Optical coherence tomography (OCT) is a medical optical imaging technique that can achieve high-resolution 3D images (1-15μm)1 by detecting infrared light scattered by body tissue. It has an imaging depth range of about 3mm in highly-scattered tissue and has already been widely used for external imaging, e.g., for eye diseases and skin cancer diagnosis. It would be useful to apply OCT to internal imaging, since about 85% of all cancers originate from the surface layers of internal organs. As these layers are within its imaging depth range, OCT can detect cancer at an early stage as well as precisely determine tumor margins. However, it is very challenging to apply this technique to internal organs because of stringent size and imaging time constraints.
One promising approach is to use microelectromechanical systems (MEMS) scanning micromirrors. Several research groups have demonstrated actuation mechanisms to generate rotational scanning for micromirrors for endoscopic OCT. Among them, electrostatic micromirrors can be very fast and consume low power, but the angular scan range is limited and the required driving voltage is too high (~100V) to be safe for patients. Electromagnetic micromirrors have also been used for OCT probes, but the assembly process is complicated and the potential electromagnetic interference caused by the use of permanent magnets is another concern.
Another problem these mirrors share is that they all have a small fill factor (the ratio of the mirror area to the device area). For a given resolution, the required mirror area is fixed. Thus, a small fill factor leads to a large device, which in turn increases the imaging probe size. Electrothermal mirrors, on the other hand, can achieve a large scan range at low drive voltage, but most of them have large initial tilt angles and significant shift of rotation centers, causing serious optical alignment and coupling problems.
We developed initial-flat and lateral-shift-free electrothermal micromirrors that typically scan large angles at less than 10V. They are also fast and typically have fill factors of 25%, compared to 10% or less for other MEMS mirrors. The device footprint is much smaller for the same effective mirror aperture size, so further miniaturization of the probe is possible. Figure 1 shows a MEMS scanning micromirror that we developed.2 The mirror plate is 1×1mm2 and the chip size is 2×2mm2, resulting in a 25% fill factor.
Figure 1. Scanning electron micrograph of an electrothermal MEMS mirror. Chip size: 2mm×2mm.
Another challenge is to make the size of the probe as small as possible while minimizing the manufacturing cost and complexity of assembly. We designed, manufactured and assembled two kinds of probes. The first-generation probes have an outer diameter of 5.8mm, and they have already been applied to some animal experiments.3 In the second-generation design, the electrical wiring of the MEMS mirror is much improved by employing a flexible printed circuit board, as shown in the 3D model of the probe in Figure 2(a). We reduced the diameter of the second-generation probes to 3.4mm. Figure 2(b) shows a fully assembled probe (diameter: 2.7mm) before inserting into a biocompatible plastic tube.
Figure 2. MEMS endoscopic probe, showing the MEMS mirror and a graded index (GRIN) lens. PCB: Printed circuit board.
Figure 3 shows 3D OCT images of a live mouse's ear and tongue that we obtained experimentally. As can be seen in the images, different layers can be delineated. The image volume is 2.3×2.3×1.6mm3 and the axial resolution is about 10μm. 3D in vivo images of a human fingertip have also been obtained, as shown in Figure 4. These probes can be inserted into the biopsy channels of conventional endoscopes.
Figure 3. 3D OCT images of the ear and tongue of a live mouse.
Figure 4. 3D in vivo image of a human fingertip.
The probe size will be further reduced by improving both MEMS design and probe packaging design. More in vivo experiments on both animals and humans have been planned and more results will come in the near future. Experimental results have shown that MEMS-based endoscopic OCT imaging has great potential in clinical applications.
This work is supported by the National Science Foundation under award #0725598.
Huikai Xie, Jingjing Sun
University of Florida
Huikai Xie is an associate professor of electrical and computer engineering at the University of Florida. He obtained his PhD from Carnegie Mellon University in 2002. His research interests include MEMS, microsensors, microactuators, biosensors, and biophotonics.
Jingjing Sun obtained her BS in optical engineering from Tianjin University, Tianjin, China. She is now pursuing her PhD at the Department of Electrical and Computer Engineering at the University of Florida. Her research interests are optical imaging, optical coherence tomography, endoscopic imaging probe design, and image processing.
Lei Wu is director of engineering, WiOptix Inc., Gainesville, Florida. He obtained his PhD in electrical and computer engineering from the University of Florida in 2009.
1. 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, no. (5035), pp. 1178-1181, 1991.
3. H. Ling, S. Guo, K. M. Thieman, B. T. Wise, A. Pozzi, H. Xie, M. Horodyski, The potential of optical coherence tomography in meniscal tear characterization, Proc. SPIE 7168, pp. 71682O, 2009. doi:10.1117/12.808577