MEMS based fiber optical microendoscopes

Displays xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Displays journal homepage: www.elsevier.com/locate/displa

Review

MEMS based fiber optical microendoscopes Zhen Qiu a, Wibool Piyawattanametha b,c,⇑ a

School of Medicine, University of Michigan, Ann Arbor, MI 48109, USA Faculty of Engineering, King Mongkut’s Institute of Technology Ladkrabang, Bangkok 10520, Thailand c Advanced Imaging Research (AIR) Center, Faculty of Medicine, Chulalongkorn University, Bangkok 10330, Thailand b

a r t i c l e

i n f o

Article history: Received 31 October 2014 Received in revised form 5 December 2014 Accepted 6 December 2014 Available online xxxx Keywords: MEMS scanner Confocal Two-photon OCT Fluorescence imaging

a b s t r a c t Fiber-optical microendoscopy has recently been an essential medical diagnostic tool for patients in investigating tissues in vivo due to affordable cost, high quality imaging performance, compact size, high-speed imaging, and flexible movement. Microelectromechanical systems (MEMS) scanner technology has been playing a key role in shaping the miniaturization and enabling high-speed imaging of fiber-optical microendoscopy for over 20 years. In this article, both review of MEMS based fiber-optical microendoscopy for optical coherence tomography, confocal, and two-photon imaging will be discussed. These advanced optical endoscopic imaging modalities provide cellular and molecular features with deep tissue penetration enabling guided resections and early cancer assessment. Ó 2015 Elsevier B.V. All rights reserved.

Contents 1. 2.

3. 4.

5.

6.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Optical microscopy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Optical coherence tomography. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Confocal microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Two-photon microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MEMS based OCT microendoscope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MEMS based confocal microendoscope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Single-axis confocal microendoscope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.1. Integrated single-axis confocal microendoscope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.2. MEMS based single-axis confocal microendoscope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Dual-axis confocal microendoscope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . MEMS based two-photon (2P) microendoscope. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1. MEMS scanner based 2P microendoscope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. Scanning fiber based 2P microendoscope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Abbreviations: US, ultrasound; CT, computed tomography; MRI, magnetic resonance imaging; NIR, near infrared; NA, numerical aperture; WD, working distance; UV, ultraviolet; FOV, field of view; SAC, single-axis confocal; DAC, dualaxis confocal; MEMS, microelectromechanical systems; CM, confocal microscopy; OCT, optical coherence tomography; 2P, two-photon. ⇑ Corresponding author at: Faculty of Engineering, King Mongkut’s Institute of Technology Ladkrabang, Bangkok 10520, Thailand. Tel.: +66 87 936 5000. E-mail address: [email protected] (W. Piyawattanametha).

00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

1. Introduction One of the major public health problems in human history is the Cancer disease killing over 7 million people each year [1]. Currently, the clinical diagnosis of most cancers and their precursors is based on gross structural features obtained from biopsy proce-

http://dx.doi.org/10.1016/j.displa.2014.12.001 0141-9382/Ó 2015 Elsevier B.V. All rights reserved.

Please cite this article in press as: Z. Qiu, W. Piyawattanametha, MEMS based fiber optical microendoscopes, Displays (2015), http://dx.doi.org/10.1016/ j.displa.2014.12.001

2

Z. Qiu, W. Piyawattanametha / Displays xxx (2015) xxx–xxx

dures, such as extent of local invasion, the presence of enlarged regional lymph nodes, and detection of lesions (>1 cm) in organs [2]. While this biopsy process has been the standard of care, it has many disadvantages, such as long diagnosis time, invasiveness, artifacts, sampling error, time consumption, relatively high cost, and interpretive variability. Therefore, the understanding of the underlying fundamental system biology is less explored due to the technical limits. It is not until 1895 that the scientific and medical communities are forever in debt to an accidental discovery of X-rays made by a German physicist Wilhelm Conrad Röntgen. Afterward, many noninvasive imaging modalities based on variety physical properties such as ultrasound (US), computed tomography (CT), and magnetic resonance imager (MRI) have been invented to perform disease analysis, diagnosis, prognosis, staging, treatment, and follow-up [3]. Even though, the above modalities are useful for delineating the deep extent of advanced carcinomas, they are insensitive to detect small, earlier intraepithelial lesions, which are more readily cured [4]. In contrast to the above modalities, optical imaging of tissue can be carried out noninvasively in real-time and in vivo, yielding high spatial resolution (submicron to micron scale). Moreover, the optical imaging modalities are inexpensive, robust, and portable because of advances in computing, optical fiber, semiconductor, and microelectromechanical systems (MEMS) technologies [5–7]. Current imaging modalities used in fiber optical microendoscopy include: optical coherence tomography (OCT) [8–10], confocal microscopy (CM) [11–13], and twophoton (2P) microscopy [14–16]. Those imaging modalities are often combined with the development of contrast agents targeting cancerous receptors enhancing accurate cancer detection [17,18]. Typically, MEMS technology has been integrated at the distal end as a scanning element with either raster scanning, random access, or lissajous scanning mode inside these microendoscopy to achieve two-dimensional (2D) en face scan imaging. Some of the advantages are fast speed (kHz-MHz), small in size (a few millimeters scale), low to medium power consumption, ease of integration, batch fabrication, and low cost. Sections below will briefly introduce each optical imaging modality, and its fiber optical endoscopic version based on MEMS scanner technology. 2. Optical microscopy 2.1. Optical coherence tomography

2.2. Confocal microscopy CM concept was first introduced by Marvin Minsky in the 1957 when he was a postdoctoral fellow at Harvard University [11]. The modality uses linear light-tissue interactions to generate high image contrast with micron-scale resolution [11–13]. The principal advantage of CM is its ability to record section information of three-dimensional (3D) tissue data with cellular definition by rejecting out of focus light from its unique optical sectioning property via a pinhole (Fig. 2a). The achievable field of view (FOV) of CM is typically limited (<100 lm2) and it requires the use of exogenous fluorophores to enhance image contrast. In standard operation, CM can be employed in two imaging modes namely reflectance and fluorescence. The former relies on the backscattered light from within the tissue and provide structural and anatomical information of cells and tissues whereas the latter records light generated by fluorescence contrast agents that target specific microstructures and has high sensibility and specificity. Fluorescence signal from linear light-tissue interaction is produced when a single excitation photon in the ultraviolet (UV) or visible regime is absorbed by electrons in tissue biomolecules that then transition into higherenergy (excited-state) levels. The electrons emit visible fluorescence photons when it spontaneously relaxes to the ground state, as shown in Fig. 2b. In this process, the fluorescence intensity (F)

Ó ELSEVIER 2000

OCT was first demonstrated in early 1990s [8]. Since then, numerous applications of OCT for both biomedical and materials applications have emerged. At the same time, the resolution and capabilities of OCT technology have improved dramatically. OCT is the optical analog of US which can perform micron-scale resolu-

tion, cross-sectional tomographic imaging of the internal microstructure in materials and biological systems. OCT performs imaging by measuring the magnitude and echo time delay of backscattered light. In the standard embodiment of OCT imaging, an incident light beam is directed at the object to be imaged, and the time delay and magnitude of backscattered or backreflected light is measured in the axial or longitudinal direction. The beam is scanned in the transverse direction, and rapid successive axial measurements are performed, as shown in Fig. 1a. The result is a 2D data set, which represents the optical reflection or backscattering in a cross-sectional plane through the material or tissue. Frequency domain OCT employs a Michelson interferometer. The input light from a broadband frequency sweeping light source is divided into the reference arm and sample arm. The light beams on both arms are reflected back and form an interference signal at one port of the beam splitter. This interference signal is read by a photodetector and provides the depth information of the sample through inverse Fourier transform. By using the short coherence length of a broadband light source, the resolution of OCT can reach 1–15 lm depending on the light source employed, shown in Fig. 1b. The penetration depth of OCT is normally 1– 3 mm, which is sufficient to image the depth of the epithelial layer, where most cancers are originated.

