1. INTRODUCTION

Embryonic heart development is a dynamic process involving genetic, mechanical, chemical, and biological factors. The cardiovascular system is one of the first systems to develop in the embryo, and heart formation and blood flow occur during the very early stages of development. It has long been thought that perturbations that occur during embryonic heart development lead to congenital heart defects, the most common malformation among newborns [1]. Blood flow is known to influence gene and protein expression in the mature cardiovascular system (see review [2]). Recent studies have also shown that blood flow influences the structural development of the embryonic heart [3, 4]. At present, the actual mechanism by which the heart pumps blood at the earliest stages of development is still under great scrutiny [5, 6]. One main reason for this is the inability to noninvasively measure blood flow while imaging cardiac structures in the earliest beating hearts.

Cardiovascular development has been difficult to observe in vivo. Current versions of widely used imaging techniques have limitations of spatial or temporal resolution or imaging depth or are impractical for longitudinal studies. Magnetic resonance microscopy (MRM) has been shown to be capable of producing 3D cardiac images from the chicken embryo with resolutions as high as $25\mu {\mathrm{m}}^3$ [7, 8]. However, for sufficient image contrast and resolution, the chick embryo hearts were perfused with contrast agents, and image acquisition times lasted approximately $30\mathrm{h}$. The heart was also arrested in diastole to eliminate motion artifacts during the long acquisition period. These requirements for high-resolution images using MRM inhibit the ability to conduct longitudinal studies. Similarly, 4D microcomputed tomography (micro-CT) of the mouse heart has been reported with $200\mu \mathrm{m}$ resolution [9] using gating techniques or $80\mu \mathrm{m}$ resolution without gating [10]. The latter resolution capability, however, still required data acquisitions up to $30\min$ and sacrifice of the animal [11].

Microscopy techniques such as microparticle velocimetry [12] and confocal microscopy [13] are commonly used to study cardiovascular development in chicken embryo and zebrafish animal models. Microparticle velocimetry measures whole-field blood velocity by tracking the motion of injected tracers, such as fluorescent liposomes. Confocal microscopy is an invaluable technique for studying zebrafish heart development [3]. However, the limited imaging depth of confocal microscopy techniques has limited detailed studies of blood flow dynamics through the embryonic heart to zebrafish. These microscopy techniques are also dependent on the injection of fluorophores with unknown cytotoxic effects during development.

Ultrasound biomicroscopy (UBM) studies of early heart development have utilized high-frequency transducers $(40–55\mathrm{MHz})$ to achieve axial and lateral resolutions as high as 28 and $62\mu \mathrm{m}$, respectively [14]. UBM is commonly used to image embryonic mouse hearts and to assess their cardiac function [15, 16]. More recently, high-resolution UBM has been demonstrated for imaging the embryonic chick heart, at stages as early as Hamburger–Hamilton (HH) stage 12 [17]. In addition to 2D imaging capabilities, UBM can provide Doppler blood flow measurements in the living embryo. Though noninvasive, ultrasound techniques require the transducer to be in acoustic contact with the sample. UBM provides sufficient resolution to identify the embryonic chicken heart at HH 12; however, identification of distinct structures, such as the myocardial and endocardial layers in the early heart tube, is still not possible. Additionally, in vivo volumetric imaging of the chicken embryo has not been possible with UBM.

Optical coherence tomography (OCT) has been shown to be well suited for imaging embryonic hearts due to its high resolution $(5–20\mu \mathrm{m})$ and up to $2\mathrm{mm}$ penetration depth [18, 19, 20, 21, 22]. OCT is an imaging modality that noninvasively provides in vivo cross-sectional images based on backreflected light [23]. The development of Fourier-domain OCT techniques [24], including swept-source and spectrometer-based spectral-domain OCT (SDOCT), has now enabled high-speed imaging while maintaining high sensitivity [25, 26, 27, 28]. OCT not only has the capability of providing one-, two-, and three-dimensional images of the live embryo, but it can also provide functional information in the form of Doppler blood flow imaging [29, 30, 31]. In vivo quantitative blood flow measurements in the human retina as well as chick embryo vasculature have also recently been demonstrated [32, 33].

The relative timing of some important stages of heart development in the human, mouse, chick, and zebrafish are provided in Table 1 . Heart development is commonly studied in mouse, chick, and zebrafish embryo animal models. Zebrafish are often used because they have known genomes and rapid developmental cycles. Their small size and transparency also make them attractive models for optical imaging techniques. Chick and mouse embryos, however, are preferred animal models because their cardiovascular development and function are closely related to humans. An additional advantage of using chick embryos is that they develop outside of the womb, making them easier to access for imaging and more viable for longitudinal studies.

In the chick, the heart tube fuses by approximately $30\mathrm{h}$ of development or HH stage $8+∕9$- [36]. Following the formation of the heart tube at HH 10 $\left(36\mathrm{h}\right)$, coordinate contractions begin. The heart tube expands in size and the first indications of the tube differentiating into the outflow tract and primitive ventricle are observed [37]. Additionally, the heart tube begins to bend or loop to one side, indicating entrance into the looping process that will last several stages and is an important step leading to septation into four chambers [37]. Detailed scanning electron microscopy images of the chicken embryo heart through key developmental stages have been presented in [37].

The heart tube consists of three layers: myocardium, endocardium, and an incompressible material called cardiac jelly, which is sandwiched between the other two. After stage HH 12, the cardiac jelly in the outflow and inflow regions of the heart tube thickens and creates a bulge or cushion into the tube lumen. These endocardial cushions will eventually develop into valves and septa of the heart [38]. Finite-element simulations of blood flow through an embryonic heart tube modeled with and without endocardial cushions [6] suggest that these cushions serve an even greater purpose than valve precursors in that they are necessary for facilitating net forward blood flow out of the embryonic heart tube.

