A microscope-based label-free microfluidic cytometer capable of acquiring two dimensional light scatter patterns from single cells, pattern analysis of which determines cellular information such as cell size, orientation and inner nanostructure, was developed. Finite-difference time-domain numerical simulations compared favorably with experimental scatter patterns from micrometer-sized beads and cells. The device was capable of obtaining light scattering patterns from the smallest mature blood cells (platelets) and cord blood hematopoietic stem/progenitor cells (CD34 + cells) and myeloid precursor cells. The potential for evaluation of cells using this label-free microfluidic cytometric technique was discussed.
©2010 Optical Society of America
Flow cytometry has been widely used in biology and medicine [1–3]. Recent research demonstrated that the integration of optics with microfluidics [4,5] could potentially be applied to flow cytometry [6–8]. Microfluidic flow cytometers are the result of an amalgamation of microfluidics with flow cytometry and have several advantages over conventional flow cytometers. First, the microfluidic channel has a micrometer-sized cross-sectional area that dramatically reduces the total sample volume required for analysis. Second, the development of lab-on-a-chip (LOC)  techniques may present the portable cytometer as a next generation of commercially-available diagnostics tools. Large-scale production of portable cytometers could greatly reduce the cost of these devices.
Conventional flow cytometry measures the fluorescence signals from a single cell to obtain inner organelle information [1,3]. Labeling or staining of cells requires complex, time-consuming and expensive procedures and may also alter their functions. Compared with fluorescence method, the light scattering technique can be used as a label-free method for the study of cells [7,8,10–15]. Light scattering measurements have been performed in conventional cytometers for particle or cell size approximation, by integrating the small angle forward-scattered light intensity in a 5 degree cone angle and by measuring the side-scattered light intensity around the 90 degree polar angle, which is perpendicular to the propagation direction of the incident light [12,14,16]. The two-dimensional (2D) pattern of the scattered light, especially the side-scattered light (scattering around the 90 degree polar angle) from coherent light sources, provides significant information about cellular organelles . A combination of light scattering with microfluidic cytometry may find potential label-free applications in clinics.
Motivated by the need for low-cost, point-of-care portable cytometers, we have recently developed a 2D light scattering microfluidic cytometric technique [7,8]. This cytometer was employed to collect wide angle spatial spectra of the scattered light that were subsequently used to develop a fast Fourier transform (FFT)-based technique for the accurate measurements of cell size . The collected angular spectra also clearly displayed the 2D scattered light patterns from the mitochondria in a single cell . The main source for the background noise in this cytometer was the light scattered from the contaminants in the fluid and the microchannel side walls that could dominate the signals from the single scatterer that is of interest. To improve the signal to noise ratio (SNR) of the 2D cytometer, a substrate was first coated with a thin film and then etched to fabricate a small observation window (approximately 400 µm in diameter). Even with an improved SNR, it was still difficult to detect some important but weakly scattering cells such as platelets and embryonic-like stem cells , because their small sizes caused the scattered light to be indistinguishable from the background noise.
In a previous publication , we localized a single bead in a microfluidic channel and viewed it from above the chip using a standard microscope image system. We obtained a 2D scatter pattern on a charge coupled device (CCD) sensor beneath the microfluidic chip with no optical lens in between the microfluidic chip and the CCD sensor. We noticed the similarity between the ‘defocused scatter image’, defined as the bead being out of focus on the microscope, and the 2D scatter patterns obtained from the lens-free CCD. To study the ‘defocused scatter image’, we have combined and simplified the imaging system by using only a CCD sensor and a microscope objective . Defocused scatter pattern or diffraction images of plastic beads have also been recently reported in Ref. .
The method described here combines 2D light scattering from a microfluidic cytometer and optical microscope imaging into a technique that we call microscope-based label-free microfluidic cytometry (LFMC). By incorporating a microscope objective into a microfluidic cytometer, we have the capability to image a single cell in the microfluidic flow and then obtain its 2D scatter patterns by defocusing through this same microscope objective. The numerical aperture (NA) of the microscope objective helps to reduce the background stray light and improves the capability of the LFMC to detect very small cells such as platelets, as will be demonstrated in this work. Since the detection angle was reduced with the microscope objective, fabrication of a micrometer-sized observation window is no longer necessary. The microscope-based LFMC could allow low-cost fabrication techniques to be used to prepare the microfluidic chips, without the necessity of standard microfabrication tools and clean room facilities such as sputtering machines. In this paper, we verify the information obtained by the microscope-based LFMC by comparing the experimental scatter patterns of standard micrometer-sized polystyrene beads with finite-difference time-domain (FDTD) [7,20–24] simulations. Applying this LFMC for cell analysis, we find that the cell size, orientation and inner nanostructure of single cells may be determined without any labeling.
