Home
Atlas
Data
Project
Technical
Links
Contact Info
Specimen Selection and Preparation

Whole human embryo specimens from Carnegie stages 13 to 23 were selected from the Carnegie Collection of Human Embryos at the National Museum of Health and Medicine, in the Armed Forces Institute of Pathology (AFIP). These stages approximately represent 32 to 56 days post-ovulation. Formalin-fixed specimens without signs of mechanical damage and without obvious signs of malformations were selected whenever possible. In some cases, embryos with suspected malformations were imaged because more suitable specimens were not available to represent those stages of development. All embryos were photographed with a digital camera on an optical microscope prior to magnetic resonance microscopy (MRM) .

The larger and well-fixed embryos (over stage 17) were immersed in Fomblin LC08 (Ausimont USA, Inc., Thorofare, NJ) during MRM imaging to provide a background without signal. Smaller and poorly-fixed specimens were imaged in formalin because they would not have tolerated the density of the Fomblin medium. However, formalin does not provide as much contrast between the embryo and the background medium as the Fomblin does. All embryos were placed in airtight plastic containers and inserted into an imaging coil. Four MR imaging coils were built to accommodate the 10-fold size differences of these embryos: a Helm-Holz coil for the smallest embryos, a solenoid coil for the next sizes, and then two bird-cage coils for the larger specimens.

MRM Imaging

All embryos were imaged in a 9.4 Tesla, super-conducting magnet at Duke University’s Center for In-vivo Microscopy. Each embryo was imaged with predominantly T1-weighted, T2-weighted, and diffusion-weighted pulse sequences to provide 3 perfectly registered three-dimensional data sets. T1-weighted imaging was achieved with a three-dimensional spin-echo GRASS sequence, TR = 95 ms, TE = 3.5 ms, flip angle alpha = 60 degrees, and NEX = 2. T2-weighted imaging was obtained with a three-dimensional spin-echo sequence, TR = 1,000 ms, TE = 50 ms, alpha = 90 degrees, and NEX = 1. Diffusion-weighted imaging was obtained with a three-dimensional spin-echo sequence, TR = 1,000 ms, TE = 15 ms, alpha = 90 degrees, NEX = 1, and diffusion time of 3 ms in the long axis of the embryo. Other details of the pulse sequences can be found below. Images were acquired as 128 sagittal slices with 256 x 256 pixels per slice. The field of view was set to produce isotropic voxels (three-dimensional pixel elements with equal dimensions in the x, y, and z axis). This aspect ratio matched the embryo dimensions that were very close to 2 x 2 x 1 for the anterior-posterior, superior-inferior, and left-right axis, respectively. Image resolution ranged from 39 microns per pixel to 156 microns per pixel, depending on the size of the embryo (see table below).


Image Processing

The embryos were separated from background noise in the sagittal images using thresholding and seeding (the "wand" tool) and masking tools in Adobe Photoshop. These same tools were used to produce the segmented data sets containing single organ systems. The segmented (organ) data sets maintain the original slice numbering system and therefor do not always begin with slice #1 if the organ was not found in slice #1. Coronal and axial images were obtained from the cleaned sagittal images using VoxelView_Ultra 2.5 software (Vital Images, Fairfield, IA) running on a Silicon Graphics workstation. QuickTime animations were created using volume rendering in VoxelView_Ultra, and psuedo-timelapse animations were created using the Gryphon Morph 2.5 software.

 
Magnetic Resonance Imaging Parameters
Carnegie Stage Post Ov. Days (approx.) Scan # TR TE Flip Angle Scan Type Coil Resolution* NEX
13 28 11634 95 3.5 60 T1 HelmHolz 39.1 2
11635 1000 15 90 Diffusion HelmHolz 39.1 1
11636 1000 50 90 T2 HelmHolz 39.1 1
14 32 11627 95 3.5 60 T1 HelmHolz 39.1 2
11628 1000 15 90 Diffusion HelmHolz 39.1 1
11629 1000 50 90 T2 HelmHolz 39.1 1
15 33 11537 95 3.5 60 T1 Solenoid 54.7 2
11538 1000 15 90 Diffusion Solenoid 54.7 1
11539 1000 50 90 T2 Solenoid 54.7 1
16 37 11894 95 3.5 60 T1 Solenoid 62.5 2
11897 1000 15 90 Diffusion Solenoid 62.5 1
11898 1000 50 90 T2 Solenoid 62.5 1
17 41 11422 95 3.5 60 T1 Solenoid 46.9 2
11424 1000 15 90 Diffusion Solenoid 46.9 1
11423 1000 50 90 T2 Solenoid 46.9 1
18 44 11430 95 3.5 60 T1 Bird Cage 78.1 2
11431 1000 15 90 Diffusion Bird Cage 78.1 1
11432 1000 50 90 T2 Bird Cage 78.1 1
19 47 11239 95 3.5 60 T1 Bird Cage 70.3 2
11237 1000 15 90 Diffusion Bird Cage 70.3 1
11238 1000 50 90 T2 Bird Cage 70.3 1
20 50 11906 95 3.5 60 T1 Bird Cage 101.6 2
11907 1000 15 90 Diffusion Bird Cage 101.6 1
11908 1000 50 90 T2 Bird Cage 101.6 1
21 52 Not Available
22 54 11525 95 3.5 60 T1 Bird Cage 140.6 2
11526 1000 15 90 Diffusion Bird Cage 140.6 1
11527 1000 50 90 T2 Bird Cage 140.6 1
23 56 11616 95 3.5 60 T1 Bird Cage 156.3 2
11617 1000 15 90 Diffusion Bird Cage 156.3 1
11618 1000 50 90 T2 Bird Cage 156.3 1
* Resolution is reported in microns per voxel (microns per cubic pixel) with x, y, and z dimensions all being equal.
© 2001 Bradley R. Smith