Research Papers

Quantifying Function in the Early Embryonic Heart

[+] Author and Article Information
Brennan M. Johnson

School of Biomedical Engineering,
Colorado State University,
Fort Collins, CO 80523

Deborah M. Garrity

Department of Biology,
Colorado State University,
Fort Collins, CO 80523

Lakshmi Prasad Dasi

School of Biomedical Engineering,
Colorado State University,
Fort Collins, CO 80523
e-mail: lakshmi.dasi@colostate.edu

1Corresponding author. Present address: Department of Mechanical Engineering, Colorado State University, Room A103D Engineering, 1374 Campus Delivery, Fort Collins, CO 80523-1374.

Contributed by the Bioengineering Division of ASME for publication in the Journal of Biomechanical Engineering. Manuscript received September 12, 2012; final manuscript received January 31, 2013; accepted manuscript posted February 19, 2013; published online April 2, 2013. Assoc. Editor: Naomi Chesler.

J Biomech Eng 135(4), 041006 (Apr 02, 2013) (11 pages) Paper No: BIO-12-1402; doi: 10.1115/1.4023701 History: Received September 12, 2012; Revised January 31, 2013; Accepted February 19, 2013

Congenital heart defects arise during the early stages of development, and studies have linked abnormal blood flow and irregular cardiac function to improper cardiac morphogenesis. The embryonic zebrafish offers superb optical access for live imaging of heart development. Here, we build upon previously used techniques to develop a methodology for quantifying cardiac function in the embryonic zebrafish model. Imaging was performed using bright field microscopy at 1500 frames/s at 0.76 μm/pixel. Heart function was manipulated in a wild-type zebrafish at ∼55 h post fertilization (hpf). Blood velocity and luminal diameter were measured at the atrial inlet and atrioventricular junction (AVJ) by analyzing spatiotemporal plots. Control volume analysis was used to estimate the flow rate waveform, retrograde fractions, stroke volume, and cardiac output. The diameter and flow waveforms at the inlet and AVJ are highly repeatable between heart beats. We have developed a methodology for quantifying overall heart function, which can be applied to early stages of zebrafish development.

Copyright © 2013 by ASME
Your Session has timed out. Please sign back in to continue.



Grahic Jump Location
Fig. 1

Flow chart of methodology

Grahic Jump Location
Fig. 2

Zebrafish embryos and imaging equipment. (a) Zebrafish embryos next to a United States penny for scale. (b) Ventral view of embryonic zebrafish heart. Blood cells can easily be seen through the transparent tissues surrounding the heart. As blood flows through the heart, it proceeds through the inlet (i), atrium (a), atrioventricular junction (AVJ), and finally through the ventricle (v). Scale bar indicates 100 μm. (c) Experimental setup, consisting of bright field stereomicroscope, high speed camera, and computer.

Grahic Jump Location
Fig. 3

Image Processing. A comparison is made between the unprocessed data (top row) and the processed data (bottom row). For each frame, a sliding average of the intensity of the previous frames is subtracted from the current frame. This removes static portions of the image and creates a black background. Moving portions of the image are easily seen as gray or white pixels. The upper ST plot segment (top middle) reveals notable background artifacts, which result in multiple inaccurate variance peaks (top right). Background artifacts are greatly reduced in the processed images (bottom left and middle), resulting in significantly more accurate variance peaks (bottom right) and, consequently, more accurate velocity estimation.

Grahic Jump Location
Fig. 4

Creation of spatiotemporal plots. Part (a) shows a series of three frames with mock cells moving in the vertical direction. The vertical gray column marks the pixels that lie along an arbitrary reference line. The pixels along this line are plotted side by side to create the spatiotemporal plot in (b). As cells move along the reference line, their changing position is seen as an angled streak in the ST plot. The vertical axis of the ST plot represents position along the reference line, while the horizontal axis represents time. The velocity of the cells along the reference line can be determined by calculating the slope (dy/dt) of the streaks in the ST plot.

Grahic Jump Location
Fig. 5

ST plot analysis with Radon transforms. Part (a) shows a cartoon of an embryonic heart. ST plot reference lines are placed as shown at the atrial inlet and AVJ. A sample section of a typical ST plot is shown in (b). The ST plot must be analyzed in smaller sections, called bins. One such bin is outlined in (b). Part (c) shows the Radon transform analysis at 0 deg, 90 deg, and 138 deg. The projection of this bin is summed along each of these projection angles (θ) to produce a corresponding plot. When the projection angle matches the streak angle (138 degree sign (°) in this case), the resulting plot contains prominent peaks. A complete Radon transform contains an analysis of the ST plot bin from every angle, where Radon transform values are represented by intensity, as seen in (d). Part (e) shows a plot of the Radon transform variance for each angle. The projection angle with the highest variance is perpendicular to the streak angle in the ST plot. The velocity (pixels/frame) is then tan(θ–90 deg).

Grahic Jump Location
Fig. 6

Measurement of lumen diameter. ST plots were created from reference lines which spanned the width of the lumen at the atrial inlet (a) and AVJ. The lines in (b) mark the flow boundary used for measuring orifice diameter, ‘D’. Measurements were used to plot diameter versus time. Time was normalized such that one complete cardiac cycle spans from 0 to 1.

Grahic Jump Location
Fig. 7

Cardiac physiological parameters. Plots were created for orifice diameter (a), blood velocity (b), flow rate (c), and accumulated blood volume (d) versus time at each orifice. Results are shown for C2. Time was normalized such that one complete cardiac cycle spans from 0 to 1.

Grahic Jump Location
Fig. 10

Span size optimization. Span size was optimized at low‐moderate velocities (top) and peak velocities (bottom). Span size was varied and results were shown of (from left to right) the processed image, the resulting ST plot, the resulting Radon transform, and the variance of the derivative of the intensity of the Radon transform.

Grahic Jump Location
Fig. 9

Bin size optimization. Bin size was optimized based on an ST plot segment with changing velocity. The local ST plot streak angle was manually measured as a baseline. Bin size was varied and results were compared to the baseline. Results are shown (from left to right) of the ST plot bin, the resulting Radon transform, and the variance of the derivative of the intensity of the Radon transform.

Grahic Jump Location
Fig. 8

Changes in flow rate and volume at the atrial inlet and AVJ. Plots of the estimated flow rate versus time (a) and (c) and accumulated orifice volume versus time (b) and (d) for each experimental case. Time was normalized such that one complete cardiac cycle spans from 0 to 1.




Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In