The key components of the LFMC are the fiber coupling of the light source into the side of the microfluidic channel to illuminate a single cell or bead, the microfluidic channel to propagate the cell (or bead) to the illumination point, and the microscope objective for imaging the cell (or bead) and obtaining its scatter pattern onto a CCD sensor array. A schematic of the microfluidic and optical components of the microfluidic cytometer is shown in Fig. 1 .
The designed microfluidic channel structure was first printed out onto a 120 µm thick polymer sheet (Canon transparency type E, Canon USA, Inc., NY, USA) using an office HP laser printer (HP LaserJet 5000 Series PS). The channel was then cut out from the polymer sheet along the designed structure using a surgical blade. At this stage, the gasket for the microfluidic channel structure was a whole piece (~18 mm by 63 mm). Two additional channels were cut out on either side of the microfluidic channel to accommodate the placement of optical fibers (105/125 µm multimode fiber, Thorlabs, NJ, USA) for guiding the laser to illuminate a cell (or a bead) and for guiding the laser out of the channel. The gasket was now two pieces, which were aligned and fixed onto the standard microscope slide (12-544-1, 25 mm by 75 mm, Fisher Scientific Company, ON, Canada) using distilled water. Another glass slide with drilled pump-in/pump-out holes was put on top. Note the gasket was smaller than the standard glass slide. Norland optical adhesive 81 (Norland Products Inc., NJ, USA) was then applied on the edges and cured by UV light to sandwich the gasket pieces between the two microscope slides. The microfluidic channel fabricated was about 600 µm in width, 120 µm in height, and 46 mm in length.
To achieve a cell concentration such that one cell was within the observation volume, a cell solution was prepared by diluting to approximately 2000 cell per milliliter from a known concentration. The flow in the channel was pressure-driven by using a syringe. As a cell arrived at the observation area, it was immobilized by manipulating the syringe to apply positive and negative pressures to the flow. Coherent light from a 532 nm laser diode (DPSS laser, Laserglow Technologies, ON, Canada) operating at 2 mW was coupled into one of the fibers to illuminate the cells or beads flowing through the channel. The scattered light was collected by a microscope objective with a NA of 0.25 onto a CCD detector (ICX204AK, Sony, Japan) with an integration time of 1/15 s.
The rationale for the microscope-based LFMC is to image a single cell that is illuminated perpendicularly to a viewing system and then to defocus the imaging system to obtain a 2D scatter pattern. In Figs. 2(a) and 2(b), we show the procedures for obtaining the 2D scatter patterns from a 4 µm polystyrene bead (Invitrogen, CA, USA). When a single bead was immobilized in the microfluidic channel, the microscope objective was translated approximately 300 µm from its best focal position to obtain a 2D scatter pattern. At its best focal position, the resolution of the microscope objective is approximately 1.3 µm at 532 nm illumination, according to the Rayleigh criterion. The 2D scatter patterns obtained by defocusing may contain information about the internal nanostructures in a single cell with sizes significantly smaller than 1.3 µm as will be discussed in section 3.
The numerical 2D scatter patterns were obtained by using AETHER [7,23], our FDTD numerical code based on Yee’s algorithm . The AETHER solved Maxwell’s equations and gave numerical solutions for light scattering from a single scatterer, including non-homogeneous, irregular-shaped cells . A scatterer was located in a 3D grid and a Liao boundary condition terminated the 3D grid . Because of the small size of cells the incident laser light was modeled as a plane wave, propagating along the + z direction and polarized along x direction (for example, as will be illustrated in Fig. 3 ). The AETHER was implemented in a FORTRAN 90 code and was run on the WestGrid clusters. A typical 3D simulation of a cell with a diameter of 10 μm required approximately 24 h using 64 GB of memory on WestGrid, with a spatial step size of 40 nm in AETHER.