Structural and functional development of the chick embryo heart occur simultaneously. Around HH $10+∕\mathrm{HH}$ 11, just after the heart tube begins to beat, blood flow is first observed. Unlike adult hearts, it is believed that at this early stage of development blood flow is not used for convective transport of oxygen and nutrients or removal of waste [39]. Blood flow, in fact, may be necessary for normal structural development [39]. Modification of intracardiac blood flow patterns by ligation of the vitelline vein can alter ventricular structure and reduce cardiac performance [4, 40, 41] perhaps by inducing changes in shear-stress-related gene expression [42]. Although these studies have established a link between blood flow and structural development, the precise genes and structural changes altered by early blood flow and implications for normal cardiogenesis are still incompletely understood. Limitations in research toward this understanding strongly lie in the inability to image structure and quantify blood flow simultaneously.

In this paper we present what we believe to be the first in vivo volumetric images of the embryonic chicken heart as it fuses into a heart tube and enters the looping process (HH 9–HH 15). In vivo volumetric imaging has also enabled quantitative measurement of hemodynamics. We have previously introduced spectral Doppler velocimetry for quantitative studies of hemodynamics and direct correlation with contractile dynamics of developing cardiovasculature [33]. Here we present spectral Doppler velocimetry measurements from the embryonic heart as it develops from the onset of blood flow to the beginning stages of cushion formation (HH 11–HH 14).

2. EXPERIMENTAL PROCEDURES

2A. Chicken Embryos

We incubated fertilized Hubert Ross chicken eggs at $38°\mathrm{C}$ for a total of $3\text{\hspace{0.17em} days}$. Before the first imaging session, a small part of the shell and chorionic membrane was removed for optical access to the embryo [Fig. 1a ]. Beginning at $24\mathrm{h}$ of development, the eggs were removed from the incubator, Doppler and volumetric images were acquired, then eggs were returned to the incubator for further development. Imaging sessions were conducted under normal room temperature $(23°\mathrm{C})$ conditions. This process was repeated every $3\mathrm{h}$ over $72\mathrm{h}$ of development. Each imaging session lasted less than $3\min$ to minimize the time the embryo spent outside of the incubator. A total of four embryos were studied from stages HH 9 through HH 15.

2B. Optical Coherence Tomography System

This study employed a SDOCT system specialized for small animal imaging, which is based on an InGaAs line-scan camera operating at an $18.9\mathrm{kHz}$ A-scan rate [Fig. 1b] [43]. The system was illuminated by a superluminescent diode (InPhenix) centered at $1310\mathrm{nm}$ [$\Delta \lambda =84\mathrm{nm}$ full width at half-maximum (FWHM)]. The optical power (approximately $30\mathrm{mW}$) was split into sample and reference arms using a $50∕50$ fiber optic coupler (AC Photonics). The reference arm power was attenuated to achieve near shot-noise-limited performance [26]. Sample arm light was coupled into the optical path of a stereo-zoom microscope (Zeiss) modified with a magnetic mirror (Optics in Motion, Inc.) for two-dimensional lateral scanning of the beam across the sample. The designed spot size of the SDOCT microscope was $10\mu \mathrm{m}$ scannable over a maximum $9\mathrm{mm}\times 9\mathrm{mm}$ area. Spectral interference was detected using a custom-designed spectrometer with a $512\text{\hspace{0.17em} pixel}$ InGaAs CCD camera (512LX, Sensors Unlimited). The spectrally dispersed interferometric light was focused into a $17\mu \mathrm{m}$ diameter spot (FWHM at $1310\mathrm{nm}$, $25\mu \mathrm{m}$ at $1280\mathrm{nm}$) on the $50\mu \mathrm{m}$ pixels, as predicted by optical system modeling (ZEMAX) of the spectrometer.

The measured signal-to-noise ratio from an attenuated ($-55\mathrm{dB}$ neutral-density filter) ideal reflector was $45\mathrm{dB}$ with $5\mathrm{mW}$ total optical power on the sample, resulting in a calculated $100\mathrm{dB}$ system dynamic range. The optical power was attenuated to $5\mathrm{mW}$ to reduce autocorrelation caused by bright reflections at the surface of the egg yolk. Although exposure limits for embryos are not well known, there are published reports, based on American National Standards Institute (ANSI) standards, that say it is safe to use more than $5\mathrm{mW}$ continuous-wave power at $1310\mathrm{nm}$ for imaging human cornea [44], which is less than other OCT chick embryo studies [45]. Using software from Bioptigen, Inc., this system acquires, processes, and displays depth-resolved (B-mode) images consisting of $512\times 512\text{\hspace{0.17em} pixels}$ in real time at $26\text{\hspace{0.17em} frames}∕\mathrm{s}$. There is approximately $11\mathrm{ms}$ of delay between frames due to the limited processing speed of the computer. The axial resolution for this system was $12\mu \mathrm{m}$, which was the measured FWHM of the point-spread function from a reference reflector. The signal falloff at an imaging depth of $1.5\mathrm{mm}$ was approximately $-12\mathrm{dB}$ [Fig. 1c].