Fig. 1. (a) OCT generates cross-sectional images by performing measurements of the echo time delay of light at different transverse positions. The result is a 2D data set representing the backscattering in a cross-sectional plane of the tissue. Used with permission. (b) OCT working principle based on Michelson-type interferometer. The configuration measures the echo time delay of reflected light by using low-coherence interferometry. Reflections or backscattering from the object being imaged are correlated with light travelling through a reference path.

Please cite this article in press as: Z. Qiu, W. Piyawattanametha, MEMS based fiber optical microendoscopes, Displays (2015), http://dx.doi.org/10.1016/ j.displa.2014.12.001

3

Ó NPG 2005

Z. Qiu, W. Piyawattanametha / Displays xxx (2015) xxx–xxx

Ó NPG 2005

Fig. 2. (a) Single-axis confocal microscopy (CM) working principle. (b) An excitation photon (UV or visible wavelenght) excites electrons into a higher energy state that then spontaneously relax back to the ground state to generate a visible fluorescence photon whose intensity (F) is proportional to the excitation intensity (I). Light scattered by media outside of the focal plane is spatially filtered by a pinhole (e.g. an optical fiber core). (c) Excitation light trajectory.

Fig. 3. (a) Two-photon microscopy working principle. (b) Two NIR photons (e.g. 960 nm) arrive simultaneously to excite electrons into a higher energy state and spontaneously relax back to the ground state to generate a visible fluorescence photon whose intensity (F) is proportional to the square of the excitation intensity (I2). The non-linear effect is spatially confined to the focal plane to produce optical sectioning intensity (I). Light scattered by media outside of the focal plane is spatially filtered by a pinhole (e.g. optical fiber core) to perform optical sectioning. Used with permission. (c) Excitation light trajectory of 2P.

depends linearly on the incident excitation intensity (I). Tabletop confocal microscopes are commonly used in research laboratories and to perform sub-cellular resolution (<5 lm) imaging in cells and biology tissue specimens ex vivo. Optical sectioning is achieved with use of a pinhole placed in between the objective lens and the detector that allows only the light that originates from within a tiny volume in the focal plane below the tissue surface to be collected. Scattered light from all other origins outside of this focus does not have the proper trajectory to enter the pinhole and thus become spatially filtered, as shown in Fig. 2a. Fig. 2c shows excitation light trajectory in fluorescein. CM can be sub-divided into two main types namely (1) single-axis confocal (SAC) and 2) dual-axis confocal (DAC). In the SAC configuration, a high numerical aperture (NA) objective (typically > 0.5) is needed to achieve subcellular resolution, and this geometry achieves limiting working distance (WD). The scanning mechanism (or scanner) is usually placed on the pinhole side of the objective or in the pre-objective position. On the contrary, the DAC design uses two low NA objectives enabling off-axis illumination and collection. The latter architecture provides a long WD allowing a scanning mechanism to be placed post-objectively. For optical fiber based confocal microendoscopy, the single-mode optical fiber core is used as a pinhole. 2.3. Two-photon microscopy 2P microscopy is a fluorescence imaging technique that allows imaging of living tissue up to around one millimeter. The modality employs two-photon absorption that is nonlinear light-tissue interactions to generate image contrast [14]. It is a concept first described by Maria Goeppert-Mayer in 1931 [15]. Fluorescence is generated when two lower-energy (longer wavelength) photons in the near-infrared (NIR) regime arrive at tissue biomolecules simultaneously, as shown in Fig. 3a [16]. Typically, each photon has half of the energy needed to generate 2P effect. Instead of using a conventional UV or visible light source to excite the fluorophores,

2P excitation utilizes NIR laser to allow deeper tissue penetration. Due to the low probability of simultaneous or near simultaneous absorption of 2P by a single fluorophore, use of femtosecond laser, ultrashort 70–200 femtosecond (fs) with high peak intensity pulses of light, is required. The result obtained from this technique is a restricted focal volume only in the focal plane, Fig. 3c. Therefore, the probability of absorbing two photons increases with the square of the photon flux, and the fluorescence intensity (F) is proportional to the square of the excitation intensity (I2). This modality improves both axial depth discrimination and image contrast. Furthermore, photobleaching and photodamage to the samples are reduced. Sophisticated 2P imaging laboratory instruments have been developed to collect fluorescence imaging with sub-cellular resolution and tissue penetration. This technology is being adapted for in vivo imaging as an microendoscope by using the core of a photonic crystal fiber (PCF) or hollow air core fiber [19] to deliver the high peak intensity, ultra-short pulse to the tissue with minimal waveform distortions. In addition, other imaging methods are being developed with other nonlinear effects such as multiphoton (>2) absorption, second harmonic generation, and coherent anti-Stokes Raman scattering (CARS) [19]. For example, the latest three-photon table-top microscopy were demonstrated in vivo imaging on a mouse brain with up to 1.2 mm penetration depth [20].

3. MEMS based OCT microendoscope The first MEMS based OCT microendoscope was introduced by Pan et al. employing a one-dimensional (1D) electrothermally actuated MEMS mirror in 2001 [21]. In 2006, Piyawattanametha et al. from M.C. Wu’s group at the University of California, Los Angeles (UCLA) developed a novel angular vertical combdrive (AVC) actuated 2D MEMS scanner that was fabricated using surface and bulk micromachining processes [22], as shown in Fig. 4. Comb-

Please cite this article in press as: Z. Qiu, W. Piyawattanametha, MEMS based fiber optical microendoscopes, Displays (2015), http://dx.doi.org/10.1016/ j.displa.2014.12.001

Z. Qiu, W. Piyawattanametha / Displays xxx (2015) xxx–xxx

Ó IEEE 2005

4

Ó OSA 2007

Fig. 4. Electrostatically actuated 2D MEMS AVC gimbal scanner for MEMS based OCT microendoscope, (a) schematic (b) SEM of a scanner.

Fig. 5. MEMS based OCT system for in vivo imaging, (a) OCT system schematic for imaging with the MEMS endoscope. PC, polarization control. AGC, air–gap coupling. Det, detector. Amp, amplifier. D/A, digital to analog converter. (b) Schematic drawing of the OCT microendoscope head. OCT image acquired with the MEMS scanning catheter. (c) In vivo cross-sectional images of human skin. sc, stratum corneum; e, epidermis; d, dermis. Scale bar, 500 lm.

drive actuator is a type of electrostatic actuations that is commonly used in MEMS scanners with torsional rotation due to its relatively high actuation force. When voltage is applied between the movable and the fixed combdrive electrodes, the moving electrode rotates about their torsion axis until the restoring torque and the electrostatic torque are equal resulting in the rotation of MEMS scanners [22]. Soon after the success of a batch fabrication, those electrostatic MEMS scanners were integrated and packaged into a novel OCT endomicroscopic catheter in J.G. Fujimoto’s laboratory at the Massachusetts Institute of Technology (MIT) for in vivo ultrahigh resolution 3D and en face imaging. To the best of our knowledge, it is the first electrostatic MEMS scanner based microendoscope and used for in vivo small animal imaging [23], as shown in Fig. 5. Different from the AVC actuated scanner, another type of gimbal-less monolithic silicon micromirror with dedicate amplification mechanism was developed by Milanovic et al. [24] for tiptilt-piston applications. An additional mirror plate is bonded onto the top of actuation structure, which provides a very high fill-in factor with more than 90%. The early stage design can perform tilt and tip scan with frequency less than 3 kHz. The piston motion (zaxis out-of-plane displacement) is relatively small, less than few microns. The similar gimbal-less micromirror was successfully

used for a 3D endoscopic OCT system [25]. Based on electrostatic actuation, a cantilever-based scanner was also developed and integrated into an OCT catheter system [26]. Most recently, gimbal-less electrostatic scanners have been significantly improved and commercialized by Mirrorcle Technologies, Inc and they have been recently integrated into a handheld ultrahigh speed swept source OCT system [27], as shown in Fig. 6. High-resolution large FOV in vivo images on an eye have been demonstrated with the imaging system, shown in Fig. 7. Instead of steering light beams by micromirror, moving a focus lens by electrostatic MEMS actuators is an alternative approach for 3D imaging. Park et al. [28] have developed a forward imaging OCT endoscopic catheter using MEMS lens actuators. A group led by H. Toshiyoshi from the University of Tokyo has demonstrated a new all-optical MEMS fiber based OCT microendoscopes [29], shown in Fig. 8. Recently, those MEMS scanners have been commercialized by Santec Inc. for high speed OCT products. In addition to the MEMS scanner design for OCT medical microendoscopes, researchers also made efforts improving the alignment, wiring and packaging strategies. For example, the Silicon-on-Bench (SiOB) can potentially increase the accuracy of the micro-optics alignment. Wiring and electricity delivery are realized by soldering ball that is self-assembled inside the silicon trench. Another wire-

Please cite this article in press as: Z. Qiu, W. Piyawattanametha, MEMS based fiber optical microendoscopes, Displays (2015), http://dx.doi.org/10.1016/ j.displa.2014.12.001

5

Ó OSA 2014

Z. Qiu, W. Piyawattanametha / Displays xxx (2015) xxx–xxx

Ó OSA 2014

Fig. 6. Handheld ultrahigh speed swept source optical coherence tomography instrument using a MEMS scanning mirror, (a) optical design, (b) photo of the handheld scan head, and (c) schematic drawing showing inside optical path.

