On November 16, participants will begin with a hands-on reconstruction of a cryo-electron tomography (cryo-ET) dataset. We have several datasets available for this training, but I have decided to use a dataset deposited in the Electron Microscopy Data Bank (EMDB). This will allow participants not only to work with a real, publicly available dataset during the course, but also to access and practice with the same dataset after the workshop.

Following the reconstruction, we will move into 3D volume visualization and segmentation. This portion will introduce participants to several approaches for extracting and analyzing structural features from reconstructed volumes, including:

• AI/computational approaches to 3D image analysis and segmentation
• The Watershed algorithm for automated or semi-automated segmentation
• Manual segmentation and interactive volume annotation
• Segmentation and image analysis using established software packages such as ImageJ, Amira and Avizo (this is a paid software) 
• AI-based segmentation and analysis using Aivia (depending on the number of participants) 
• Comparison of different segmentation approaches and their applications to biological datasets

There will also be a scripting component integrated into the course. Depending on the number and interests of participants, we plan to include a dedicated segment demonstrating how scripts can be developed to automate portions of the reconstruction, image-processing, and segmentation workflows. Participants with experience in Python, C++, and/or Perl should find this particularly useful, as the scripting component will provide an opportunity to develop customized tools and workflows rather than relying entirely on graphical interfaces.

The second part of the day will focus on 3D reconstruction from cryo-FIB and Plasma FIB datasets, followed by visualization and segmentation of the resulting volumes. This will provide participants with exposure to large-volume datasets generated from FIB-based imaging and demonstrate how computational approaches can be applied to these datasets for structural and biological interpretation.

Virtual

# Participants will be introduced to the basics of LSFM and the computational hurdles working with LSFM datasets (that usually require special consideration due to unique acquisition geometries and large data size). 

# Sample datasets will be available to explore multiview registration and stitching with BigStitcher that utilizes imglib2 and the BigDataViewer. 

# Strategies for subsequent analysis (segmentation, tracking) on reconstructed volumes will also be presented.  

Opening/importing images;  Data formats, multidimensional data, multichannel images, RGB color, bit depth;

Brightness and Contrast, Channel tools and overlays, separating and merging channels,

Line profile, Spatial calibration, scale bars, making manual measurements

Thresholding; Dealing with uneven background (Background subtraction), image filtering)

Analyze Particles

Pixel Classifiers, Weka segmentation