2C. Volumetric Image Reconstruction

Each volume dataset consisted of $512\times 256\times 256\text{\hspace{0.17em} pixels}$ ($\text{lateral}\times \text{depth}\times \text{number}$ of slices) that were scanned across a $5\times 5\mathrm{mm}$ area ($5\mathrm{mm}\times 1.8\mathrm{mm}\times 5\mathrm{mm}$ volume datasets). To increase the frame rate, the lateral sampling was decreased to approximately $20\mu \mathrm{m}$ lateral resolution. Volumetric reconstructions and orthogonal cross sections of the OCT volume datasets were produced using the VolView OSA visualization platform, currently called OSA ISP (OSA and Kitware, Inc.). To isolate the heart tube structure from the entire embryo, semitransparent surface renderings were made by manual segmentation of the heart tube wall, in three dimensions, using commercial software (Mercury Systems, Inc.).

2D. Doppler and Spectral Doppler Velocimetry

In Doppler SDOCT imaging, Doppler-shifted sample arm light reflected from moving scatterers in the sample induces phase shifts between sequential spectral interferograms collected at the same sample position [46, 47]. The relative phase shift observed is correlated to the velocity of the moving scatterers using the following expression [29]:

Without knowing the angle of flow, Doppler OCT imaging can only provide relative frequency shifts induced by fluid motion. We have developed a technique called spectral Doppler velocimetry (SDV) to correlate quantitative measurement of depth-resolved blood flow dynamics with high-speed M-mode OCT imaging, at a user-defined location in the sample [33]. SDV is a SDOCT analog to pulsed Doppler ultrasound [16]. However, unlike pulsed Doppler ultrasound, which averages blood flow measurements within a focal volume, SDV is inherently depth resolved and therefore produces hemodynamic measurements at all depths, simultaneously. SDV measurements were made by acquiring Doppler M-mode measurements along a single lateral position at $3800\text{\hspace{0.17em} lines}∕\mathrm{s}$. Steps taken to acquire SDV measurements are illustrated in Fig. 2 . First, real-time Doppler B-mode imaging is used to locate the region of interest [Fig. 2a]. Once the line of interest is identified, we acquire 2048 lines of Doppler M-mode at that position. Without moving the embryo, a volumetric dataset is subsequently acquired, which is then used to measure the angle of flow to determine velocity from Eq. (1) and can later be used to pinpoint the SDV measurement location [Fig. 2b]. The angle of flow was measured by rendering the vessel of interest (Mercury Systems, Inc.) and measuring the angle of the vessel relative to the y axis, which corresponds to the direction of the Doppler OCT beam [Fig. 2b, green angle]. This technique to use volume renderings to measure the angle of flow enables accurate calculation of velocity by accounting for measurements taken at oblique angles relative to the vessel.

2E. Phase Unwrapping

Doppler OCT measurements are constrained by the integration time of the system, where the integration time is set by the readout time of the CCD camera. When fluid flow induces a Doppler phase shift greater than $2\pi$ during the integration period, the measured signal becomes phase-wrapped and velocity is not uniquely extractable from the phase. Also, as expressed in Eq. (1), the minimum and maximum detectable velocities are dependent on the flow angle. In false-color Doppler OCT images, wrapped phase appears as red–blue rings (e.g., see Fig. 5 HH 14, $0\mathrm{ms}$, below). Figure 3 contains a plot of the theoretical minimum and maximum (non-phase-wrapped) detectable velocities for our system. In most cases the vessel of interest is oriented at an angle between 30 and $60\mathrm{deg}$, thus corresponding to maximum detectable velocities between 12 and $18\mathrm{mm}∕\mathrm{s}$. To address velocities greater than this range, we implemented a previously developed cellular-automata method for phase unwrapping [49]. Cellular-automata is a computational method for removing discontinuities by performing local neighborhood tests in a nondirectional manner. Detailed steps in performing this technique for 2D phase unwrapping are described by Ghighlia et al. [50].

3. RESULTS

3A. In Vivo Volume Images of Embryonic Heart Development

Three-dimensional OCT images were acquired, in ovo, from live embryos over development from stages HH 9 to HH 15. This period of development is of primary interest to developmental biologists because it covers the initial formation of the embryonic heart tube, through the onset of blood flow, into the looping stage (prior to septation into a four-chambered system). Previously described technologies are unable to adequately image at these stages of development, especially in an in ovo setting. Figure 4 contains volumetric images of an embryonic heart at stages HH 9, HH 12, and HH 15.

Three-dimensional OCT reconstructions (Fig. 4, second column) were constructed from a stack of 256 cross-sectional B-mode OCT images (Fig. 4, first column) that were acquired in $9.8\mathrm{s}$. These reconstructions provide gross inspection of anatomical features and morphology over development. In Fig. 4, third column, we show surface renderings of just the heart tube at the three stages of development. These renderings enable visualization of the heart tube without obstruction from other anatomical structures, such as the neural tube. These semitransparent renderings were produced by manual segmentation of the heart tube wall, including the myocardial and endocardial layers, as well as the cardiac jelly, which resides between the two layers. The light purple indicates regions of the heart tube wall, whereas the darker region is the lumen of the heart tube where blood passes through. Also in the third column in Fig. 4 are representative light microscopy images of excised embryonic chick hearts at the same stages of development.

In the top row of Fig. 4 we show that at stage HH 9 the endocardial tubes have begun to fuse into a heart tube starting from the head and continuing toward the tail. At this stage, the heart tube is only a few hundred micrometers in diameter and sits underneath the developing neural tube. The size and location of the forming heart tube at this stage prevents other imaging technologies from imaging it in vivo.

In Fig. 4, middle row, it is clear that by stage HH 12 the endocardial tube has fully fused as well as begun the looping stage, as evidenced by the bend in the ventricle region. At this point, the heart tube wall has thickened in large part due to the formation of cardiac jelly. In the volumetric reconstructions, it is evident that the embryo has developed significantly from stage HH 9 to stage HH 12. The embryo as a whole has turned slightly to the side and individual somites can be identified (Fig. 4, center image, arrows labeled s). These are key developmental factors that are used in staging embryos [36].