Fig. 7. 6  6 mm2, 350  350 A-scan volume (motion-corrected) of the optic nerve head. (a) En face fundus image. (b and c) Horizontal and vertical cross sectional images indicated by the colored lines. Scale bars are 1 mm.

bonding free technique using through-hole via silicon on insulator (SOI) wafer was proposed by Samuelson et al. [30] for much more compact packaging. In the most recent progress, an endoscopic swept-source optical coherence tomography based on a 2D electrothermal MEMS scanner with large z-axis stroke and tilting angle has been developed for in vivo imaging on mouse [31]. Based on the similar electrothermal driving principle, a miniature microendoscope catheter with a novel electrothermal MEMS scanner based on SiOB technology was developed [32]. Electromagnetic MEMS 2D scanners have been developed for OCT microendoscopes [33,34]. However, the outer electrical coils usually occupy a decent amount of space and prevent further miniaturization of microendoscopes. The group led by H. Zappe from

the University of Freiburg has made effort in the tunable microoptics for many years. A recent pneumatically actuated tunable lens has been integrated with magnetically actuated scanner into an OCT microendoscope system with SiOB platform technology [35]. Based on thin-film lead-zirconate-titanate (PZT) base MEMS scanner, researchers have developed a proof-of-concept endoscopic OCT system [36,37].

4. MEMS based confocal microendoscope The development of MEMS based confocal microendoscopes has made rapid progress over the past decades since the seminal work demonstrated by Dickensheets and Kino [38] from Stanford Uni-

Please cite this article in press as: Z. Qiu, W. Piyawattanametha, MEMS based fiber optical microendoscopes, Displays (2015), http://dx.doi.org/10.1016/ j.displa.2014.12.001

Z. Qiu, W. Piyawattanametha / Displays xxx (2015) xxx–xxx

Ó IEICE 2010

6

Ó OSA 1996

Fig. 8. All optical MEMS fiber based OCT microendoscope. (a) Schematic of the microendoscope system. (b) Photo of the packaged probe. (c) SEM of a MEMS scanner.

Ó OSA 1994

Fig. 9. Micromachined confocal optical scanning microscope (lCOSM) (a) Side view showing the zig–zag beam path. (b) Exploded isometric view showing the lens plate, spacer element, and scan mirror element that constitute the lCOSM, which consists of a single-mode optical fiber for illumination and detection, two tortional mirrors for scanning, and a binary transmission grating as the objective lens.

Fig. 10. Integrated single axis confocal microendoscope. (a) XY-scanning of a single-mode optical fiber is performed in the distal end with a cantilever mechanism. (b) Compact design allows for miniature microscope to be integrated in a standard medical endoscope.

versity in 1996. The microendoscope employs two cascaded 1D MEMS scanners as a scanning element to perform lissajous scanning during image acquisition, as shown in Fig. 9. Based on optical architecture, we can divide CM into two main configurations, which are single-axis and dual-axis confocal microendoscopy. The single-axis architecture is where the optical fiber and objective are co-located along the same optical axis. A high NA objective is usually used to achieve sub-cellular resolution and efficient light collection, and the same lens provides both illumination and collection of light [39–41]. Another alternative lens

such as gradient index (GRIN) lenses have been used to achieve millimeter dimensions. These rod-shaped objectives have a refractive index, n, that decreases continuously as a functional of the radial coordinate and can be reduced in size but increased in length with flat surfaces that allow for easy assembly. The WD is determined by the geometry of the objective and is typically on the order of a few hundred microns. Because of this space limitation, the scanning mechanism is typically located on the fiber side of the objective (pre-objective position). In this arrangement, some of the illumination light (or noise) scattered by tissue present in

Please cite this article in press as: Z. Qiu, W. Piyawattanametha, MEMS based fiber optical microendoscopes, Displays (2015), http://dx.doi.org/10.1016/ j.displa.2014.12.001

7

Ó IEEE 2003

Z. Qiu, W. Piyawattanametha / Displays xxx (2015) xxx–xxx

Ó SPIE 2012

Fig. 11. Electromagnetic scanner based confocal microendoscope by Olympus, (a) cross sectional drawing of the endoscope head; (b) electromagnetic MEMS scanner design.

Fig. 12. OD 5.5 mm diameter dual-axis confocal microendoscope scanhead. (a) Two collimated beams are focused by a parabolic mirror. Real-time en face scanning is performed by a 2D MEMS scanner. (b) Photograph of the microendoscope without its cap showing a 2D MEMS scanner mounted on the axial translation stage; scale bar is 3 mm.

between the objective and focal volume can be collected reducing image contrast. The dynamic range of this configuration is, therefore, limited, and images are usually collected in horizontal or en face cross-sections (plane parallel to the tissue surface). Hence, the dual-axis architecture [42–44] was developed to overcome some of these limitations and uses crossed off-axis illumination and collection of light along separate optical axes to overcome out of focus light tissue scattering. Such novel approach provides a significantly higher dynamic range and allows for the collection of vertical cross-sectional images (plane perpendicular to the tissue surface). This orientation is used by pathologists to evaluate for abnormalities in the differentiation pattern of the epithelium. 4.1. Single-axis confocal microendoscope 4.1.1. Integrated single-axis confocal microendoscope The first confocal microendoscope demonstrated for clinical use employed the SAC architecture and was developed by Optiscan Pty. Ltd. [41]. This instrument has been integrated into the insertion tube of a medical endoscope (Pentax Precision Instruments EC3870K, Tokyo, Japan). Scanning of the distal tip of a single-mode optical fiber is performed using a cantilever mechanism, as shown in Fig. 10a. A GRIN objective with an NA  0.6 focuses the beam into the tissue. Axial scanning is performed with a nitinol Z-actuator that translates the focal volume over a distance of 0–250 lm sub-surface. The diameter of the confocal microendoscope unit is around 5 mm. The overall size of the medical endoscope is 12.8 mm in diameter, as shown in Fig. 10b, which is slightly larger

than that of conventional instruments. This instrument can achieve transverse and axial resolutions of 0.7 and 7 lm, respectively. The images are collected in horizontal cross-sections at a frame rate of either 0.8 or 1.6 Hz and digitized to dimensions of 1024  1024 pixels, resulting in an FOV of 500  500 lm2. The excitation wavelength is at 488 nm. The fluorescence emission is at over 505 nm and is collected through the core of a single-mode fiber.

4.1.2. MEMS based single-axis confocal microendoscope Electromagnetic MEMS microscanner based SAC microendoscope was developed by Olympus Inc. for a proof-of-concept, shown in Fig. 11 [45]. Later on, based on the MEMS electrostatic staggered vertical comb (SVC) scanner was integrated into microendoscopes with a custom made objective lens co-developed by O. Solgaard’s group at Stanford University and R.R. RichardsKortum’s group at the Rice University developed [46,47]. Based on similar scanning mirror design as the Stanford University team, X. Zhang’s group from the University of Texas at Austin developed a handheld miniature fiber-optics based fluorescence hyperspectral confocal imaging system using their custom-made SVC MEMS scanners [48]. The system was used for oral cancer detection with a large FOV by hand-made mosaicing. Most recently, electrothermal or electromagnetic scanners based confocal microendoscope have also been developed for 3D imaging [49]. However, those proofof-concept designs have not been used for in vivo imaging due the relatively large enclosure.