We tested our device, and performed simulations and analysis using blood cells of various sizes, orientation and degree of maturation. Platelet concentrates were obtained from donors at the Canadian Blood Services (Edmonton, AB, Canada). Cord blood was obtained with mothers’ informed consent after delivery. Cord blood hematopoietic stem/progenitor cells (HSPC) which express the CD34 antigen were separated by immunomagnetic selection using MACS technology according to the manufacturer’s instructions (Miltenyi Biotec, Auburn, CA, USA). For some experiments we stained CD34 + cells with Mito Tracker Red (M-7512, Molecular Probes Inc., Eugene, OR, USA) and examined mitochondrial distribution in these cells by confocal fluorescence microscopy. Cord blood CD34 + HSPC were also ex vivo expanded and differentiated towards the myeloid lineage in a serum-free liquid culture (StemSpan, StemCell Technologies Inc., Vancouver, BC, Canada) in the presence of recombinant human (rh) interleukin-3 (10 ng/mL) and rh granulocyte macrophage-colony stimulating factor (5 ng/mL) (both from Peprotech, Rocky Hill, NJ, USA) as described in our earlier work . Cell cultures were incubated at 37°C in a humidified atmosphere supplemented with 5% CO2 for up to 11 days. Under these conditions, on day 11, almost 100% of cells expressed CD33, a marker of the myeloid lineage .
3. Results and discussion
The detection capability of the LFMC for scattered light angular range was determined by geometric analysis. Figure 3 is a diagram depicting the propagation of a scattered light ray in the LFMC through a layer of water (100 µm, refractive index 1.334), a glass substrate (1.0 mm, refractive index 1.5) and the air (refractive index 1.0), into the microscope lens, with the light ray subsequently detected by a CCD sensor. Note that a microscope lens with a NA of 0.25 will limit the angular range that can be detected by the CCD sensor. Based on geometric analysis, this 2D cytometer was capable of obtaining light scattered over a 22 degree cone angle (79 degree to 101 degree in polar angle, as compared with the detection of light scattering in a 60 degree cone angle centered around the 90 degree scattering by using the lens-free cytometer ).
First, to demonstrate the angular detection capability of the LFMC experimentally, we performed experiments using the standard 4 µm and 9.6 µm diameter polystyrene beads (Invitrogen, CA, USA). Figures 4(a) and 4(b) show the 2D scatter patterns from the 4 µm and 9.6 µm beads, respectively. Finite-difference time-domain light scattering simulations were performed on these beads. They have included far-field transformation of the scattered light and the angular range is determined by ray tracing simulations to describe light propagation in the optical elements of the cytometer, cf. Figure 3. The refractive index for the beads was 1.591 for a laser wavelength of 532 nm. The surrounding medium was assumed to have a refractive index of 1.334. Figures 4(c) and 4(d) show the AETHER generated 2D scatter patterns for the 4 µm and 9.6 µm beads collected in a 22 degree cone angle, respectively. From both experimental and simulation results, four fringes for the 4 µm bead, and nine fringes for the 9.6 µm bead were observed. Good agreement between experimental and simulation results confirmed that the LFMC obtained light scattering patterns in a cone angle of approximately 22 degrees, which is in agreement with the geometric analysis as described above.
This 2D LFMC can provide label-free characterization of single cells and their organelles including cell nucleus, mitochondria, and cytoplasm. These organelles are the main contributors to the total light scattering from cells. Other organelles such as ribosome or lysosome contribute significantly less to the total scattering due to their smaller volume fractions [7,14]. Among the blood cells, some cells have many organelles (e.g., white blood cells), some have no nucleus (e.g., red blood cells and platelets), and some have few mitochondria (e.g., platelets); hence our interest in studying these representative cell types using the 2D LFMC. In this work, we chose to study platelets, HSPC (CD34 + cells) from cord blood and ex vivo expanded and differentiated myeloid precursor cells.