At stage HH 15 (Fig. 4, bottom row) the heart tube is in the looping stage, bringing the outflow track and atrial regions of the heart tube closer together. Volumetric imaging always provides better insight into the shape and size of anatomical features than traditional video microscopy in which only a projection is visualized. As can be seen here, during looping the heart tube has a corkscrew shape, making volume imaging imperative for accurate visualization of anatomical features relative to each other in the beating heart.

3B. Correlation of Blood Flow with Heart Development

Two-dimensional Doppler OCT imaging provides noninvasive cross-sectional visualization of the embryonic heart with sufficient spatial resolution to identify key anatomical features and relate them with relative fluid flow. A series of Doppler OCT images of blood flow through the primitive ventricle of the heart tube at stages HH 11 and HH 14 are shown in Fig. 5 . As is standard in color Doppler ultrasound, we display false-color Doppler OCT images (red–blue) superimposed on SDOCT intensity images (gray scale); this imaging mode is performed in real time. To clearly visualize the structural features of the heart tube, the Doppler images were thresholded to display Doppler values greater than 15% of maximum. As a result, some of the Doppler signal near the edges of the heart tube may have been thesholded out, creating cigar-shaped Doppler signals. All quantitative Doppler data was based on nonthresholded values. Blood flow begins near stage HH 11 of development; therefore these images were acquired during the very first hours of blood flow initiation. From these cross-sectional views, the myocardium and endocardium are clearly resolved (arrows labeled m and ec, respectively). Blood flows from the primitive ventricle out of the heart tube through the outflow tract (dotted arrow). Then by stage HH 14 the Doppler OCT images reveal the continued development of the embryonic heart tube. At the end of the outflow tract, the endocardial cushions have begun to form. These cushions are the precursors to the aortic valves and are believed to have significant influence on the mechanism of blood flow [6].

To correlate blood flow dynamics with heart development, we acquired volumetric, Doppler, and SDV measurements from the outflow tract at these stages. Figure 6 contains volumetric reconstructions, surface renderings, and SDV measurements from these two stages of development. The volume renderings [Figs. 6a, 6d] show that at stage HH 11 the heart tube began to bend in the ventricle region, indicating the initial phase of the looping process. By stage HH 14, the heart tube bulged toward the right in a later phase of looping. Along with providing identification of structural stage, these renderings were used to measure the flow angle and calculate the blood flow velocity using Eq. (1). SDV measurements (velocity versus depth versus time) and M-mode (depth or A-scan versus time) are simultaneously acquired along the dashed yellow line through the center of the outflow tract shown in the insets in Figs. 6a, 6d. The combination of SDV measurements [Figs. 6b, 6e] and M-mode [Fig. 6c, 6f] enable correlation of blood flow with expansion and dilation of the outflow tract. Volume reconstructions of the HH 11 and HH 14 hearts are also provided in the bottom of Figs. 6a, 6d, respectively. At both stages, the bulk of the outflow occurred as the diameter of the outflow tract decreased [Figs. 6b, 6f, gray dashed lines], indicating that the region under observation was actively moving the blood by local contraction. This is consistent with a peristaltic model of pumping by the chick heart [6]. At contraction, there is nearly zero net blood flow [Figs. 6b, 6e, blue dashed lines], due to blood being constricted by the closing of the outflow tract. At this point in the heart beat cycle, however, there was a difference between stages HH 11 and HH 14. The HH 11 outflow shows a definite backflow as the outflow tract begins to expand [Fig. 6b, red arrow]. This backflow is related to the undeveloped endocardial cushions (valve precursors) at stage HH 11, which allows rapid regurgitation of blood into the outflow region. As shown in Fig. 5, at HH 14, the endocardial cushions have begun to form. As a result there is less regurgitation of blood. Peak blood flow velocity only slightly increased over the two stages of development, from $11\text{to}14\mathrm{mm}∕\mathrm{s}$. These measurements are within the range of $14.3\text{to}30\mathrm{mm}∕\mathrm{s}$ reported using laser Doppler velocimetry [51], microparticle image velocimetry [12], and pulsed Doppler ultrasound [17, 52] techniques. These reported measurements were taken in embryos between stage 12 [51] and stage 24 [17] and, to our knowledge, are the closest reported measurements related to our studies.

4. DISCUSSION

The volume datasets shown here were acquired in vivo, producing motion artifacts that affect image quality. At stage HH 9 prior to heartbeat initiation, volume renderings reveal a smooth surface (Fig. 4, top row). Coordinated contractions began at stage HH 10 at a rate between $120\text{to}180\text{\hspace{0.17em} beats}∕\min$. Motion artifacts due to beating do not degenerate single B-mode images (Fig. 4, first column) because they were acquired in only $27\mathrm{ms}$ (with an $11\mathrm{ms}$ time gap between frames). However, volume datasets comprising multiple B-mode images acquired in $9.8\mathrm{s}$ cover 14–30 beats; therefore the volumetric reconstructions (Fig. 4, second column, HH 12 and HH 15) and renderings (Fig. 4, third column, HH 12 and HH 15) contained ripples. Similar motion artifacts are present in Figs. 2, 6. Since we were primarily interested in the overall morphology, the motion artifacts in the volume renderings were not a concern. For quantitative measurement of structural dynamics in three dimensions, gated [20] or higher-speed [45] OCT techniques would be required. In this case, volume renderings acquired by our SDOCT system provide staging information from HH 9 to HH 15, pinpoint the SDV measurement location, and determine the vessel angle to quantify blood flow.