Please cite this article in press as: Z. Qiu, W. Piyawattanametha, MEMS based fiber optical microendoscopes, Displays (2015), http://dx.doi.org/10.1016/ j.displa.2014.12.001

Z. Qiu, W. Piyawattanametha / Displays xxx (2015) xxx–xxx

Ó SPIE 2012

8

Fig. 13. OD 5.5 mm DAC microendoscope, (a) Demonstration in an Olympus XT-160 medical endoscope. (b) Distal end of the microendoscope; scale bar = 5 mm. (c) resolution test, (d) mosaiced large FOV en face images of normal colonic mucosa at a depth of 60 lm and (e) a representative histologic image stained with hematoxylin and eosin (H&E) of normal colonic mucosa. Scale bar = 100 lm. The white rectangle represents an individual en face image (362  134 lm2) obtained using the DAC microendoscope.

4.2. Dual-axis confocal microendoscope The DAC architecture is a novel confocal design that separates the illumination and collection beams along different optical axes, using the region of overlap between the two beams (focal volume) to define the resolution [50–53], as shown in Fig. 12a. Low NA objectives are used to enable a long WD so that the scanner can be placed on the tissue side of the lens or in the post-objective position. In this configuration, the two beams are always incident to the objective on axis, thus the scan mirror can sweep a diffraction-limited focal volume over an arbitrarily large FOV, limited only by the maximum deflection angle of the mirror. This feature allows for the size of the instrument to be scaled down to millimeter dimensions without losing performance. In addition, very little of the light that is scattered by tissue along the illumination path (noise) is collected in this geometry, resulting in a much higher dynamic range. Consequently, optical sections can be collected in the vertical cross-section to achieve deep tissue imaging with NIR light. This view shows the relationship among tissue microstructures as they vary with depth and provides important diagnostic information about variations in the differentiation pattern of the tissue. Handheld and miniaturized dual-axis imaging tools have been used for both in vivo and ex vivo on transgenic mouse models (cancer model or brain imaging with green-fluorescent protein/red-fluorescent protein (GFP/RFP) endogenous marker) and human skin imaging [54–59]. In the outside diameter (OD) 5.5 mm miniature microendoscope instrument, a MEMS scanner performs beam scanning while maintaining the fixed region of intersection between the two focal beams below the tissue surface. This device has an active mirror surface (600  650 lm2) for each beam that is connected together by a strut and is designed to deflect both beams equally to preserve the overlapping focal volume without introducing aberrations. This barbell-shaped scanner rotates on a gimbal around an inner and outer spring in two dimensions, as shown in Fig. 12b. Orthogonal rows of SVCs actuate the micromirror using electrostatic forces. The maximum optical scan angle is +/ 3.3° on the inner axis and +/ 1.0 ° on the outer axis, and the resonance frequencies of the device are 3.54 kHz an 1.1 kHz, respectively. Images can be acquired up to 30 Hz with a maximum FOV of 800  400 lm2 using post-objective scanning. The MEMS based dual-axis confocal microendoscope has been applied for in vivo high-resolution imaging for the human low gastrointestinal (GI) tract first the time in 2012 [60], in vivo images are shown in Fig. 13a–c. Large FOV

images (over mm scale) can be reconstructed by custom-made mosaicing algorithms [59]. Individual human colon crypts can be visualized in real time and in vivo, as shown in Fig. 13d and e, and potentially used for optical biopsy and colorectal cancer early detection in the near future. Taking advantages of the superior dynamic range and long WD characteristics of the dual-axis microendoscope [61,62], multispectral fluorescent deep imaging with vertical cross-sectional imaging can be achieved [63–65]. Latest version of dual-axis confocal microendoscope employed a 2D MEMS scanner mounted on a thin-film PZT actuation platform to achieve large out-of-plane (piston mode) vertical displacement (>500 lm) for dynamic focus tracking [66], as shown in Fig. 14.

5. MEMS based two-photon (2P) microendoscope Endoscope-compatible instruments are also being developed for collection of 2P excited fluorescence. While rapid progress has also been made recently in this area, this nonlinear effect is much weaker and presents a number of technical challenges for generating a sufficient throughput of fluorescence photons for a practical clinical system. The development of PCF [19] represents an important milestone for in vivo microendoscopy by effectively transmitting the excitation light. These hollow-core fibers preserve the width of the ultrashort pulse as it propagates over distances of several meters from the instrument unit to the endoscope without incurring distortions to provide the high peak intensities needed to generate 2P excited fluorescence. Similar to confocal microendoscopy, the scanning mechanism also plays a key role in the design of the miniature 2P microendoscopes. Both bulk PZT based scanning sheet/tubing [67] and MEMS scanners [68] have been developed to address this important feature. Submicron to micrometer lateral resolutions can be realized with 2P microendoscopy. However, ongoing stringent requirements in neuroscience are the main driving force behind the development of MEMS scanner based endoscopic imaging probes. New endoscopic probes enable new imaging capabilities to inspect the nervous system across multiple spatial scales at faster speeds and over larger tissue volumes than previously possible. Moreover, portable, miniaturized fluorescence microscopes with only 2.9 g allow brain imaging in freely behaving mice [69]. 3D acquisition is achieved with actuating MEMS based micromotor along optical axis during en face scans. Clinical imaging for cancer diagnosis in human for these MEMS scanner based endoscopic probes has yet to be demonstrated.

Please cite this article in press as: Z. Qiu, W. Piyawattanametha, MEMS based fiber optical microendoscopes, Displays (2015), http://dx.doi.org/10.1016/ j.displa.2014.12.001

9

Ó IEEE 2014

Z. Qiu, W. Piyawattanametha / Displays xxx (2015) xxx–xxx

Ó OSA 2009

Fig. 14. Vertical-rotational micro-scanner based cross-sectional imaging with dual-axis confocal microendoscope. (a) Schematic of the OD 5.5 mm packaging; (b) verticalrotational MEMS scanning stage based on active outer vertical displacement and passive inner resonant scanning. Inset: SEM, variant with a solid dog-bone shaped mirror surface for dual-axis confocal microendoscopy.

Ó OSA 2009

Fig. 15. MEMS scanner based 2P microendoscope. (a) CAD schematic shows packaging strategy and integration of MEMS scanner with micro-optics. (b) Optical configuration consists of separate illumination and fluorescence collection fibers coupled into a custom assembly of GRIN microlenses. (c) Photography of complete instrument in a package with dimensions of 2.0  1.9  1.1 cm3 and a mass of 2.9 g, scale bar 1 cm.

Fig. 16. A MEMS scanner. (a) SEM of a 2D MEMS scanner. (b) detail of torsional spring. (c) detail of outer combdrive banks. (d) frequency response and result in resonances of 1.08 kHz and 0.56 kHz for the fast (inner) and slow (outer) axes, respectively.

5.1. MEMS scanner based 2P microendoscope A miniature MEMS 2D scanner based 2P microendoscope has been developed by Piyawattanametha et al. for in vivo imaging the microvasculature of the brain of living mice. The exploded isometric computer-aided-design (CAD) schematic of the imaging instrument is shown in Fig. 15a. Tiny lens mounts are used to align the collimating optics with the optical fibers. A direct-current (DC) electromagnetic micromotor (Faulhaber Group, GmbH, Germany) that is integrated into the microendoscope drives the focusing shuttle to adjust the position of the PCF. Control wires for the

MEMS scanner and micromotor are wire-bonded onto electrodes located on the printed circuit board. This configuration uses separate optical fibers to deliver the ultrashort pulse and to collect fluorescence, as shown in Fig. 15b. The incident light (red arrow, in Fig. 15b) is comprised of 110-fs pulses delivered by a PCF that provides up to 1 nJ of energy. This beam is collimated by an aspheric lens and reflected by the MEMS micromirror into a micro-optical assembly (GRINTECH GmbH, Jena, Germany) that consists of four GRIN lenses, designated as scan, tube, collection, and objective. The scan and tube lenses work as a beam expander. This beam under-fills the back aperture of the objective lens, which has a

Please cite this article in press as: Z. Qiu, W. Piyawattanametha, MEMS based fiber optical microendoscopes, Displays (2015), http://dx.doi.org/10.1016/ j.displa.2014.12.001

Z. Qiu, W. Piyawattanametha / Displays xxx (2015) xxx–xxx

Ó OSA 2009

10

Fig. 17. 2P fluorescence images of neocortical microvasculature. (a–c) In vivo images of neocortical capillaries with eight frames averaging over 2 s at 4 Hz. Scale bar 50 lm. (d) Line images taken by driving only the scanner’s outer axis, at its resonant frequency (560 Hz). Flowing erythrocytes appear as dark streaks in relief.