Next, we verified that the LFMC could obtain 2D light scattering patterns from the smallest human blood cells. Platelets, the smallest mature cells circulating in blood, are disc-shaped with a diameter of about 3 µm and a thickness of about 1 µm. The experimentally-obtained platelet scatter patterns are shown in Figs. 5(a) , 5(b) and 5(c), and were compared with the AETHER light scattering simulations. For the simulation of platelets, the cell was assumed to be ellipsoidal [Fig. 5(d)], the projection on the xz-plane being a circular region with a 3 µm diameter, and the projection on the yz-plane being elliptical with a minor axis of 1 µm along y-direction and a major axis of 3 µm along z-direction. The direction of fluid flow and laser propagation were assumed to be along the x-axis and z-axis (Fig. 3), respectively. Three spherical mitochondria in the platelet model were assumed to be present, each with a diameter of 500 nm. The refractive index of the mitochondrion was 1.42 inside a cell with a refractive index of 1.38 . The surrounding medium had a refractive index of 1.334. Figure 5(e) shows a platelet with a different orientation in the microfluidic channel, rotated 90 degrees counterclockwise from the previous case [Fig. 5(d)]. Figure 5(f) shows a platelet with an orientation in between these two cases. In Figs. 5(d), 5(e) and 5(f), the different cell components are shown in different colors: the cell cytoplasm is magenta and the mitochondria are blue. The corresponding AETHER 2D scatter patterns for the platelet models (d), (e) and (f) are shown in Figs. 5(g), 5(h) and 5(i), respectively. The experimental results agree well with the AETHER simulations. Since the cell microstructures (nucleus and cytoplasm) give fringe structures in 2D scatter patterns  and because platelets do not contain nucleus, the platelet cytoplasm is the main contributor to the 2D scatter patterns of the platelets. Simulations for homogeneous platelet models with different orientations (cell models are the same as in Figs. 5(d), 5(e) and 5(f) but without mitochondria) give similar scatter patterns as in Figs. 5(g), 5(h) and 5(i), respectively. The fringe numbers increase when the effective size (size of the platelet projection along the z axis) of the platelet increases. Three fringes [Fig. 5(g)], two fringes [Fig. 5(i)], and no fringe [Fig. 5(h)] were predicted for different orientations. For the experimental scatter patterns, we observed platelets with three fringes (see Fig. 5(a)) and with no obvious fringe [see Fig. 5(b)]. According to our simulations, the 3 fringes correspond to an effective size of 3 µm, while the scatter pattern with no obvious fringe is for an effective size of 1 µm. These values correspond to the size parameters for platelets, which have a disc-shape structure with a diameter of 3 µm and a thickness of 1 µm. We also obtained platelet scatter patterns with two fringes [Fig. 5(c)], which may be explained with the cell model in Fig. 5(f) that gives a scatter pattern as in Fig. 5(i). The study of cell orientation is important for understanding light scattering from cells  and their active response to external stimuli . The light scattering method shown here can be used for the label-free determination of cell orientation in a microfluidic flow by observing the fringe numbers which are mainly determined by the effective size of a cell along the direction of the incident wave vector.
Previously, we reported that the highest dominant frequency component in the Fourier spectra of the light scattering signal can be used for determination of cell size . Alternatively, as shown in the present report, one can simply count the number of diffraction fringes to determine the size of beads (Fig. 4) or in the case of the platelets (Fig. 5) to gain information about the orientation of cells. The FFT based method  has also been successfully applied for the size determination of plastic beads at different flow velocities as shown in a recent publication by another group . Our results (Fig. 5) indicate that the orientation effects should be considered when using the fringes of the scatter patterns for size determination of a single scatterer.
Since the scatter light intensity increases with the size of a cell, the successful detection of platelets indicated that our device was capable of measuring the scatter patterns from various types of human blood cells. Specifically, we were interested in investigating immature HSPC (CD34 + cells) and more mature myeloid precursor cells.
The scatter patterns from myeloid precursor cells and CD34 + cells (Fig. 6 ) are quite different as compared with the platelet scatter patterns. In contrast to the fringe patterns observed from the platelets, the myeloid precursor and CD34 + cell scatter patterns are dominated by small scale 2D structures. The differences between the platelet scatter patterns and those of the myeloid precursor and the CD34 + cells could be explained, based on recently reported results, where the homogeneous microstructures in cells give rise to 2D fringe scatter patterns and the randomly distributed nanometer scale mitochondria generate the small scale 2D structures .