To correlate flow through the heart tube in relation to the heart tube itself, we utilized M-mode imaging [Figs. 6c, 6f]. These images are acquired simultaneously with SDV measurements providing direct correlation of flow with the heart tube diameter. This permits a higher temporal resolution through a single region than would ordinarily be possible. Although we have suggested here that endocardial cushion formation contributes to the blood flow velocity patterns observed over time, other developmental processes will need to be individually analyzed, such as the expression and regression of cardiac jelly, the role of looping, and changes in the cytoskeleton [53]. The inherent advantage that SDV measurements are spatially resolved, in depth, has not yet been completely exploited. Blood flow velocity profiles through the heart tube could potentially be utilized to measure volumetric flow rates [32] and flow-induced shear stress [54], and we are planning to implement the measurements in the future.

Heart development is one of the most dynamic events experienced by the embryo. The heart and cardiovascular system are continually adding, removing, changing, and remodeling tissues and structures. Congenital heart disease, a common and potentially life-threatening disorder, begins in this milieu. Whether the inciting factors are inherent to the tissue (expression of genes leading to anatomical defects) or whether blood flow alterations and mishaps precede the abnormal structural defect (blood flow and shear stress inciting altered gene expression) has not been fully understood. This is largely because there have not been imaging methods with the combined spatial resolution, temporal resolution, and imaging depth to sufficiently measure structural changes—simultaneously with accurate measurements of blood flow—in chick embryo hearts at these early stages of formation. Here we have demonstrated the first in vivo volumetric images of the developing chicken embryo heart and correlation of blood flow and structural dynamics at the beginning stages of embryonic development using SDOCT in conjunction with spectral Doppler velocimetry. We expect that continued work using these tools will provide critical missing information regarding the role of fluid dynamics in the timing of heart development.

ACKNOWLEDGMENTS

The authors acknowledge the contribution of Margaret Kirby for providing the chicken embryo preparations, Laura Barbosky for her assistance in staging the embryos, and Tzuo Law for implementation of the cellular-automata phase-unwrapping algorithms. This study was supported by the National Institutes of Health (NIH) (R21-EB006338).

Table 1. Milestones of Early Heart Development in Different Species^a

View Table

 

Fig. 1 SDOCT microscope system for small animal imaging. (a) Chicken embryo preparation, (b) SDOCT system setup. A low-coherence light source $(\lambda =1310\mathrm{nm})$ was used in a fiber-based Michelson interferometer design where the optical power was split using a $50∕50$ coupler into reference and sample arms. The interferogram was measured using a custom-made spectrometer containing a 512 element InGaAs CCD detector (Sensors Unlimited). Two-axis scanning of the SDOCT beam across the sample was performed using an adapted Zeiss stereo zoom microscope. (c) SDOCT signal falloff as a function of depth. $-12\mathrm{dB}$ signal falloff was measured at $1.5\mathrm{mm}$ imaging depth. SLD, superluminescent diode (InPhenix); L, lens; M, mirror; M2, dual-axis scanning mirror (Optics in Motion); G, grating (Wasatch).

Download Full Size | PDF

 

Fig. 2 SDV measurement and volume rendering of chicken embryo vessel. (a) Doppler OCT image (blue) superimposed on a SDOCT intensity image of a cross section of a vessel. The vertical dashed line indicates the location where SDV measurements were acquired. (b) 3D surface rendering of the vessel. The surface rendering was used to measure the angle of blood flow (green) relative to the SDOCT scanning beam. Here two orthogonal OCT planes are displayed for which the $x–y$ plane corresponds to the image shown in (a). Scale $\text{bar}=200\mu \mathrm{m}$.

Download Full Size | PDF

 

Fig. 3 Detectable flow velocity range. Plot of the theoretical minimum and maximum detectable velocities, as per Eq. (1) using our $19\mathrm{kHz}$, $1310\mathrm{nm}$, SDOCT system. Higher velocities than the maximum plotted here are detectable but result in phase-wrapping artifacts.

Download Full Size | PDF

 

Fig. 4 Volume images of chicken embryo heart development from HH 9 to HH 15. Top row, SDOCT cross sections in $x–y$ and $y–z$ planes, volume reconstruction, and surface rendering of the fusing heart tube at HH 9 (View 1). Middle row, SDOCT cross sections in the $x–y$ and $y–z$ planes, volume reconstruction, and surface rendering of the fused heart tube at HH 12 (View 2). Bottom row, SDOCT cross sections in the $x–y$ and $y\text{–}z$ planes, volume reconstruction, and surface rendering of the looping heart tube at HH 15 (View 3). Light purple represents the heart tube wall consisting of the myocardium, cardiac jelly, and endocardium layers. Dark purple represents the heart tube lumen. The microscope images the in the third column were taken from representative embryos at or near the same stage of development. Cross sections and volume reconstructions were made using the VolView OSA visualization platform, currently called OSA ISP. nt, neural tube; ht, heart tube; s, somites; a, atrial buds; v, primitive ventricle; oft, outflow tract. Scale $\text{bar}=500\mu \mathrm{m}$.

Download Full Size | PDF

 

Fig. 5 Time series of Doppler SDOCT images of blood flow through the outflow tract. A series of Doppler B-mode images (red–blue) are overlaid on SDOCT images of the primitive ventricle and outflow tract at HH 11 (top) and HH 14 (bottom), respectively. At HH 14, the primitive ventricle increased in diameter and the endocardial cushion (ec) is clearly visible at 80 and $560\mathrm{ms}$. All images were acquired at $12\text{\hspace{0.17em} frames}∕\mathrm{s}$. $0\mathrm{ms}$ was set at diastole, or when the outflow tract was most contracted. Dotted arrows show the direction of flow. e, endocardium layer; m, myocardium layer; cj, cardiac jelly. Scale $\text{bar}=250\mu \mathrm{m}$.