5.2. Scanning fiber based 2P microendoscope Cantilever [75,76] or PZT tubing [77–81] based scanning fiber can also be used for ultra-thin 2P microendoscope because of its

Ó PNAS 2012

working distance of 280 lm. A dichroic microprism with a shortpass cut-off at 700 nm than deflects the NIR beam to the tissue. Fluorescence (green arrows, in Fig. 15b) is then collected by the full aperture of the objective lens and delivered through a multimode fiber to the photomultiplier tube (PMT). The image resolution (FWHM) was characterized with 100-nm diameter fluorescence beads and found to be 1.29 ± 0.05 lm and 10.3 ± 0.3 lm in the transverse and axial dimensions, respectively. The housing, shuttle, and base plate are made from a special conductive plastic (polyetheretherketone) to minimize any static charge that may damage the MEMS microscanner. A photo of the assembled instrument contained within a package with dimensions of 2.0  1.9  1.1 cm3 and having a mass of 2.9 g is shown in Fig. 15c, scale bar 1 cm. The size and weight of this MEMS based microendoscope is compatible with imaging the brain microvasculature in vivo in freely moving mice. An electrostatically actuated MEMS scanner [68] is designed around a 2D 750  750 lm2 micromirror that rotates on a gimbal in the xy-plane, as shown in the SEM in Fig. 16a, scale bar 250 lm. The mirror surface has a radius of curvature >1 m and average surface roughness of <16 nm. There are two inner and four outer combdrive banks of electrostatic, SVC actuators that steer the micromirror. The inner and outer combdrive banks have 103 and 53 combdrive fingers, respectively. These combdrive fingers interdigitate and provide electrostatic torque in one direction, while the torsional springs supply a restoring torque in the opposite direction. The inner (fast axis) and outer (slow axis) springs that sus-

pend the mirror and outside gimbal frame are 259  6 lm2 and 416  8 lm2, respectively. The whole MEMS chip die footprint is 3.2  3.0 mm2. A pair of control lines drives the opposing pairs of combdrive banks for the inner and outer axes of the scanner are about ±5° and ±4°, respectively. In dynamic operation, scan rates can be adjusted from near static to over the mechanical resonant frequencies of 1.08 kHz and 0.56 kHz for the inner (solid) and outer (dashed) axes, respectively, as shown by the frequency response in Fig. 16b. A raster-scan pattern, with one axis of the scanner oscillating at resonance and the other at the frame rate of 1–15 Hz, is generally used. Images with dimensions of 400  135 pixels2 are collected, resulting in a 295  100 lm2 FOV. The ability of this MEMS based microendoscope to perform functional imaging was demonstrated by attaching it to the cranium of live mice for collection of 2P excited fluorescence images of neocortical microvasculature in vivo. Visualization of capillaries is seen in Fig. 17a–c, scale bar 50 lm. The images shown are averaged over eight frames that were acquired in 2 s at a frame rate of 4 Hz. The mice were first anesthetized with an intraperitoneal injection of ketamine and xylazine. Furthermore, individual erythrocytes can be tracked as they flow through these vessels. Data were acquired by operating the MEMS micromirror in line-scan mode, where the outer axis is scanned at resonance while the inner axis is held stationary. The red blood cells that move in a parallel direction to the line scan appear as dark streaks. Moreover, the flow velocities can be measured from the slope (dy/dt) of these dark streaks near the center of the line scan, as shown in Fig. 17d. With the same SVC electrostatic scanner, Hoy et al. [70,71] demonstrated a 2P microendoscope system and also used it for micro-surgery with high intensity ultrafast laser power. Similar MEMS scanner based microendoscopes have also been developed for different applications. Tang et al. [72,73] demonstrated a handheld 2P imaging probe by using gimbal-less electrostatically actuated MEMS scanner. In addition, some resonant microscanners have been commercialized by OPUS microsystem company for laser scanning applications. Researchers can directly integrate those scanners into their 2P microendoscope prototypes to replace bulk galvo scanners [74].

Fig. 18. Scanning fiber based 2P microendoscope for fluorescence and second harmonic generation (SHG) imaging. (a) Schematic of the distal end of the scanning microendoscope (upper) and photograph of the assembled microendoscope (lower). PZT: tubular piezoelectric actuator; DCF: double clad fiber. (b) Microendoscopy imaging system layout. PMT: photomultiplier tube; PA: preamplifier; DAQ: data acquisition unit. Representative SHG images of a mouse cervical tissue section at day 18 of gestation acquired with a scanning microendoscope (c) and a bench-top microscope (d). The microscopy image was taken with a 20  0.95 NA objective with the image size chosen to match microendoscope FOV. The SHG microendoscopy image quality is approaching or comparable to the bench-top microscopy.

Please cite this article in press as: Z. Qiu, W. Piyawattanametha, MEMS based fiber optical microendoscopes, Displays (2015), http://dx.doi.org/10.1016/ j.displa.2014.12.001

Z. Qiu, W. Piyawattanametha / Displays xxx (2015) xxx–xxx

11

Table 1 Summary of the properties and characteristics of various optical microendoscopy. Imaging modality

Spatial Imaging resolution (lm) rate (Hz)

Advantages

Wide-field (non-sectioned) 100–300

>24

Single-axis confocal

0.5–5

>2

Dual-axis confocal

3–6

>15

 

>60

  

OCT

1–15

       



Two-photon

0.5–2

>5

     

Disadvantages

Full frame imaging High speed imaging Inexpensive light source No moving parts Commercial devices exist High sensitivity Provide functional information Miniaturization through proximal or distal ends Commercial devices exist Effective out-of-focus rejection of scattered light for high contrast Deep tissue imaging depth (400 lm) Relatively isotropic resolution Impressive miniaturization through a singlemode optical fiber High sensitivity, dynamic range, and imaging speed Penetration depth: 1–3 mm Commercial devices exist High resolution and contrast Deep tissue penetration (500 lm–1 mm) Less photobleaching and phototoxicity Commercial devices exist

 Poor resolution and contrast  No depth sectioning

 Limited contrast  Short tissue imaging depth (50–100 lm)  High NA optics lead to short working distances and increase image aberrations  Low NA optics limit sensitivity  Challenging alignment of a dual-beam configuration

 Non-fluorescence imaging  Detector array: expensive and small dynamic range  Depth resolution dependent on NA

 Very expensive laser source and optics  Need dispersion compensation or special optical fibers to maintain pulse shape

compact size and robustness. Wu et al. [79] from X. Li’s biomedical imaging group has developed a PZT scanning fiber based 2P microendoscope for in vivo imaging in mouse model, as shown in Fig. 18a. The scan head of the 2P microendoscope consists of a small tubular piezoelectric actuator that spirally scans a doubleclad fiber. This fiber is effective for delivery of femtosecond laser pulses for excitation and collection of fluorescence [80]. Excitation is delivered through the core (3.5 lm diameter and 0.19 NA), and fluorescence is collected by the core and by the inner cladding (103 lm diameter and 0.24 NA). The beam is focused by a GRIN lens with 0.22 pitch and a 1.8 mm diameter. The distal end of the instrument is housed in a thin-wall hypodermic tube with an overall diameter of 2.4 mm. The en face image is produced by a resonant scan of the double-clad fiber in a spiral pattern. The optical fiber extends out of the tubular piezoelectric actuator as a freestanding cantilever. The outer surface of the piezoelectric tube is divided symmetrically into four quadrants that form two orthogonal pairs of drive electrodes. Each pair of opposing quadrants is driven independently out of phase at the mechanical resonance of the fiber cantilever at 1690 Hz. At this frequency, images consisting of 512 circular scans are generated at 3.3 Hz. The diameter of the FOV is 160 lm using the GRIN lens with a magnification of 0.5. Fig. 18b shows imaging system layout. The potential of 2P microendoscopy to perform optical sectioning was demonstrated with rat oral mucosa stained with acridine orange. 3D volumetric image was generated by collecting horizontal (en face) cross-sections. Moreover, second harmonic generation (SHG) images can be acquired using the similar imaging system and used for in vivo study of cervical tissue with a comparable imaging quality to that by bench-top microscopy, as shown in Fig. 18c and d. Table 1 summarizes the properties and characteristics of various optical microendoscopy.