A simple cell model was used for the AETHER simulation of the light scattering from the myeloid precursor cells [Fig. 6(c)]. In this model, a myeloid precursor cell was assumed to be a 10 µm diameter sphere with a 6 µm diameter nucleus (in cyan) located at the center of the cell. There were also 120 mitochondria arbitrarily distributed inside the cell, each with a diameter of 1 µm. The refractive index for the myeloid precursor cell cytoplasm, nucleus, and mitochondria were assumed to be 1.35, 1.39, and 1.42, respectively . Figure 6(e) shows the simulated scatter pattern for the myeloid precursor cell model [Fig. 6(c)], which qualitatively agrees with the experimental results (quantitative analysis was performed as shown in Fig. 8 ). Both the experimental and simulation scatter patterns are dominated by the small scale 2D structures, indicating that the mitochondria are the main contributors to the 2D light scatter patterns of the myeloid precursor cells. The mitochondria dominate the light scattering from cells due to their higher refractive index (i.e., they are optically dense) and volume fraction as compared to other cell components .
A difference did exist between the scatter patterns of the myeloid precursor cell [Fig. 6(a)] and the CD34 + cell [Fig. 6(b)]. We propose that the differences between Fig. 6(a) and 6(b) are due to the distribution of the mitochondria. Mitochondria are the main contributors for the 2D structures in Figs. 6(a) and 6(b), but our models considered the randomly distributed mitochondria in myeloid precursor cell [Fig. 6(c)] and the aggregated mitochondria in CD34 + cell [Fig. 6(d)]. As shown in Fig. 6(d), there are 70 mitochondria with a diameter of 1 µm and a refractive index of 1.42, arbitrarily distributed in an ellipsoid centered at the origin with two long axes of 8 µm, and a short axis of 4 µm. The short axis is in the yz-plane and rotated 45 degrees from the +z-axis. Figure 6(f) shows the AETHER 2D scatter pattern for the CD34 + cell model [Fig. 6(d)], which compares favorably with the experimental CD34 + cell scatter pattern as shown in Fig. 6(b). To ensure that our model of the CD34 + cell is realistic, we imaged mitochondria in CD34 + cells using confocal fluorescence microscopy (Fig. 7 ). The images confirm mitochondrial aggregation in CD34 + cells, which agree qualitatively with our cell model.
We have presented in this report the scattered light diffraction fringes from beads or platelets and described the methods of determining the particle size or cell orientation. Note that the beads or platelets were treated as homogenous spheres or ellipsoids, giving the regular 2D fringe patterns. When the scattered light angular spectrum is dominated by irregularly distributed small-scale 2D structures as in Fig. 6, corresponding to the scatter pattern produced by mitochondria, the real space analysis of the intensity maxima may become very effective, as compared with the Fourier method we developed for sizing cells . We show below a method for cell discrimination by analyzing the 2D scatter patterns in spatial domain.
We analyzed the 2D scatter patterns in Fig. 6. All 2D scatter patterns were normalized to have a maximum intensity of 240 arbitrary units (a. u.). In Fig. 8, two parameters were used for the cell discrimination. One was the number of the distinctive scattered light regions in a 2D scatter pattern, and the other was the average area for these regions. We used ImageJ  to process the data. We set the “noise tolerance” value in ImageJ to 10, that is, a local intensity maximum was only considered if it stood out from the surrounding by more than 10 a. u. in the normalized scatter pattern. The distinctive scattered light region was then defined as the continuous area around this local intensity maximum with values differed from the local intensity maximum value by less than 10 a. u..
We performed this analysis for scatter patterns of 10 different CD34 + cells and 10 myeloid precursor cells, shown as open triangles and open squares in Fig. 8, respectively. The solid symbols indicate the average values for both groups of cells. We demonstrated that the CD34 + cells are distinctive from the myeloid precursor cells. The CD34 + cells have 6 ± 1.4 distinctive scattered light regions with an average area of 1.8 ± 0.5 squared degrees. For the myeloid precursor cells, there were 23 ± 8.0 distinctive scattered light regions with an average area of 0.8 ± 0.3 squared degrees. In CD34 + cells there are less small-scale 2D structures and their average areas are larger compared with the myeloid precursor cells. The AETHER results are shown in Fig. 8 as plus ( + ) and cross ( × ) signs for the myeloid precursor cell and the CD34 + cell, respectively. The simulation result for myeloid precursor cell pattern [Fig. 6(e)] has 23 distinct regions with an average area of 0.9 squared degrees and the CD34 + cell pattern [Fig. 6(f)] has 8 regions with an average area of 1.8 squared degrees. The AETHER simulation results agree well with the statistical results from the experimental patterns for both types of cells. By analyzing the 2D scatter patterns in real space we may be able to discriminate various stages of hematopoietic stem cell differentiation.