Download Full Size | PDF

 

Fig. 6 SDV measurements at HH 11 and HH 14. Surface renderings and volume reconstructions of the heart tube at (a) HH 11 (View 4) and (d) HH 14 (View 5). Insets show Doppler OCT images of blood flow out of the outflow tract (out of the plane). SDV measurements were acquired along the yellow dashed line. The heart tube is in the midst of looping at HH 14 as evidenced by the U-shaped rendering and appearance of the inflow tract in (d). Blood flow velocity dynamics from (b) HH 11 and (e) HH 14. (c), (f) M-mode images where the vertical axis is depth and the horizontal is time. These M-mode images are acquired simultaneously with the velocity measurement, enabling correlation of heart tube contractions. SDV measurements were taken along the green dotted line in (c) and (f). oft, outflow tract; ift, inflow tract. Scale $\text{bar}=250\mu \mathrm{m}$.

Download Full Size | PDF

1. American Heart Association, “Congenital heart defects in children fact sheet” (American Heart Association, 2004), available at www.americanheart.org/children.

2. R. S. Reneman, T. Arts, and A. P. Hoeks, “Wall shear stress—an important determinant of endothelial cell function and structure—in the arterial system in vivo. Discrepancies with theory,” J. Vasc. Res. 43, 251–269 (2006). [CrossRef]   [PubMed]  

3. J. R. Hove, R. W. Köster, A. S. Forouhar, G. Acevando-Bolton, S. E. Fraser, and M. Gharib, “Intracardiac fluid forces are an essential epigenetic factor for embryonic cardiogenesis,” Nature 421, 172–177 (2003). [CrossRef]   [PubMed]  

4. N. T. Ursem, C. S. Stekelenburg-de Vos, J. W. Wladimiroff, R. E. Poelmann, A. C. Gittenberger-de Groot, N. Hu, and E. B. Clark, “Ventricular diastolic filling characteristics in stage-24 chick embryos after extra-embryonic venous obstruction,” J. Exp. Biol. 207, 1487–1490 (2004). [CrossRef]   [PubMed]  

5. A. S. Forouhar, M. Liebling, A. Hickerson, A. Nasiraei-Moghaddam, H.-J. Tsai, J. R. Hove, S. E. Fraser, M. E. Dickinson, and M. Gharib, “The embryonic vertebrate heart tube is a dynamic suction pump,” Science 312, 751–753 (2006). [CrossRef]   [PubMed]  

6. L. A. Taber, J. Zhang, and R. Perucchio, “Computational model for the transition from peristaltic to pulsatile flow in the embryonic heart tube,” J. Biomech. Eng. 129, 441–449 (2007). [CrossRef]   [PubMed]  

7. X. Zhang, T. M. Yelbuz, G. P. Kofer, M. A. Choma, M. L. Kirby, and G. A. Johnson, “Improved preparation of chick embryonic samples for magnetic resonance microscopy,” Magn. Reson. Med. 49, 1192–1195 (2003). [CrossRef]   [PubMed]  

8. T. M. Yelbuz, X. Zhang, M. A. Choma, H. A. Stadt, M. Zdlanowicz, G. A. Johnson, and M. L. Kirby, “Approaching cardiac development in three dimensions by magnetic resonance microscopy,” Circulation 108, e154–e155 (2003). [CrossRef]   [PubMed]  

9. C. T. Badea, B. Fubara, L. W. Hedlund, and G. A. Johnson, 4-D micro-CT of the mouse heart,” Mol. Imaging 4, 110–116 (2005). [PubMed]  

10. D. W. Holdsworth, M. Drangova, and A. Fenster, “A high-resolution XRII-based quantitative volume CT scanner,” Med. Phys. 20, 449–462 (1993). [CrossRef]   [PubMed]  

11. M. J. Paulus, S. S. Gleason, S. J. Kennel, P. R. Hunsicker, and D. K. Johnson, “High resolution x-ray computed tomography: an emerging tool for small animal cancer research,” Neoplasia 2, 62–70 (2000). [CrossRef]   [PubMed]  

12. P. Vennemann, K. T. Kiger, R. Lindken, B. C. W. Groenendijk, S. Stekelenburg-de Vos, T. L. M. ten Hugen, N. T. C. Ursem, R. E. Poelmann, J. Westerweel, and B. P. Hierck, “In vivo micro particle image velocimetry measurements of blood-plasma in the embryonic avian heart,” J. Biomech. 39, 1191–1200 (2006). [CrossRef]  

13. M. Liebling, A. S. Forouhar, M. Gharib, S. E. Fraser, and M. E. Dickinson, “Four-dimensional cardiac imaging in living embryos via postacquisition synchronization of nongated slice sequences,” J. Biomed. Opt. 10, 054001 (2005). [CrossRef]   [PubMed]  

14. F. S. Foster, M. Y. Zhang, Y. Q. Zhou, G. Liu, J. Mehi, E. Cherin, K. A. Harasiewicz, B. G. Starkoski, L. Zan, D. A. Knapik, and S. L. Adamson, “A new ultrasound instrument for in vivo microimaging of mice,” Ultrasound Med. Biol. 28, 1165–1172 (2002). [CrossRef]   [PubMed]  