move morphologic assessment form post-procedure to intra-procedure, fundamentally creating a new paradigm shift of healthcare environment with patients. Optical microendoscopy discussed in this review, such as OCT, confocal, and multi-photon, has created a new paradigm shift real time morphological findings as it can perform in vivo optical sectioning for imaging deeply below the tissue surface with sub-cellular to cellular resolution. Therefore, they have become the mainstay in clinics as early cancer detection tools for visualizing the mucosa of hollow organs in vivo, such as oropharynx, esophagus, lung, stomach, colon, and rectum. However, no optical imaging modality, discussed so far, has demonstrated clear advantages over others. This is because specific medical diagnostic applications will dictate the requirements for spatial resolution, penetration depth, frame rate, FOV, WD, etc., from imaging systems. Endoscopic MEMS based fiber optical imaging modalities offer one of the best diagnostic approaches to noninvasively image in real time the cellular features of cancer in vivo. Currently, these MEMS based fiber optical imaging modalities are relatively inexpensive, robust, and portable because of advances in computing, fiber optics, and micro/nano technology. Among various MEMS actuation mechanisms, electrostatic actuation is the dominant choice in MEMS based fiber optical endoscopes due to lower power consumption, small footprint, and ease of integration. Future work may lead to the integration of all the aforementioned optical modalities into one endoscopic imaging probe coupled with the use of contrast agents to expand system performance into different organs or diseases. We hope that discussions such as this can provide a useful framework for progress along this exciting new frontier in MEMS based fiber optical microendoscopy.

6. Conclusion

This work is partially supported by grants from the FraunhoferBessel Research Award from the Alexander von Humboldt Foundation, Germany; the Newton Fund, British Council, UK; the King Mongkut’s Institute of Technology Ladkrabang, Thailand; the Thailand Research Fund, Thailand; and the National Research Council, Thailand.

The simplest view of in vivo pathology is that the doctor obtains diagnostic morphologic information in real time and relies only on the in vivo morphologic finds to make on-the-spot decisions about patient management and intervention. In vivo imaging would

Acknowledgements

Please cite this article in press as: Z. Qiu, W. Piyawattanametha, MEMS based fiber optical microendoscopes, Displays (2015), http://dx.doi.org/10.1016/ j.displa.2014.12.001

12

Z. Qiu, W. Piyawattanametha / Displays xxx (2015) xxx–xxx

References [1] American Cancer Society, Cancer Facts and Figures 2010, American Cancer Society, Atlanta, Ga, USA, 2010. [2] R.D. Strayer (Ed.), Rubin’s Pathology: Clinicopathologic Foundations of Medicine, 5th ed., Lippincott Williams & Wilkins, Philadelphia, 2007. [3] R. Weissleder, M.J. Pittet, Imaging in the era of molecular oncology, Nature 452 (2008) 580–589. [4] M.G. Pomper, Molecular Imaging: An Overview, Acad. Radiol. 8 (2001) 1141– 1153. [5] O. Solgaard, A.A. Godil, R.T. Howe, L.P. Lee, Y.A. Peter, H. Zappe, Optical MEMS: from micromirrors to complex systems, J. Microelectromech. Syst. 23 (3) (Jun. 2014). [6] S.T.S. Holmstrom, U. Baran, H. Urey, MEMS laser scanners: a review, J. Microelectromech. Syst. 23 (2) (2014) 259–275. [7] B.A. Flusberg, E.D. Cocker, W. Piyawattanametha, et al., Fiber-optic fluorescence imaging, Nat. Meth. 2 (12) (2005) 941–950. [8] D. Huang, E.A. Swanson, C.P. Lin, J.S. Schuman, W.G. Stinson, W. Chang, M.R. Hee, T. Flotte, K. Gregory, C.A. Puliafito, J.G. Fujimoto, Optical coherence tomography, Science 254 (5035) (Nov. 1991) 1178–1181. [9] J.G. Fujimoto, C. Pitris, S.A. Boppart, M.E. Brezinski, Optical coherence tomography: an emerging technology for biomedical imaging and optical biopsy, Neoplasia 2 (1) (2000) 9–25. [10] M.R. Hee, J.A. Izatt, E.A. Swanson, D. Huang, C.P. Lin, J.S. Schuman, C.A. Puliafito, J.G. Fujimoto, Optical coherence tomography of the human retina, Arch. Ophthalmol. 113 (1995) 325–332. [11] M. Minsky, Microscopy Apparatus, US Patent No. 3013467, 1961. [12] T.D. Wang, G. Triadafilopoulos, Endoscopy, in: J. Moore, G. Zouridakis (Eds.), Biomedical Technology and Devices Handbook, CRC Press LLC, FL, 2003. pp. 11:1-32. [13] J.B. Pawley (Ed.), Handbook of Biological Confocal Microscopy, third ed., Springer Verlag, New York, 2006. [14] W. Denk, J.H. Strickler, W.W. Webb, Two-photon laser scanning fluorescence microscopy, Science 248 (4951) (1990) 73–76. [15] M. Goeppert-Mayer, Über Elementarakte mit zwei Quantensprüngen, Annals Phys. 9 (3) (1931) 273–295. [16] Fritjof Helmchen, Winfried Denk, Deep tissue two-photon microscopy, Nat. Meth. 2 (12) (2005) 932–940. [17] S.F. Elahi, T.D. Wang, Future and advances in endoscopy, J. Biophoton. 4 (7–8) (2011) 471–481. [18] N. Khemthongcharoen, R. Jolivot, S. Rattanavarin, W. Piyawattanametha, Advances in imaging probes and optical microendoscopic imaging techniques for early in vivo cancer assessment, Adv. Drug Deliv. Rev. 74 (July 2014) 53–74. [19] F. Ling, M. Gu, Fibre-optic nonlinear optical microscopy and endoscopy, J. Microsc. 226 (3) (2007) 195–206. [20] N.G. Horton, K. Wang, D. Kobat, C.G. Clark, F.W. Wise, C.B. Schaffer, C. Xu, In vivo three-photon microscopy of subcortical structures within an intact mouse brain, Nat. Photon. 7 (3) (2013) 205–209. [21] Y. Pan, H. Xie, G.K. Fedder, Endoscopic optical coherence tomography based on a microelectromechanical mirror, Opt. Lett. 26 (24) (2001) 1966–1968. [22] W. Piyawattanametha, P.R. Patterson, D. Hah, H. Toshiyoshi, M.C. Wu, Surfaceand bulk-micromachined two-dimensional scanner driven by angular vertical comb actuators, J. Microelectromech. Syst. 14 (6) (2005) 1329–1338. [23] A.D. Aguirre, P.R. Hertz, Y. Chen, J.G. Fujimoto, W. Piyawattanameth, L. Fan, M.C. Wu, Two-axis MEMS scanning catheter for ultrahigh resolution threedimensional and en face imaging, Opt. Exp. 15 (5) (2007) 2445–2453. [24] V. Milanovic, G.A. Matus, D.T. McCormick, Gimbal-less monolithic silicon actuators for tip-tilt-piston micromirror applications, IEEE J. Sel. Top. Quant. Electron. 10 (3) (2004) 462–471. [25] J.M. Zara, S. Yazdanfar, K. Divakara Rao, J.A. Izatt, S.W. Smith, Electrostatic micromachine scanning mirror for optical coherence tomography, Opt. Lett. 28 (8) (2003) 628–630. [26] W. Jung, D.T. McCormick, J. Zhang, L. Wang, N.C. Tien, Z. Chen, Threedimensional endoscopic optical coherence tomography by use of a two-axis microelectromechanical scanning mirror, Appl. Phys. Lett. 88 (16) (2006) 163901–163903. [27] C.D. Lu, M.F. Kraus, B. Potsaid, J.J. Liu, W.J. Choi, V. Jayaraman, A.E. Cable, J. Hornegger, J.S. Duker, J.G. Fujimoto, Handheld ultrahigh speed swept source optical coherence tomography instrument using a MEMS scanning mirror, Biomed. Opt. Exp. 5 (1) (2014) 293–311. [28] H.C. Park, C. Song, M. Kang, Y. Jeong, K.H. Jeong, Forward imaging OCT endoscopic catheter based on MEMS lens scanning, Opt. Letters 37 (13) (2012) 2673–2675. [29] M. Nakada, C. Chong, A. Morosawa, K. Isamoto, T. Suzuki, H. Fujita, H. Toshiyoshi, Optical coherence tomography by all-optical MEMS fiber endoscope, IEICE Electron. Exp. 7 (6) (2010) 428–433. [30] S.R. Samuelson, L. Wu, J. Sun, S. Choe, B.S. Sorg, H. Xie, A 2.8-mm imaging probe based on a high-fill-factor mems mirror and wire-bonding-free packaging for endoscopic optical coherence tomography, J. Microelectromech. Syst. 21 (6) (2012) 1291–1302. [31] D. Wang, L. Fu, X. Wang, Z. Gong, S. Samuelson, C. Duan, H. Jia, J.S. Ma, H. Xie, Endoscopic swept-source optical coherence tomography based on a two-axis microelectromechanical system mirror, J. Biomed. Opt. 18 (8) (2013) 086005– 086009.

