Multi-angle illumination with pixel super-resolution enables lensfree on-chip tomography

A lensfree computational microscope uses multiple angles of partially-coherent illumination with source-shifting-based pixel super-resolution algorithms to image biological specimens on a chip.
29 November 2011
Serhan Isikman, Waheb Bishara, Uzair Sikora, Oguzhan Yaglidere and Aydogan Ozcan

Three-dimensional microscopy techniques have found increasing use in the life sciences because they provide high-resolution structural information of biological specimens.1 Numerous approaches to achieve 3D optical imaging have been demonstrated, for example, optical coherence tomography, confocal microscopy, diffraction tomography,2 optical projection tomography,3 and light-sheet microscopy.4 Many of these imaging platforms have complex and relatively bulky architectures, which restrict their use to advanced settings and impede their integration with lab-on-a-chip platforms. To address these issues, we recently introduced a compact and cost-effective tomographic microscopy platform for use in resource-limited settings and lab-on-a-chip devices.

Lensfree optical tomography (LOT) is an on-chip 3D microscopy technique for high-throughput imaging of biological specimens at a spatial resolution of <1μm×<1μm×<3μm in the x, y, and z dimensions, respectively.5 LOT relies on digital holographic microscopy to achieve sectional imaging of micro-sized objects. The technique uses simple light sources, such as LEDs, and an optoelectronic sensor array, such as a CCD or CMOS chip. The 3D structure of the target object is digitally reconstructed by measuring in-line transmission holograms of the sample as a function of the illumination angle. Additionally, by removing the lenses, we can reduce the size and complexity of the imaging system while increasing the field-of-view and depth-of-field that can be probed. Overall, LOT can perform sectional imaging of a large sample volume—approximately 15–100mm3—in a single data acquisition step.5

Figure 1. (a) Scheme of the lensfree optical tomography (LOT) setup, where partially coherent illumination is rotated along two orthogonal arcs to record multi-angle in-line holograms of samples placed on the sensor-chip. (b) Distinct details can be observed in the tomograms through the depth of a Caenorhabditis elegans worm, demonstrating the slicing performance of LOT.

Figure 2. A photograph (left) and schematic diagram (right) of the field-portable lensfree tomographic microscope. Individual LEDs butt-coupled to multimode optical fibers provide multiple angles of illumination. Optical fibers are electromagnetically actuated to implement source-shifting based pixel super-resolution at each illumination angle.

In LOT, the sample of interest is placed directly on a sensor array and is illuminated using partially-coherent light, such as an LED that is spatially filtered by a pinhole with a diameter of ∼0.05–0.1mm: see Figure 1(a). This partially-coherent source emanating through a large aperture simultaneously enables hologram recording of micro-objects on the sensor-chip while significantly reducing the coherent noise terms from speckle and multiple-reflections. To achieve tomographic imaging, lensfree holograms are recorded at different illumination angles within a range of ±50° with 2° increments. We limit the angular range to ±50° because the degraded pixel response at higher incident angles added significant artifacts to the holograms. To compensate for the resulting missing spatial frequency information, a dual-axis tomography scheme is used, where illumination is varied along two orthogonal arcs: see Figure 1(a).

At each angle of illumination, we acquire multiple sub-pixel shifted holograms by shifting the light source to different positions with ∼50–100μm increments, which do not need to be precisely known. Using pixel super-resolution (PSR) algorithms, we can digitally synthesize a single high-resolution hologram from multiple lower resolution holograms.6, 7 Next, we reconstruct these high-resolution holograms using iterative phase recovery techniques to obtain 2D projection images of the objects for each angle of illumination. Finally, we back-project these projection images to compute tomographic images of the specimens, such as cells or microorganisms like Caenorhabditis elegans: see Figure 1(b).5

Owing to its architectural simplicity, LOT also lends itself to a compact, light-weight, and cost-effective architecture (see Figure 2). To this end, we also demonstrated a field-portable tomographic microscope based on LOT.8 In this portable device, which weighs ∼110g, separate LEDs coupled to multimode optical fibers (∼0.1mm core diameter) are devoted to each illumination angle. The tips of these fibers are electromagnetically actuated to achieve source-shifting using small magnets and coils to record shifted holograms and implement PSR at each viewing angle. Angular increments of 4° along a single-arc were used in this device, as opposed to 2° in the dual-axis bench-top demonstration. We measured the axial resolution (along the z-axis) to be <7μm and obtained a lateral resolution of ∼1μm.8

In summary, we developed a device capable of achieving tomographic microscopy of micro-objects over large imaging volumes. LOT can provide a useful tool for lab-on-a-chip platforms and high-throughput imaging applications in low-resource settings.5–8 We hope to extend the angular range of illumination to, for example, ±80° using emerging sensor-array architectures. These may offer improved pixel-response at even higher incident angles and should allow us to achieve near-isotropic 3D resolution at the sub-micrometer scale.

Serhan Isikman, Waheb Bishara, Uzair Sikora, Oguzhan Yaglidere, Aydogan Ozcan
Department of Electrical Engineering University of California, Los Angeles (UCLA)
Los Angeles, CA

Aydogan Ozcan received his PhD from Stanford University's electrical engineering department. After a postdoctoral fellowship at Stanford, he was appointed as a faculty member at Harvard Medical School, Wellman Center for Photomedicine. He joined UCLA in 2007, where he is currently an associate professor.

1. V. Ntziachristos, Going deeper than microscopy: The optical imaging frontier in biology, Nat. Methods 7, pp. 603-614, 2010. doi:10.1038/nmeth.1483
2. Y. Sung, W. Choi, C. Fang-Yen, K. Badizadegan, R. R. Dasari, M. S. Feld, Optical diffraction tomography for high resolution live cell imaging, Opt. Express 17, pp. 266-277, 2009. doi:10.1364/OE.17.000266
3. J. Sharpe, U. Ahlgren, P. Perry, B. Hill, A. Ross, J. Hecksher-S⊘rensen, R. Baldock, D. Davidson, Optical projection tomography as a tool for 3D microscopy and gene expression studies, Science 296, pp. 541-545, 2002. doi:10.1126/science.1068206
4. J. Huisken, J. Swoger, F. D. Bene, J. Wittbrodt, E. H. K. Stelzer, Optical sectioning deep inside live embryos by selective plane illumination microscopy, Science 305, pp. 1007-1009, 2004. doi:10.1126/science.1100035
5. S. O. Isikman, W. Bishara, S. Mavandadi, F. W. Yu, S. Feng, R. Lau, A. Ozcan, Lens-free optical tomographic microscope with a large imaging volume on a chip, Proc. Nat. Acad. Sci. USA 108, pp. 7296-7301, 2011. doi:10.1073/pnas.1015638108
6. W. Bishara, T.-W. Su, A. Coskun, A. Ozcan, Lensfree on-chip microscopy over a wide field-of-view using pixel super-resolution, Opt. Express 18, pp. 11181-11191, 2010. doi:10.1364/OE.18.011181
7. W. Bishara, U. Sikora, O. Mudanyali, T.-W. Su, O. Yaglidere, S. Luckhart, A. Ozcan, Holographic pixel super-resolution in portable lensless on-chip microscopy using a fiber-optic array, Lab Chip 11, pp. 1276-1279, 2011. doi:10.1039/C0LC00684J
8. S. O. Isikman, W. Bishara, U. Sikora, O. Yaglidere, J. Yeah, A. Ozcan, Field-portable lensfree tomographic microscope, Lab Chip 11, pp. 2222-2230, 2011. doi:10.1039/C1LC20127A
Recent News