Cross Sectional Dual Axis Endomicroscopy

Add One

When biopsies are reviewed by pathologists in histology, they are typically viewed in ‘cross-section’ – a slice is cut perpendicular to the surface of the tissue. In contrast, microscopy performed directly on the patient (in vivo or endomicroscopy) usually generates en face images – slices in the plane of the tissue surface. Some endomicroscopes designs allow vertical cross-sections to be generated from stacks of en face images at varying depths, but this is nowhere near to being a real time approach. A group from the University of Michigan has now developed a handheld endomicroscope which can generate vertical cross-sections with a frame rate of 10 Hz.

The instrument works on the principle of dual-axes microscopy, a technique which was invented at Stanford University1. Dual-axes microscopy is a way of generating optically depth-sectioned images using low numerical aperture optics. Optical sectioning is necessary when imaging thick tissue to prevent the image being swamped by out-of-focus contribution from layers of tissue above and below the focal plane. In conventional confocal microscopy the excitation beam is focused to a particular depth within the sample (the focal plane) and scanned transversally. Returning fluorescence or reflected light is de-scanned and focused onto a pinhole. Sectioning occurs because light from the focal plane is focused on the pinhole and passes through, while light from out-of-focus planes is mostly rejected. The disadvantage of confocal microscopy for in vivo imaging is two-fold. Firstly, it’s necessity to scan the beam across the sample, limiting the frame rate. Secondly, the depth resolution is inversely proportional to the square of the numerical aperture of the objective lens. So a high numerical aperture lens is required, making the whole thing difficult to miniaturise.

The idea of dual-axes microscopy is to separate the excitation and collection optical paths. The excitation path is arranged so as to illuminate the sample at an angle, while the collection path is similarly arranged at a different angle, with the two paths coinciding at the focal plane. The result is an optical sectioning effect similar to confocal microscopy, but with lower numerical aperture lenses. This allows for a large working distance between the objective lens and the tissue, so the scanning mechanism (a pair of MEMS mirrors) can fit between them. This means that the beam always passes through the centre of the lens, reducing aberrations and allowing for a larger field of view.

The Stanford Group developed the dual axis microscope firstly as bench-top device, and then later as a hand-held instrument2. These previous implementations produced images similar to other endomicroscopes – scanning in both transversal directions to create an en face image parallel to the tissue surface.  New work led by Thomas Wang at Michigan University has led to a device that instead generates cross-sections3. Rather than a two axis MEMS mirror, it uses a single axis mirror for transversal scanning in one direction only. Depth scanning is then performed using an actuator to translate the mirror towards or away from the tissue.

The device has a field of view of 800 microns (transversal) by 400 microns (depth), although the practical depth range is limited by the signal to noise ratio. The images shown in the paper, from mouse colonic tissue, are superficially similar to optical coherence tomography cross sections. In practice they are somewhat different as they are formed from fluorescence rather than reflectance, and the practical imaging range is much shallower.

The prototype device is fairly large, due mainly to the z-axis actuator, and would be difficult to use in vivo. The authors are planning the next generation to have an outer diameter of 5 mm which would make it suitable for a much wider range of applications. Whether clinicians will prefer an en face or cross-sectional view in practice is another question. The ideal instrument would, of course, be capable of imaging in both modes. There is no obvious reason why this couldn’t be done in principle, although increasing complexity would tend to make miniaturisation more difficult. 


  1. T. Wang, M. Mandella, C. Contag, and G. Kino, “Dual-axis confocal microscope for high-resolution in vivo imaging,” Opt. Lett.  28, 414-416 (2003)
  2. J. Liu, M. Mandella, H. Ra, L. Wong, O. Solgaard, G. Kino, W. Piyawattanametha, C. Contag, and T. Wang, “Miniature near-infrared dual-axes confocal microscope utilizing a two-dimensional microelectromechanical systems scanner,” Opt. Lett.  32, 256-258 (2007)
  3. Z. Qiu, Z. Liu, X. Duan, S. Khondee, B. Joshi, M. Mandella, K. Oldham, K. Kurabayashi, and T. Wang, “Targeted vertical cross-sectional imaging with handheld near-infrared dual axes confocal fluorescence endomicroscope,” Biomed. Opt. Express  4, 322-330 (2013)

Leave a Comment

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>