3d datasets reconstruction software  (InstaRecon Inc)

Journal: iScience

Article Title: A method for investigating spatiotemporal growth patterns at cell and tissue levels during C-looping in the embryonic chick heart

doi: 10.1016/j.isci.2022.104600

Figure Lengend Snippet: A comprehensive pipeline for 3D multi-scale study of C-looping This figure shows an overview of the pipeline used in this study. For a full, comprehensive detail on every aspect in the methods please see . The workflow is composed of two main parts, the experimental workflow and the computational workflow. In the experimental workflow, fertilized chicken embryos were harvested at HH10 and HH11 gestation age. Whole-mount fluorescent staining was performed to obtain a 3D image of the whole heart with individual cells labeled. Embryos were then serially incubated in Glycerol for optical clearing of the whole tissue. Using a confocal microscope, entire heart area imaged on custom made chamber for mounting whole embryo. A following 3D reconstruction of confocal images resulted in a super-image of the whole chicken heart that includes information from cell level through to the whole organ. Next, we were interested to obtain information on the tissue to organism levels from the same heart samples. Upon the completion of confocal imaging, tissue samples were washed and stained to be imaged with a micro-CT scanner at sub-micron resolution to acquire a 3D image stack of the chicken embryos, capturing information from tissue level to whole organ. Ultimately, this experimental workflow resulted in two sets of 3D dataset: (1) data at cell, tissue, and organ levels, and (2) data at tissue, organ, and organism levels. These datasets were subsequently used in the computational workflow. To begin the computational analysis and modeling, a shape representation of the heart was needed. To define an anatomically realistic shape, we segmented the geometry of the heart using a custom-made, semi-automatic pipeline, in Amira software. The segmentation resulted in labeled masks from which digitized data points were sampled to generate a 3D point cloud of the chicken heart. Using the Finite Element Method (FEM), a template mesh was constructed using high-order shape functions to mathematically represent the anatomy of hearts. We morphed the template mesh by using fitting techniques to fit and customise mesh to the 3D point cloud representation of the heart. Anatomical landmarking and temporal ordering helped to capture the spatial and temporal dynamics of C-looping. We also acquired information at cell level. We developed a fully automatic algorithm using deep learning techniques (convolutional neural networks) to segment single cells from the entire tissue. The result was a comprehensive, high-resolution single cell scale map of the myocardium. Once we had the dataset with both whole geometry and cell information, we spatially aligned all samples to a reference coordinate system to remove any confounding transformation effects. Next, in Amira software, a number of important cell features were extracted for the analysis of cell shape, volume, and orientation. This information was required to map all cell features as fields onto the constructed FE model of the heart. This resulted in a spatiotemporal dataset of heart with embedded cells and cellular feature. A comprehensive analysis of cell features and feature variance revealed differential growth patterns during C-looping. To understand how C-looping happens at the tissue level, we analyzed the growth mechanism using kinematics modeling by computing the deformation gradient tensor to describe the deformation of tissue material points from the initial time point to the next time point. Using this deformation tensor, we also obtained volume changes by computing the Jacobian from the deformation gradient. Furthermore, by performing a singular value decomposition on the deformation tensor, we obtained the tissue stretch and orientation information. The previously extracted cell features were used to measure changes of the myocardial cell number, size, and shape, and orientation throughout the course of C-looping. The resulting datasets from both the tissue- and cell-level analyses were combined to investigate their correlations during growth, and to examine whether their relationship changes spatially and temporally. The main idea for this analysis is to understand how cell-level features affect the volume and orientation changes observed at the tissue level.

Article Snippet: The acquired 3D datasets were reconstructed using InstaRecon ®CBR software.

Techniques: Staining, Labeling, Incubation, Microscopy, Imaging, Micro-CT, Software, Construct, Transformation Assay

Journal: iScience

Article Title: A method for investigating spatiotemporal growth patterns at cell and tissue levels during C-looping in the embryonic chick heart

doi: 10.1016/j.isci.2022.104600

Figure Lengend Snippet:

Article Snippet: The acquired 3D datasets were reconstructed using InstaRecon ®CBR software.

Techniques: Recombinant, Software