[32] Y. Xu, J. Singh, T. Selvaratnam, N. Chen, Two-axis gimbal-less electrothermal micromirror for large-angle circumferential scanning, IEEE J. Sel. Top. Quant. Electron. 15 (5) (2009) 1432–1438. [33] K. Aljasem, L. Froehly, A. Seifert, H. Zappe, Scanning and tunable micro-optics for endoscopic optical coherence tomography, J. Microelectromech. Syst. 20 (6) (2011) 1462–1472. [34] K.B. Kim, B.H. Park, G.N. Maguluri, T.W. Lee, F.J. Rogomentich, M.G. Bancu, B.E. Bouma, J.F. de Boer, J.J. Bernstein, Two-axis magnetically-driven MEMS scanning catheter for endoscopic high-speed optical coherence tomography, Opt. Exp. 15 (26) (2007) 18130–18140. [35] N. Weber, D. Hertkorn, H. Zappe, A. Seifert, Polymer/silicon hard magnetic micromirrors, J. Microelectromech. Syst. 21 (5) (2012) 1098–1106. [36] K.H. Gilchrist, R.P. McNabb, J.A. Izatt, S. Grego, Piezoelectric scanning mirrors for endoscopic optical coherence tomography, J. Micromech. Microeng. 19 (9) (2009) 095012–095022. [37] T. Naono, T. Fujii, M. Esashi, S. Tanaka, A large-scan-angle piezoelectric MEMS optical scanner actuated by a Nb-doped PZT thin film, J. Micromechan. Microeng. 24 (1) (2014) 015010. [38] D.L. Dickensheets, G.S. Kino, A micromachined scanning confocal optical microscope, Opt. Lett. 21 (10) (1996) 764–766. [39] R. Rouse, A. Kano, J.A. Udovich, S.M. Kroto, A.F. Gmitro, Design and demonstration of a miniature catheter for a confocal microendoscope, Appl. Opt. 43 (31) (2004) 5763–5771. [40] C. Liang, K. Sung, R. Richards-Kortum, M.R. Descour, Design of a highnumerical-aperture miniature microscope objective for an endoscopic fiber confocal reflectance microscope, Appl. Opt. 41 (22) (2002) 4603–4610. [41] P.M. Delaney, M.R. Harris, R.G. King, Fiber-optic laser scanning confocal microscope suitable for fluorescence imaging, Opt. Lett. 33 (4) (1994) 573– 577. [42] T.D. Wang, In-Vivo Microscopy, in: X. Intes, F. Azar, Translational Multi-Modality Optical Imaging, Artech House, Norwood, MA, 2008, pp 19–32. [43] T.D. Wang, M.J. Mandella, C.H. Contag, G.S. Kino, Dual axes confocal microscope for high resolution in vivo Imaging, Opt. Lett. 28 (6) (2003) 414– 416. [44] T.D. Wang, C.H. Contag, M.J. Mandella, N.Y. Chan, G.S. Kino, Dual axes confocal microscope with post-objective scanning and low coherence heterodyne detection, Opt. Lett. 28 (20) (2003) 1915–1917. [45] K. Murakami, A. Murata, T. Suga, H. Kitagawa, Y. Kamiya M. Kubo, K. Matsumoto, H. Miyajima, M. Katashiro, A miniature confocal optical microscope with MEMS gimbal scanner, in: IEEE International Solid-State Sensors and Actuators Conference, Boston, MA, USA, June, 2003, pp. 587– 590. [46] K.C. Maitland, H.J. Shin, H. Ra, D. Lee, O. Solgaard, R. Richards-Kortum, Single fiber confocal microscope with a two-axis gimbaled MEMS scanner for cellular imaging, Opt. Exp. 14 (19) (2006) 8604–8612. [47] H.J. Shin, M.C. Pierce, D. Lee, H. Ra, O. Solgaard, R. Richards-Kortum, Fiber-optic confocal microscope using a MEMS scanner and miniature objective lens, Opt. Exp. 15 (15) (2007) 9113–9122. [48] Y. Wang, M. Raj, H.S. McGuff, G. Bhave, B. Yang, T. Shen, X.J. Zhang, Portable oral cancer detection using a miniature confocal imaging probe with a large field of view, J. Micromech. Microeng. 22 (6) (2012). 065001. [49] L. Liu, H. Xie, 3-D confocal laser scanning microscopy based on a full-MEMS scanning system, IEEE Photon. Technol. Lett. 25 (15) (2013) 1478–1480. [50] J.T.C. Liu, M.J. Mandella, J.M. Crawford, S. Friedland, R. Soetikno, C.H. Contag, G.S. Kino, T.D. Wang, A dual-axes confocal reflectance microscope for distinguishing colonic neoplasia, J. Biomed. Opt. 11 (5) (2006) 054019– 054028. [51] L.K. Wong, M.J. Mandella, G.S. Kino, T.D. Wang, Improved rejection of multiplyscattered photons in confocal microscopy using dual-axes architecture, Opt. Lett. 32 (12) (2007) 1674–1676. [52] J.T.C. Liu, M.J. Mandella, H. Ra, D. Lee, G.S. Kino, O. Solgaard, W. Piyawatthanametha, C.H. Contag, T.D. Wang, Miniature near-infrared dualaxes confocal microscope utilizing a two-dimensional microelectromechanical systems scanner, Opt. Lett. 32 (3) (2007) 256–258. [53] J.T.C. Liu, M.J. Mandella, J.M. Crawford, C.H. Contag, T.D. Wang, G.S. Kino, Efficient rejection of scattered light enables deep optical sectioning in turbid media with low-NA optics in a dual-axes confocal architecture, J. Biomed. Opt. 13 (3) (2008) 034020–034030. [54] H. Ra, W. Piyawattanametha, E. González-González, M.J. Mandella, G.S. Kino, O. Solgaard, D. Leake, R.L. Kaspar, A. Oro, C.H. Contag, In vivo imaging of human and mouse skin with a handheld dual-axis confocal fluorescence microscope, J. Invest. Dermatol. 131 (5) (2011) 1061–1066. [55] H. Ra, E. González-González, B.R. Smith, S.S. Gambhir, G.S. Kino, O. Solgaard, R.L. Kaspar, C.H. Contag, Assessing delivery and quantifying efficacy of siRNA therapeutics in the skin using a dual-axis confocal microscope, J. Biomed. Opt. 15 (3) (2010) 036027–036031. [56] J.T.C. Liu, M.J. Mandella, N.O. Loewke, H. Haeberle, H. Ra, W. Piyawattanametha, O. Solgaard, G.S. Kino, C.H. Contag, A dual-axis confocal microscope utilizing a gradient-index relay lens and unstable Lissajous scanning for image-guidance during surgery, J. Biomed. Opt. 15 (2) (2010) 026029–026034. [57] W. Piyawattanametha, H. Ra, M.J. Mandella, K. Loewke, T.D. Wang, G.S. Kino, O. Solgaard, C.H. Contag, 3-D Near infrared fluorescence imaging using a MEMSbased miniature dual-axes confocal microscope, IEEE J. Sel. Top. Quant. Electron. 15 (5) (2009) 1344–1350.

