with and without compensation for distance variation. eSchematic of a rectifier and an LED integrated into the parylene contactlens platform to demonstrate energy harvesting scenario

Fig. 7 Simulation of local SAR field in HFSS human bodymodel. A 2-W power source is used to excite the coil which is 12mm away from the eye. Thepeak local SAR field is 0.021 W/kg. a Front view. b Side view

75 Page 8 of 9 Biomed Microdevices (2015) 17:75

Acknowledgments The authors are thankful to Jongbum Kim for tak-ing SEM pictures.

References

R.J. Cameron, C.M. Kudsia, R.R. Mansour, Microwave filters for com-munication systems. (Wiley-Interscience, 2007), pp. 279–293

A.P. Chandrakasan, N. Verma, D.C. Daly, Annu. Rev. Biomed. Eng. 10,247–274 (2008)

G. Chen, H. Ghaed, R. Haque, M.Wieckowski, Y. Kim, G. Kim, D. Fick,D. Kim, M. Seok, K.Wise, D. Blaauw, D. Sylvester, IEEE Int. SolidState Circuits Conf. (ISSCC), 310–312 (2011)

N.M. Farandos, A.K. Yetisen, M.J. Monteiro, C.R. Lowe, S.H. Yun, Adv.Healthcare Mater. 4, 6 (2015)

D. Ha, W.N. de Vries, S.W. John, P.P. Irazoqui, W.J. Chappell, Biomed.Microdevices 14, 1 (2012)

D. Ha, T.-C. Lee, D.J. Webery, W.J. Chappell, IEEEMTT-S Int. Microw.Symp. Dig. (IMS), 1–4, (2014)

A. Kurs, A. Karalis, R.Moffatt, J.D. Joannopoulos, P. Fisher, M. Soljačić,Science 317, 83 (2007)

M. Leonardi, E.M. Pitchon, A. Bertsch, P. Renaud, A. Mermoud, ActaOphthalmol. 87, 4 (2009)

W. Li, D.C. Rodger, E. Meng, J.D. Weiland, M.S. Humayun, Y.-C. Tai,IEEE Int. Conf. Microtechnol. Med. Biol. (MMB), 105–108 (2006)

A.R. Lingley, M. Ali, Y. Liao, R. Mirjalili, M. Klonner, M. Sopanen, S.Suihkonen, T. Shen, B.P. Otis, H. Lipsanen, B.A. Parviz, J.Micromech. Microeng. 21, 12 (2011)

J. Maeng, B. Kim, D. Ha,W.J. Chappell, IEEE Trans.Microwave TheoryTech. 59, 12 (2011)

J. Maeng, C. Meng, P.P. Irazoqui, Biomed. Microdevices 17, 7 (2015)K. Mansouri, F.A. Medeiros, A. Tafreshi, R.N. Weinreb, Arch.

Ophthalmol. 130, 12 (2012)N.A. McNamara, K.A. Polse, R.J. Brand, A.D. Graham, J.S. Chan, C.D.

McKenney, Am J. Ophthalmol. 127, 6 (1999)E. Meng, R. Sheybani, Lab Chip 14, 17 (2014)J. Pandey, Y. Liao, A. Lingley, R. Mirjalili, B. Parviz, B. Otis, IEEE

Trans. Biomed. Circuits Syst. 4, 6 (2010)G.A. Salvatore, N. Münzenrieder, T. Kinkeldei, L. Petti, C. Zysset, I.

Strebel, L. Büthe, G. Tröster, Nat. Commun. 5, 2982 (2014)P. Si, A. Hu, J. Hsu, M. Chiang, Y. Wang, S. Malpas, D. Budgett, IEEE

Ind. Electr. Appl., 235–239 (2007)M.A. Spivack, Rev. Sci. Instrum. 41, 11 (1970)X. Sun, X. Peng, Y. Zheng, X. Li, H. Zhang, J. Microelectromech. Syst.

23, 4 (2014)N. Thomas, I. Lähdesmäki, B.A. Parviz, Sensors Actuators B 162, 1

(2012)Ç. Varel, Y.C. Shih, B.P. Otis, T.S. Shen, K.F. Böhringer, J. Micromech.

Microeng. 24, 4 (2014)B.Winn, D.Whitaker, D.B. Elliott, N.J. Phillips, Invest. Ophthalmol. Vis.

Sci. 35, 3 (1994)R. Wu, S. Raju, M. Chan, J.K. Sin, C.P. Yue, IEEE Electron Device Lett.

34, 1 (2013)H. Yao, A.J. Shum, M. Cowan, I. Lähdesmäki, B.A. Parviz, Biosens.

Bioelectron. 26, 7 (2011)H. Yao, C. Marcheselli, A. Afanasiev, I. Lähdesmäki, B.A. Parviz, IEEE

Int. Conf. Microelectron. Mech. Syst. (MEMS), 769–772 (2012)Y. Zhao, M. Nandra, C.-C. Yu, Y.-C. Tai, Int. Conf. Eng. Med. Biol. Soc.

(EMBS), 6583–6586 (2012)

Biomed Microdevices (2015) 17:75 Page 9 of 9 75