Light sheet microscopy

nicholasreder edited this page Jan 5, 2017 · 2 revisions

Vision

Advances in microscopic imaging enable visualization of heretofore unseen 3D microanatomical features, which have the potential to transform cancer diagnostics. These novel imaging techniques have led to improved diagnostics in kidney biopsies, brain structure, and embryonic development. Light-sheet microscopy and tissue clarification techniques are a particularly intriguing combination because large volumes of cancer tissue can be imaged with high spatial resolution. However, data processing, data management, and visualization of 3D structures has lagged behind the advances in data acquisition. Currently, our data processing steps require 12-24 hours of computing time and are accomplished using fragments of code written in matlab and Miji (matlab + Fiji). In addition, our existing software code is stored on multiple servers and is not well annotated, limiting reproducibility of our work. Finally, the visualization of 3D structures is suboptimal and hinders our ability to provide diagnostic insights. Thus, our current data science limitations have impeded potentially groundbreaking discoveries in cancer microanatomy.

To solve this problem, we will create a software package to optimize the processing, storage, and visualization of 3D microscopic data. Key morphologic features will be extracted from the images and displayed in a simple, easy to navigate format. All code will be well annotated and stored on GitHub to ensure reproducibility.

Objectives

Aim 1: Efficiently stitch a mosaic image from 1D TIFF image strips acquired by light-sheet microscopy. We will use parallel processing in the cloud and write code to speed data processing time and reduce stitching artifacts. Aim 2: Efficiently store large (300 gb) datasets generated by light-sheet microscope. Uninformative regions of the image will be cropped to reduce file size and datasets will be compressed in a lossless format that can be easily accessed and queried. Aim 3: Provide elegant, easy-to-interpret visualizations of 3D tubular prostate structures. The prostate gland is analogous to a branching tree, with tubular glands (akin to tree branches) lined by cell nuclei. We will produce software that efficiently processes and segments the image, detecting cell nuclei and defining the parts of the image that belong to the lumen of prostate gland tubules. The software will allow researchers and clinicians to explore the three-dimensional branching structure of the glands.

Deliverable Schedule

Current Pipeline: (1) Read dcimg files into matlab, (2) resample data and store as tiff z-stack, (3) stitch adjacent tiff strips into full area 2D tiff, (4) reconstruct stitched tiffs in 3D (imaris).

Week of 1/9: Place existing code into github, read dcimg files into Python Week of 1/16: Store dcimg files as intensity + transformation structure. Subgoal: perform same transformation on tiff files Weeks of 1/23 - 2/6: Stitch together adjacent 3D volume strips (stored as intensity + transformation). Investigate cluster solutions. Weeks of 2/13 - 2/20: Focus on making code/processing more efficient Weeks of 2/27 - 3/6: Optimize visualization of 3D images using existing software tools

You can’t perform that action at this time.
You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session.
Press h to open a hovercard with more details.