A microscope-based label-free microfluidic cytometric (LFMC) technique was developed to obtain 2D light scatter patterns from a wide range of blood cells. Comparisons between the AETHER generated and experimentally-measured scatter patterns from polystyrene beads showed that the microscope-based LFMC can accurately determine cell sizes. To improve our simulations and achieve closer agreement to our experimental results, an optical ray tracing subroutine was added to our AETHER code to model light propagation in the optical system of the cytometer .
Using this LFMC technique, we obtained 2D scatter patterns from platelets, myeloid precursor cells and CD34 + hematopoietic stem/progenitor cells. The 2D scatter patterns from the platelets (the smallest mature human blood cells) have been used for the determination of the platelet cell orientation in a microfluidic channel. The nanoscale organelle information (for example, mitochondria) in immature CD34 + cells and the more mature myeloid precursor cells was obtained by using the LFMC, which would otherwise not be achieved by using the 10 × objective lens according to the Rayleigh criterion. Analysis of the experimental CD34 + and myeloid precursor cell scatter patterns showed that the determination of hematopoietic stem cell differentiation may be achieved by using our LFMC technique.
The LFMC technique is simple and compact, requiring a CCD detector, a microfluidic chip, a diode laser source coupled into the channel by an optical fiber, and an objective lens. These components are standard equipment in most laboratories, making the technique widely accessible. This inexpensive LFMC could be applied in the study of the inner structures of live cells. Future development of the LFMC may lead to a new generation of cytometers that could have applications in medicine, including the study of stem cell differentiation or early detection of malignant cells.
The authors acknowledge the financial support of Canadian Institute for Photonic Innovations (CIPI), MicroSystems Technology Research Initiative (MSTRI) of University of Alberta, and Natural Sciences and Engineering Research Council of Canada (NSERC), and Canadian Blood Services/Canadian Institute of Health Research (CBS/CIHR). We thank WestGrid of University of Alberta for parallel computation support; the Integrated Nanosystems Research Facility (INRF) of University of Alberta for providing the confocal microscopy facility; Alois Harmony from Department of Medicine for useful discussion about confocal microscopy; and Jencet Montano from CBS for technical help.
References and links
1. B. K. Patterson, M. Till, P. Otto, C. Goolsby, M. R. Furtado, L. J. McBride, and S. M. Wolinsky, “Detection of HIV-1 DNA and messenger RNA in individual cells by PCR-driven in situ hybridization and flow cytometry,” Science 260(5110), 976–979 (1993). [CrossRef] [PubMed]
2. H. M. Shapiro, Practical Flow Cytometry, (John Wiley & Sons, Inc., Hoboken, NJ, 2003).
5. C. Monat, P. Domachuk, and B. J. Eggleton, “Integrated optofluidics: A new river of light,” Nat. Photonics 1(2), 106–114 (2007). [CrossRef]
6. M. M. Wang, E. Tu, D. E. Raymond, J. M. Yang, H. C. Zhang, N. Hagen, B. Dees, E. M. Mercer, A. H. Forster, I. Kariv, P. J. Marchand, and W. F. Butler, “Microfluidic sorting of mammalian cells by optical force switching,” Nat. Biotechnol. 23(1), 83–87 (2005). [CrossRef]
8. X. T. Su, K. Singh, C. Capjack, J. Petrácek, C. Backhouse, and W. Rozmus, “Measurements of light scattering in an integrated microfluidic waveguide cytometer,” J. Biomed. Opt. 13(2), 024024 (2008). [CrossRef] [PubMed]
10. R. Drezek, A. Dunn, and R. Richards-Kortum, “A pulsed finite-difference time-domain (FDTD) method for calculating light scattering from biological cells over broad wavelength ranges,” Opt. Express 6(7), 147–157 (2000). [CrossRef] [PubMed]