Please cite this article in press as: Z. Qiu, W. Piyawattanametha, MEMS based fiber optical microendoscopes, Displays (2015), http://dx.doi.org/10.1016/ j.displa.2014.12.001

Z. Qiu, W. Piyawattanametha / Displays xxx (2015) xxx–xxx [58] H. Ra, W. Piyawattanametha, M.J. Mandella, P. Hsiung, J. Hardy, T.D. Wang, C.H. Contag, G.S. Kino, O. Solgaard, Three-dimensional in vivo imaging by a handheld dual-axes confocal microscope, Opt. Exp. 16 (10) (2008) 7224–7232. [59] K.E. Loewke, D.B. Camarillo, W. Piyawattanametha, M.J. Mandella, C.H. Contag, S. Thrun, J.K. Salisbury, In vivo micro-image mosaicing, IEEE Trans. Biomed. Eng. 58 (1) (2011) 159–171. [60] W. Piyawattanametha, H. Ra, Z. Qiu, S. Friedland, J.T.C. Liu, K. Loewke, G.S. Kino, O. Solgaard, T.D. Wang, M.J. Mandella, C.H. Contag, In vivo near-infrared dualaxis confocal microendoscopy in the human lower gastrointestinal tract, J. Biomed. Opt. 17 (2) (2012) 0211021–0211024. [61] W. Piyawattanametha, T.D. Wang, MEMS-based dual-axes confocal microendoscopy, IEEE J. Sel. Top. Quant. Electron. 16 (4) (2010) 804–814. [62] W. Piyawattanametha, T.D. Wang, Miniature dual axes confocal microscope for real time in vivo imaging, Intech, 2009. [63] Z. Qiu, Z. Liu, X. Duan, S. Khondee, B. Joshi, M.J. Mandella, K. Oldham, K. Kurabayashi, T.D. Wang, Targeted vertical cross-sectional imaging with handheld near-infrared dual axes confocal fluorescence endomicroscope, Biomed. Opt. Exp. 4 (2) (2013) 322–330. [64] Z. Qiu, S. Khondee, X. Duan, H. Li, M.J. Mandella, B.P. Joshi, Q. Zhou, S.R. Owens, K. Kurabayashi, K.R. Oldham, T.D. Wang, Vertical cross-sectional imaging of colonic dysplasia in vivo with multi-spectral dual axes confocal endomicroscopy, Gastroenterology 146 (3) (2014) 615–617. [65] W. Shahid, Z. Qiu, X. Duan, H. Li, T.D. Wang, K.R. Oldham, Modeling and Simulation of a Parametrically Resonant Micromirror With Duty-Cycled Excitation, J. Microelectromech. Syst., doi: 10.1109/JMEMS.2014.2315518 (in press). [66] Z. Qiu, C.H. Rhee, J. Choi, T.D. Wang, K.R. Oldham, Large stroke vertical PZT microactuator with high-speed rotational scanning, J. Microelectromech. Syst. 23 (2) (2014) 256–258. [67] F. Helmchen, M.S. Fee, D.W. Tank, W. Denk, A miniature head-mounted twophoton microscope: high-resolution brain imaging in freely moving animals, Neuron 31 (6) (2001) 903–912. [68] W. Piyawattanametha, R.P.J. Barretto, T.H. Ko, B.A. Flusberg, E.D. Cocker, H. Ra, D. Lee, O. Solgaard, M.J. Schnitzer, Fast-scanning two-photon fluorescence imaging based on a microelectromechanical systems two-dimensional scanning mirror, Opt. Lett. 31 (13) (2006) 2018–2020. [69] W. Piyawattanametha, E.D. Cocker, L.D. Burns, R.P.J. Barretto, J.C. Jung, H. Ra, O. Solgaard, M.J. Schnitzer, In vivo brain imaging using a portable 2.9 g twophoton microscope based on a microelectromechanical systems scanning mirror, Opt. Lett. 34 (15) (2009) 2309–2311.

13

[70] C.L. Hoy, N.J. Durr, P. Chen, W. Piyawattanametha, H. Ra, O. Solgaard, A. BenYakar, Miniaturized probe for femtosecond laser microsurgery and twophoton imaging, Opt. Exp. 16 (13) (2008) 9996–10005. [71] C.L. Hoy, O. Ferhanog˘lu, M. Yildirim, W. Piyawattanametha, H. Ra, O. Solgaard, A. Ben-Yakar, Optical design and imaging performance testing of a 9.6-mm diameter femtosecond laser microsurgery probe, Opt. Exp. 19 (11) (2011) 10536–10552. [72] W. Jung, S. Tang, D.T. McCormic, T. Xie, Y.C. Ahn, J. Su, I.V. Tomov, T.B. Krasieva, B.J. Tromberg, Z. Chen, Miniaturized probe based on a microelectromechanical system mirror for multiphoton microscopy, Opt. Lett. 33 (12) (2008) 1324– 1326. [73] S. Tang, W. Jung, D. McCormick, T. Xie, J. Su, Y.C. Ahn, B.J. Tromberg, Z. Chen, Design and implementation of fiber-based multiphoton endoscopy with microelectromechanical systems scanning, J. Biomed. Opt. 14 (3) (2009) 034005–034011. [74] T.M. Liu, M.C. Chan, I. Chen, S.H. Chia, C.K. Sun, Miniaturized multiphoton microscope with a 24Hz frame-rate, Opt. Exp. 16 (14) (2008) 10501–10506. [75] H. Bao, M. Gu, Reduction of self-phase modulation in double-clad photonic crystal fiber for nonlinear optical endoscopy, Opt. Lett. 34 (2) (2009) 148–150. [76] H. Bao, J. Allen, R. Pattie, R. Vance, M. Gu, Fast handheld two-photon fluorescence microendoscope with a 475 micron x 475 micron field of view for in vivo imaging, Opt. Lett. 33 (12) (2008) 1333–1335. [77] R. Le Harzic, M. Weinigel, I. Riemann, B. Messerschmidt, Nonlinear optical endoscope based on a compact two axes piezo scanner and a miniature objective lens, Opt. Exp. 16 (25) (2008) 20588–20596. [78] C.J. Engelbrecht, R.S. Johnston, E.J. Seibel, F. Helmchen, Ultra-compact fiberoptic two-photon microscope for functional fluorescence imaging in vivo, Opt. Exp. 16 (8) (2008) 556–564. [79] Y. Wu, Y. Leng, J. Xi, X. Li, Scanning all-fiber-optic endomicroscopy system for 3D nonlinear optical imaging of biological tissues, Opt. Exp. 17 (10) (2009) 7907–7915. [80] Y. Zhang, M.L. Akins, K. Murari, J. Xi, M.J. Li, K. Luby-Phelps, M. Mahendroo, X. Li, A compact fiber-optic SHG scanning endomicroscope and its application to visualize cervical remodeling during pregnancy, Proc. Nation. Acad. Sci. 109 (32) (2012) 12878–12883. [81] D.R. Rivera, C.M. Brown, D.G. Ouzounov, I. Pavlova, D. Kobat, W.W. Webb, C. Xu, Compact and flexible raster scanning multiphoton endoscope capable of imaging unstained tissue, Proc. Nation. Acad. Sci. 108 (43) (2011) 17598– 17603.

Please cite this article in press as: Z. Qiu, W. Piyawattanametha, MEMS based fiber optical microendoscopes, Displays (2015), http://dx.doi.org/10.1016/ j.displa.2014.12.001