15. C. K. Phoon, O. Aristizabal, and D. H. Turnbull, “$40\mathrm{MHz}$ Doppler characterization of umbilical and dorsal aortic blood flow in the early mouse embryo,” Ultrasound Med. Biol. 26, 1275–1283 (2000). [CrossRef]   [PubMed]  

16. C. K. Phoon, O. Aristizabal, and D. H. Turnbull, “Spatial velocity profile in mouse embryonic aorta and Doppler-derived volumetric flow: a preliminary model,” Am. J. Physiol. Heart Circ. Physiol. 283, H908–H916 (2002). [PubMed]  

17. T. C. McQuinn, M. Bratoeva, A. DeAlmeida, M. Remond, R. P. Thompson, and D. Sedmera, “High-frequency ultrasonographic imaging of avian cardiovascular development,” Dev. Dyn. 236, 3503–3513 (2007). [CrossRef]   [PubMed]  

18. S. A. Boppart, G. J. Tearney, B. E. Bouma, J. F. Southern, M. E. Brezinski, and J. G. Fujimoto, “Noninvasive assessment of the developing Xenopus cardiovascular system using optical coherence tomography,” Proc. Natl. Acad. Sci. U.S.A. 94, 4256–4261 (1997). [CrossRef]   [PubMed]  

19. T. M. Yelbuz, M. A. Choma, L. Thrane, M. L. Kirby, and J. A. Izatt, “Optical coherence tomography: a new high-resolution imaging technology to study cardiac development in chick embryos,” Circulation 106, 2771–2774 (2002). [CrossRef]   [PubMed]  

20. M. W. Jenkins, F. Rothenberg, D. Roy, V. P. Nikolski, Z. Hu, M. Watanabe, D. L. Wilson, I. R. Efimov, and A. M. Rollins, “4D embryonic cardiography using gated optical coherence tomography,” Opt. Express 14, 736–748 (2006). [CrossRef]   [PubMed]  

21. W. Luo, D. L. Marks, T. S. Ralston, and S. A. Boppart, “Three-dimensional optical coherence tomography of the embryonic murine cardiovascular system,” J. Biomed. Opt. 11, 021014 (2006). [CrossRef]   [PubMed]  

22. J. Manner, L. Thrane, K. Norozi, and T. M. Yelbuz, “High-resolution in vivo imaging of the cross-sectional deformations of contracting embryonic heart loops using optical coherence tomography,” Dev. Dyn. 237, 953–961 (2008). [CrossRef]   [PubMed]  

23. D. Huang, E. A. Swanson, C. P. Lin, J. S. Schuman, W. G. Stinson, W. Chang, M. R. Hee, T. Flotte, K. Gregory, C. A. Puliafito, and J. G. Fujimoto, “Optical coherence tomography,” Science 254, 1178–1181 (1991). [CrossRef]   [PubMed]  

24. A. F. Fercher, C. K. Hitzenberger, G. Kamp, and S. Y. El-Zaiat, “Measurement of intraocular distances by backscattering spectral interferometry,” Opt. Commun. 117, 43–48 (1995). [CrossRef]  

25. R. Leitgeb, C. K. Hitzenberger, and A. F. Fercher, “Performance of Fourier domain vs. time domain optical coherence tomography,” Opt. Express 11, 889–894 (2003). [CrossRef]   [PubMed]  

26. M. A. Choma, M. Sarunic, C. Yang, and J. Izatt, “Sensitivity advantage of swept-source and Fourier-domain optical coherence tomography,” Opt. Express 11, 2183–2189 (2003). [CrossRef]   [PubMed]  

27. J. F. de Boer, B. Cense, B. H. Park, M. C. Pierce, G. J. Tearney, and B. E. Bouma, “Improved signal-to-noise ratio in spectral-domain compared with time-domain optical coherence tomography,” Opt. Lett. 28, 2067–2069 (2003). [CrossRef]   [PubMed]  

28. R. Huber, D. C. Adler, and J. G. Fujimoto, “Buffered Fourier domain mode locking: unidirectional swept laser sources for optical coherence tomography imaging at $\mathrm{370,000}\text{\hspace{0.17em} lines}∕\mathrm{s}$,” Opt. Lett. 31, 2975–2977 (2006). [CrossRef]   [PubMed]  

29. S. Yazdanfar, M. D. Kulkarni, and J. A. Izatt, “High resolution imaging of in vivo cardiac dynamics using color Doppler optical coherence tomography,” Opt. Express 1, 424–431 (1997). [CrossRef]   [PubMed]  

30. A. Mariampillai, B. A. Standish, N. R. Munce, C. Randall, G. Liu, J. Y. Jiang, A. E. Cable, I. A. Vitkin, and V. X. D. Yang, “Doppler optical cardiogram gated 2D color flow imaging at $1000\mathrm{fps}$ and 4D in vivo visualization of embryonic heart at $45\mathrm{fps}$ on a swept source OCT system,” Opt. Express 15, 1627–1638 (2007). [CrossRef]   [PubMed]  

31. N. V. Iftimia, D. X. Hammer, R. P. Ferguson, M. Mujat, D. Vu, and A. A. Ferrante, “Dual-beam Fourier domain optical Doppler tomography of zebrafish,” Opt. Express 16, 13624–13636 (2008). [CrossRef]   [PubMed]  

32. Y. Wang, B. A. Bower, J. A. Izatt, O. Tan, and D. Huang, “In vivo total retinal blood flow measurement by Fourier domain Doppler optical coherence tomography,” J. Biomed. Opt. 12, 041215 (2007). [CrossRef]   [PubMed]  

33. A. M. Davis, J. A. Izatt, and F. Rothenberg, “Quantitative measurement of blood flow dynamics in embryonic vasculature using spectral Doppler velocimetry,” Anat. Rec. (to be published).

