Quantitative phase microscopy, also known as interferometric microscopy, is a label-free imaging approach to study morphological changes in living biological cells. When light passes through a cell, it modifies the amplitude and phase of the complex optical field. The changes in amplitude and phase carry sample specific information such as cell morphology and dry mass. While amplitude measurement may be straightforward, quantifying changes in optical phase requires more sophisticated techniques such as interferometry. In the following, we highlight two near commonpath quantitative phase microscopy systems developed at LBRC:
DPM is a novel quantitative phase imaging technique that combines the single-shot feature of off-axis interferometry with the common path geometry [1]. As shown in Fig. 1, the wavefront modified by the sample is imaged on a grating that generates its multiple copies. The setup isolates the zeroth and first diffraction orders to be used a sample and reference fields, respectively, similar to typical Mach-Zehnder interferometry. This is achieved by utilzing another 4f imaging system where a pinhole placed in the Fourier plane filters all the spatial frequencies in the zeroth order beam; the spatially filtered beam serves as the reference plane wave for interference with the sample (1st order beam) at the CCD plane. The off-axis interference between the sample and reference beams creates a sinusoidal fringe pattern, see Fig. 2(a) [1,2]. The frequency of the fringe pattern is defined by the grating period, wavelength of the illumination, and the magnification of the final 4f imaging system. Using the Hilbert Transform, one can isolate the complex field of the spatially modulated order in the Fourier plane, see Fig. 2(b).
The size of the cropping window is determined by the numerical aperture of the imaging objective. The near-commonpath deisgn helps cancel common-mode noise leading to phase measurements with good sensitivity. The inverse Fourier transform of the cropped information, Fig. 2(b), yields the complex amplitude of the sample beam from which the amplitude and phase of the field can be calculated. If sample has homogeneous refractive index, then the sample height can be determined from the measured phase and the knowledge refractive index contrast between the sample and the medium. A series of such measurements can help quantify membrane fluctuation dynamics.
Figure 3 shows the movie of nanometer-scale membrane height fluctuations of a red blood cell. We have also recently shown that the sensitivity of these measurements depends on shot noise of the system which in turn depends on the camera well depth. Using a camera with 1.6 million electrons well depth, we have demonstrated sub-milliradians phase measurement sensitivity. We note that temporal averaging can help further improve the noise characteristics and hence measurement sensitivity [2]. Recently, we have also developed a DPM system that does not a physical pinhole to generate a plane wave reference beam, but instead employs a digital micromirror device (DMD) in the Fourier plane to implement a programmable pinhole [3]. The use of an active DMD offers two key advantages: flexibility in varying the interference fringe period on the camera to satisfy the pixel sampling conditions, and a simplification of the pinhole alignment process.
As briefly described above, DPM provides cell morphological information with high sensitivity based on single-shot off-axis interferometry combined with a near-common-path geometry. The approach, however, requires that the reference beam be passed through an optical pinhole, physical or implemented using a programmable DMD, which creates an alignment and optimization constraint that would constitute a demanding problem for a non-specialist in microscopy. Likewise, it is not a trivial task to replace a system hardware element such as the objective lens with another, because the pinhole size must be optimized with respect to the system optics. At LBRC, we have developed a near-common-path quantitative phase microscope, which provides a straightforward approach for generating highly sensitive phase profiles of biological samples is developed [4].
Figure 4 shows the schematic of the system where a collimated beam from fiber-coupled laser diode source is incident upon the sample that occupies one-half or less of the objective lens FOV. Instead of using a spatial pinhole filter in the Fourier plane of the final 4f system, we use a pair dove prisms as shown in the figure. The two selected diffraction orders from the grating pass through the dove prisms (oriented in different directions) such that the two beams flip in opposite directions prior to their recombination. This permits interfering the empty portion of one beam with the sample region of the other beam. The necessity that the beams interfere after being flipped mandates that a spatially coherent source be used for the approach. The SrQPM design is simple and modular requiring no maintenance, rendering it accessible to the non-specialists in optics. At the same time, it offers the same advantages as that of DPM including high-speed quantitative phase measurements, leading to cellular morphology, cell drymass and so on.
Figure 5 depicts quantitative phase imaging of a live RKO human colon-cancer cell and drosophila embryo. For the RKO cell, structures such as the cytoplasmic-nuclear boundary and nucleolus are clearly visible. Similarly, the progress of gastrulation in the middle section as well as the dorsal appendages can be observed in the quantitative phase image of the drosophila embryo. The drosophila embryos were treated with bleach to remove chorion (the outer egg shell), and later fixed with 5% formaldehyde before being transferred to 50% glycerol solution on a glass slide. We note that microfluidic devices, for which narrow microchannels are surrounded by optically flat background regions, are ideal targets for the self-reference near-common-path system.