Machine-learning-based image analysis with ilastik

Date: June 17th 2024

Time: 9am - 5pm

Location: HMS, Longwood (In-person)

Cost: Free

About

Ilastik is a free, open-source, and truly user-friendly image analysis tool that can help you leverage machine learning algorithms to easily segment, classify, track, and count your cells or other experimental data. Most operations are interactive, even on large, multi-dimensional datasets: you just draw the labels and immediately see the result. No machine learning expertise required. In this course you will learn how to perform automated pixel- and object-level segmentation, object classification, and tracking.

Participation will be limited to 20 people. The format will be a combination of tutorials and hands-on exercises.

Selection: This is not a first-come first-serve event; we evaluate all applications after the deadline. In case of over-subscription and multiple applicants from the same labs or departments, we will attempt to distribute the available seats evenly between the labs and departments.

Note: This is a one-off workshop at HMS by Dominik Kutra Senior Research Software Engineer in the Kreshuk Group at EMBL, Heidelberg, Germany.

Requirements

You should be curious about machine-learning based image analysis and bring a laptop with an external computer mouse. It would be ideal to install the latest beta version of ilastik before the start of the course: download.

Scope

• Morning:
  • Introduction to machine-learning based image analysis
  • Introduction to ilastik
  • ilastik workflows:
    • Pixel Classification
    • Autocontext
    • Neural Networks Prediction
• Afternoon:
  • ilastik workflows:
    • Trainable Domain Adaptation
    • Multicut
    • Object Classification
    • Tracking
  • Automation:
    • Batch processing
    • Fiji integration
    • Command line usage
    • Jupyter notebook usage

To stay updated with our upcoming workshops and seminars, subcribe to our email list.