11. H. Subramanian, P. Pradhan, Y. Liu, I. R. Capoglu, X. Li, J. D. Rogers, A. Heifetz, D. Kunte, H. K. Roy, A. Taflove, and V. Backman, “Optical methodology for detecting histologically unapparent nanoscale consequences of genetic alterations in biological cells,” Proc. Natl. Acad. Sci. U.S.A. 105(51), 20118–20123 (2008). [CrossRef] [PubMed]
13. M. Bartholdi, G. C. Salzman, R. D. Hiebert, and M. Kerker, “Differential light scattering photometer for rapid analysis of single particles in flow,” Appl. Opt. 19(10), 1573–1581 (1980). [CrossRef] [PubMed]
14. J. D. Wilson, C. E. Bigelow, D. J. Calkins, and T. H. Foster, “Light scattering from intact cells reports oxidative-stress-induced mitochondrial swelling,” Biophys. J. 88(4), 2929–2938 (2005). [CrossRef] [PubMed]
15. E. Maurer-Spurej, K. Brown, A. Labrie, A. Marziali, and O. Glatter, “Portable dynamic light scattering instrument and method for the measurement of blood platelet suspensions,” Phys. Med. Biol. 51(15), 3747–3758 (2006). [CrossRef] [PubMed]
16. Z. Wang, J. El-Ali, M. Engelund, T. Gotsaed, I. R. Perch-Nielsen, K. B. Mogensen, D. Snakenborg, J. P. Kutter, and A. Wolff, “Measurements of scattered light on a microchip flow cytometer with integrated polymer based optical elements,” Lab Chip 4(4), 372–377 (2004). [CrossRef] [PubMed]
17. C. McGuckin, M. Jurga, H. Ali, M. Strbad, and N. Forraz, “Culture of embryonic-like stem cells from human umbilical cord blood and onward differentiation to neural cells in vitro,” Nat. Protoc. 3(6), 1046–1055 (2008). [CrossRef] [PubMed]
18. X. T. Su, S. E. Kirkwood, H. Gul, K. Singh, M. Z. Islam, A. Janowska-Wieczorek, W. Rozmus, and Y. Y. Tsui, “Light scattering characterization of single biological cells in a microfluidic cytometer,” Proc. SPIE 7386, 738602, 738602-8 (2009). [CrossRef]
20. A. Taflove, Computational Electrodynamics: The Finite-Difference Time-Domain Method, (Artech House, Inc., Norwood, Massachusetts, 1995).
21. A. Dunn and R. Richards-Kortum, “Three-dimensional computation of light scattering from cells,” IEEE J. Sel. Top. Quantum Electron. 2(4), 898–905 (1996). [CrossRef]
22. X. Li, A. Taflove, and V. Backman, “Recent progress in exact and reduced-order modeling of light-scattering properties of complex structures,” IEEE J. Sel. Top. Quantum Electron. 11(4), 759–765 (2005). [CrossRef]
23. C. G. Liu, C. Capjack, and W. Rozmus, “3-D simulation of light scattering from biological cells and cell differentiation,” J. Biomed. Opt. 10(1), 014007 (2005). [CrossRef]
24. J. Q. Lu, P. Yang, and X. H. Hu, “Simulations of light scattering from a biconcave red blood cell using the finite-difference time-domain method,” J. Biomed. Opt. 10(2), 024022 (2005). [CrossRef] [PubMed]
25. M. Majka, A. Janowska-Wieczorek, J. Ratajczak, K. Ehrenman, Z. Pietrzkowski, M. A. Kowalska, A. M. Gewirtz, S. G. Emerson, and M. Z. Ratajczak, “Numerous growth factors, cytokines, and chemokines are secreted by human CD34(+) cells, myeloblasts, erythroblasts, and megakaryoblasts and regulate normal hematopoiesis in an autocrine/paracrine manner,” Blood 97(10), 3075–3085 (2001). [CrossRef] [PubMed]
26. L. Marquez-Curtis, A. Jalili, K. Deiteren, N. Shirvaikar, A. M. Lambeir, and A. Janowska-Wieczorek, “Carboxypeptidase M expressed by human bone marrow cells cleaves the C-terminal lysine of stromal cell-derived factor-1alpha: another player in hematopoietic stem/progenitor cell mobilization?” Stem Cells 26(5), 1211–1220 (2008). [CrossRef] [PubMed]
27. D. Watson, N. Hagen, J. Diver, P. Marchand, and M. Chachisvilis, “Elastic light scattering from single cells: orientational dynamics in optical trap,” Biophys. J. 87(2), 1298–1306 (2004). [CrossRef] [PubMed]
28. R. De, A. Zemel, and S. A. Safran, “Dynamics of cell orientation,” Nat. Phys. 3(9), 655–659 (2007). [CrossRef]
29. M. D. Abramoff, P. J. Magelhaes, and S. J. Ram, “Image Processing with ImageJ,” Biophoton. Int. 11, 36–42 (2004).