Dispersion phase microscopy

The refractive index of a material varies with wavelength - a property also known as dispersion. In some applications such as optical communications, dispersion is viewed as a deleterious effect that requires compensation. However, the same property may be used to provide a source of contrast for biological imaging or to quantify the concentration of specific biomolecules in a given sample. Traditionally, absorption measurements in mid UV have been used for visualization or concentration of biomolecules. However, using UV illumination poses significant challenges as it causes physical and chemical damage to the cells under observation, especially when long-term imaging is desired. This can be avoided by resorting to dispersion measurements in the near UV and visible spectrum and using Kramers-Kronig relationship to quantify absorption characteristics and hence concentration. Over the years, LBRC has made many contributions to this topic using point, line, and wide-field illuminations [1-5]. Some of these stduies are highlighted below:

Wide-field dispersion phase microscopy

Figure 1: Schematic of a wide-field dispersion phase microscope [1].

Figure 1 shows a wide-field dispersion phase microscope, which is essentially a near common-path interferometer [see DPM] together with wavelength selection scheme based on a broadband white light source coupled with set of color filters [1]. While the DPM provides highly stable phase measurements, the use of color flters enables wavelength switching. Seven different flters were used to select various center wavelengths: 440±20, 546±10, 560±20, 580±25, 600±20, 655±20, and 700±20 nm for dispersion characterization of biological material. For each wavelength illumination, the image of the specimen is projected onto the image plane IP2 where a holographic grating generates multiple diffraction orders. After the zeroth- and frst-order beams are isolated, the zeroth-order beam is spatially low-pass filtered through a pinhole placed at the Fourier plane of last 4f lens system. The spatially-filtered beam becomes the reference beam whereas the first-order beam serves as sample beam. Both beams interfered and generated a spatially modulated interference image, which was then captured by a CCD camera.

Figure 2: (a) Dispersion of water, PDMS, and BSA solution. (b) Dispersion of different concentrations of Hb solution. (c,d) Histograms of Hb conc. and volume of human RBCs [1].

The above system was used to quantify Hb concentration in intact red blood cells as follows. We used a PDMS microfluidic channel and filled it with distilled water. Quantitative phase images were acquired at different wavelengths mentioned above, which lead to the measurement of relative phase delay between PDMS and water. This information together with the knowledge of dispersion of water and the microfluidic channel height was used to determine the dispersion of PDMS. Next, water was replaced with bovine serum albumin (BSA) and another set of mesaurements were made, leading to the quantification of disperison of BSA solution [Fig 2(a)]. Using the same approach, the dispersion of three different concentrations of Hb solutions was also characterized [see Fig 2(b)]. The calibrated dispersion of Hb solution was further used to measure the cytoplasmic Hb concentration in live human RBCs. More specifically, fresh blood (5 ml) was diluted in phosphate-buffered saline (PBS) solution and then washed three times to remove white blood cells and platelets. The interference images of RBCs were acquired at the seven wavelengths. Using the Hb solution dispersion [Fig. 2(b)], the Hb concentration in each RBC was determined. The volume of individual RBCs was also determined from measured area and cell height. Figures 2(c) and 2(d) show the histograms of Hb concentration and cell volume [1].

Figure 3: Single-shot spectroscopic phase measurement of a 6 microns polystyrene bead. (a-c) Interferograms at single wavelengthsand (d-f) corresponding FFT images. (g) Single interferogram using all the colors simultaneously and (h) corresponding FFT image. (i-k) Quantitative phase images of the same bead at the three wavelengths, 780nm, 532nm, and 405nm, respectively, obtained from the single interferogram [3].

Follow up studies at the LBRC focused on a dual-wavelength quantitative phase microscope to measure quantitative phase images of cells at 310 nm and 400 nm [2]. The dry mass, projected area, and the ratio of integrated OPL at the two wavelengths were determined for HeLa cell population. The measured dispersion of living HeLa cells was found to be consistent with that measured directly for protein solutions using total internal reflection. Later on, the LBRC researchers further improved this approach by developing a dispersion phase microscope with single-shot dispersion measurement at three wavelengths simultaneously [3]. Figure 3 shows single-shot dispersion measurements at 780nm, 532nm, and 405nm by the system.

Self-reference line-dispersion phase microscopy

Figure 4: Experimental setup of line-illumination dispersion phase microscopy.

Simultaneous interferometric microscopy over a continuous spectrum provides the necessary platform to measure dispersive properties of the biological samples in motion such as when they are moving in a flow or dynamically moving within the field-of-view. To determine the disperive of biological samples over wide continuous spectrum, we have developed a line-scanning interferometric microscope [4]. Figure 4 shows the experimental setup where a broadband supercontinuum source is used as the light source. A cylindrical lens (L1) in combination with the condenser oil immersion lens (1.2 NA) forms a 4F imaging system to illuminate the sample with a line-focused beam along one axis at the sample plane. at the sample plane (SP) of an inverted microscope. The camera C1 is used at one of the output ports of beam splitter BS1 to acquire bright-field images by using a white light source (and without the cylindrical lens) to monitor the sample before data acquiisition. The light from the second output port of BS1 is sent into a Michelson interferometer. The two mirrors M3 and M4 of the inteferometer are tilted to obtain relative shift between the two beams such that the sample-free region serves as the reference as well as to achieve off-axis holography.

Figure 5: (a) Oblique illumination image of a HeLa cell. (b)-(d) show reconstructed wide-field quantitative phase images of the HeLa cell at 485 nm, 535 nm, and 620 nm, respectively.

The spectrograph is used to disperse the interferometric beam to be recorded using a color camera. The sample is translated across the line beam and serial RGB spectro-interferometric images are acquired. While in the wide-field spectroscopic measurements, the acquisition speed is mainly determined by the wavelength sweeping time, the speed in this system is defined either by speed of the translation stage or movement of the sample during flow. The spectrally encoded interference pattern can be recognized as modulated sinusoidal fringes carrying phase differences between the sample and the reference beams. By translating the sample across the line focused beam, we can reconstruct a wide-feld spatio-spectral quantitative phase map. Figure 5(a-d) shows the oblique illumination image of HeLa cell and the reconstructed wide-field quantitative phase image at three different wavelengths that are 485 nm, 535 nm, and 620 nm, respectively. The measured phase represents the cumulative effect of the optical properties and morphology of the sample.

References

  1. "Spectroscopic phase microscopy for quantifying hemoglobin concentrations in intact red blood cells," Optics Letters, 2009. [ Pubmed ]
  2. "Quantitative dispersion microscopy," Biomedical Optics Express, 2010. [ Pubmed ]
  3. "Single-shot quantitative dispersion phase microscopy," Applied Physics Letters, 2012. [ Pubmed ]
  4. "Self-reference line-dispersion interferometric microscopy," Applied Physics Letters, 2017. [ Pubmed ]