34. M. C. Fishman and K. R. Chien, “Fashioning the vertebrate heart: earliest embryonic decisions,” Development 124, 2099–2117 (1997). [PubMed]  

35. B. J. Martinsen, “Reference guide to the stages of chick heart embryology,” Dev. Dyn. 233, 1217–1237 (2005). [CrossRef]   [PubMed]  

36. V. Hamburger and H. L. Hamilton, “A series of normal stages in the development of the chick embryo,” J. Morphol. 88, 49–92 (1951). [CrossRef]  

37. J. Männer, “Cardiac looping in the chick embryo: a morphological review with special reference to terminological and biomechanical aspects of the looping process,” Anat. Rec. 259, 248–262 (2000). [CrossRef]   [PubMed]  

38. A. D. Person, S. E. Klewer, and R. B. Runyan, “Cell biology of cardiac cushion development,” Int. Rev. Cytol. 243, 287–335 (2005). [CrossRef]   [PubMed]  

39. W. W. Burggren, “What is the purpose of the embryonic heart beat? Or how facts can ultimately prevail over physiological dogma,” Physiol. Biochem. Zool. 77, 333–345 (2004). [CrossRef]   [PubMed]  

40. B. Hogers, M. C. De Ruiter, A. C. Gittenberger-de Goot, and R. E. Poelmann, “Extraembryonic venous obstructions lead to cardiovascular malformations and can be embryolethal,” Cardiovasc. Res. 41, 87–99 (1999). [CrossRef]   [PubMed]  

41. K. Tobita and B. B. Keller, “Maturation of end-systolic stress-strain relations in chick embryonic myocardium,” Am. J. Physiol. Heart Circ. Physiol. 279, H216–H224 (2000). [PubMed]  

42. B. C. W. Groenendijk, B. P. Hierck, J. Vrolijk, M. Baiker, M. J. B. M. Pourquie, A. C. Gittenberger-de Groot, and R. E. Poelmann, “Changes in shear stress-related gene expression after experimentally altered venous return in the chicken embryo,” Circ. Res. 96, 1291–1298 (2005). [CrossRef]   [PubMed]  

43. S. H. Yun, G. Tearney, B. Bouma, B. Park, and J. de Boer, “High-speed spectral-domain optical coherence tomography at $1.3\mu \mathrm{m}$ wavelength,” Opt. Express 11, 3598–3604 (2003). [CrossRef]   [PubMed]  

44. S. Radhakrishnan, A. M. Rollins, J. E. Roth, S. Yazdanfar, V. Westphal, D. S. Bardenstein, and J. A. Izatt, “Real-time optical coherence tomography of the anterior segment at $1310\mathrm{nm}$,” Arch. Ophthalmol. (Chicago) 119, 1179–1185 (2001).

45. M. W. Jenkins, D. C. Adler, M. Gargesha, R. Huber, F. Rothenberg, J. Belding, M. Watanabe, D. L. Wilson, J. G. Fujimoto, and A. M. Rollins, “Ultrahigh-speed optical coherence tomography imaging and visualization of the embryonic avian heart using a buffered Fourier domain mode locked laser,” Opt. Express 15, 6251–6267 (2007). [CrossRef]   [PubMed]  

46. Z. Chen, T. E. Milner, S. Srinivas, X. Wang, A. Malekafzali, M. J. C. Van Gemart, and J. S. Nelson, “Noninvasive imaging of in vivo blood flow velocity using optical Doppler tomography,” Opt. Lett. 22, 1119–1121 (1997). [CrossRef]   [CrossRef]   [PubMed]  

47. R. Leitgeb, L. Schmetterer, W. Drexler, A. Fercher, R. Zawadzki, and T. Bajraszewski, “Real-time assessment of retinal blood flow with ultrafast acquisition by color Doppler Fourier domain optical coherence tomography,” Opt. Express 11, 3116–3121 (2003). [CrossRef]   [PubMed]  

48. L. Wang, Y. Wang, S. Guo, J. Zhang, M. Bachman, G. P. Li, and Z. Chen, “Frequency domain phase-resolved optical Doppler and Doppler variance tomography,” Opt. Commun. 242, 345–350 (2004). [CrossRef]  

49. D. C. Ghiglia and M. D. Pritt, Two Dimensional Phase Unwrapping, Theory, Algorithms, and Software (Wiley, 1998).

50. D. C. Ghiglia, G. A. Mastin, and L. A. Romero, “Cellular-automata method for phase unwrapping,” J. Opt. Soc. Am. A 4, 267–280 (1987). [CrossRef]  

51. N. Hu and E. B. Clark, “Hemodynamics of the stage 12 to stage 29 chick embryo,” Circ. Res. 65, 1665–1670 (1989). [PubMed]  

52. J. T. Butcher, T. C. McQuinn, D. Sedmera, D. Turner, and R. R. Markwald, “Transitions in early embryonic atrioventricular valvular function correspond with changes in cushion biomechanics that are predictable by tissue composition,” Circ. Res. 100, 1503–1511 (2007). [CrossRef]   [PubMed]  

53. K. K. Linask and M. Vanauker, “A role for the cytoskeleton in heart looping,” ScientificWorldJournal 7, 280–298 (2007). [CrossRef]   [PubMed]  

54. T. G. van Leeuwen, M. D. Kulkarni, S. Yazdanfar, A. M. Rollins, and J. A. Izatt, “High-flow-velocity and shear-rate imaging by use of color Doppler optical coherence tomography,” Opt. Lett. 24, 1584–1586 (1999). [CrossRef]