SubjectsSubjects(version: 996)
Course, academic year 2026/2027
   
4EU+ Advanced image analysis with focus on - ImageJ, Arivis Vision 4D, SVI Huygens - MB100T01
Title: 4EU+ Advanced image analysis with focus on - ImageJ, Arivis Vision 4D, SVI Huygens
Czech title: 4EU+ Pokročilá analýza obrazu se zaměřením na software - ImageJ, Arivis Vision 4D, SVI Huygens
Form of teaching: block practicals
Guaranteed by: Biology Section (31-101)
Faculty: Faculty of Science
Actual: from 2025
Duration in semesters: 1
Semester: winter
E-Credits: 4
Examination process: winter s.:
Hours per week, examination: winter s.:1/3, C+Ex [DS]
Capacity: 20
Maximum number of enrolled students: 20
Min. number of students: unlimited
4EU+: yes
Virtual mobility / capacity: no
State of the course: taught
Language: English
Note: enabled for web enrollment
the course is taught as cyclical
Guarantor: Mgr. Zuzana Burdíková, Ph.D.
Teacher(s): Mgr. Zuzana Burdíková, Ph.D.
Ing. Martin Schätz, Ph.D.
Mgr. Zdeněk Švindrych, Ph.D.
Annotation -
This course provides a comprehensive introduction to advanced bioimage analysis, focusing on key software tools such as ImageJ, MIB, Arivis Vision 4D, and SVI Huygens. Through theoretical sessions and hands-on exercises, participants will learn fundamental and advanced concepts in image analysis, including image formation, segmentation, object detection, colocalization assessment, and deconvolution. Special emphasis is placed on the integration of artificial intelligence and deep learning tools for bioimage processing.
Last update: Sacherová Veronika, RNDr., Ph.D. (19.02.2025)
Literature - Czech

Základní literatura / Core Literature

  1. Russ, J. C. The Image Processing Handbook. 7th ed. CRC Press, 2016.
    (Základní principy digitálního zpracování obrazu, segmentace a měření.)

  2. Burger, W., Burge, M. J. Principles of Digital Image Processing. Springer, 2009.
    (Základy digitálních obrazů, rozlišení, vzorkování a artefakty.)

  3. Eliceiri, K. W., et al. Biological imaging software tools. Nature Methods, 2012.
    (Přehled nástrojů pro analýzu bioobrazů.)

  4. Culley, S. et al. Made to measure: an introduction to quantification in microscopy data. 2023.
    (Principy kvantitativní analýzy mikroskopických dat – intenzita, morfologie a počítání objektů.)

Doporučená literatura / Recommended Literature

  1. Arganda-Carreras, I. et al. Trainable Weka Segmentation: a machine learning tool for microscopy pixel classification. Bioinformatics, 2017.
    (Strojové učení pro segmentaci mikroskopických obrazů.)

  2. Harrington, K. State of the Art for Machine Learning in Bioimage Analysis. Microscopy and Microanalysis, 2023.
    (Moderní přístupy využívající strojové učení a hluboké učení v analýze bioobrazů.)

  3. McCann, M. T., Unser, M. Biomedical Image Reconstruction: From the Foundations to Deep Neural Networks. 2019.
    (Principy rekonstrukce obrazů a moderní metody včetně hlubokého učení.)

  4. Akçakaya, M. et al. Unsupervised Deep Learning Methods for Biological Image Reconstruction and Enhancement. 2021.
    (Hluboké učení pro rekonstrukci a zlepšování mikroskopických dat.)

Online zdroje / Online Resources

  1. ImageJ/FIJI Documentation
    https://imagej.net
    (Uživatelské návody zahrnují segmentaci, kolokalizaci, sledování objektů a skriptování.)

  2. Introduction to Bioimage Analysis with Fiji/ImageJ
    https://hms-iac.github.io/fiji-workshop
    (Úvod do kvantitativní analýzy bioobrazů a reprodukovatelných pracovních postupů.)

  3. Bioimage Analysis with Fiji/ImageJ Tutorials
    https://www.microlist.org
    (Zpracování obrazů, segmentace, kvantifikace a kolokalizace

Last update: Burdíková Zuzana, Mgr., Ph.D. (25.02.2026)
Course assessment methods and requirements for successful completion, grading scheme -

For the final assessment, participants will prepare an individual project that demonstrates their ability to design and describe a bioimage analysis workflow.

The project will contain several parts.

  1. Definition of a research question – Clearly state the biological problem to be analyzed.
  2. Pick up a dataset and software - Choose from provided datasets or use your own.
  3. Design a step-by-step workflow for image of image analysis to solve the given task.
Last update: Sacherová Veronika, RNDr., Ph.D. (19.02.2025)
Syllabus -

Day 1 - Fundamentals of Bioimage Analysis

Introduction to Bioimage analysis

●     What is Bioimage analysis?

●     Key features and applications

Basic concepts

●     Pixels/voxels, bit depth, dynamic range

●     Resolution, sampling, and image artifacts

Overview of software tools for Bioimage analysis

Image processing fundamentals

●     Common steps in object segmentation

●     Classical machine learning approaches: Trainable Weka Segmentation

●     Object detection methods: Watershed and Connected Components Analysis

Hands-on case study: Reproducibility in Bioimage analysis

●     Comparing manual vs. automated object counting

Introduction to AI in Bioimage analysis

Day 2: Processing Large-Scale Microscopy Datasets

Pre-processing of (not only) volume electron microscopy (VEM) data

●     Stitching and stack alignment using FIJI's TrackEM

Introduction to Microscopy Image Browser (MIB)

●     Image I/O, bit depth conversion, filtering, alignment, and stitching

Segmentation strategies in MIB

●     Part I: Layers, superpixel clustering, and GraphCut segmentation

●     Part II: Deep learning tools (DeepMIB, Segment Anything Model)

Day 3: Bioimage Analysis workflows with ArivisPro

First steps with ArivisPro

Handling large image datasets

●     Tile sorting and stitching

Lightsheet Microscopy: Introduction and applications

3D segmentation workflows

●     Machine learning-based segmentation

●     Blobs finder pipeline

●     Magic Wand tool for segmentation

3D Object Tracking & Visualization

Colocalization analysis in ArivisPro (theory & hands-on)

Day 4: Image deconvolution & Data visualization

Principles of Image Deconvolution

●     Theory behind deconvolution

●     Hands-on practice using SVI Huygens

Data Visualization & Reporting

●     Best practices for presenting results

●     Hands-on session using Python (Pandas, Matplotlib, Seaborn) and Google Colab

Day 5: Emerging Trends in Bioimage Analysis

Bioimage analysis communities & collaboration

Deep learning in Bioimage analysis

●     Introduction to AI-driven tools

●     Noise2Void for image denoising

●     StarDist for star-convex object segmentation

●     Hands-on practice

Last update: Sacherová Veronika, RNDr., Ph.D. (19.02.2025)
Learning outcomes -

After completing the course, the student will be able to:

  1. Explain the basic principles of bioimage analysis, including key concepts such as pixels, voxels, bit depth, dynamic range, resolution, and sampling.

  2. Demonstrate orientation in commonly used bioimage analysis software tools and select appropriate tools for specific microscopy datasets.

  3. Perform basic image-processing steps, including object segmentation, filtering, and object detection using classical algorithms.

  4. Apply machine-learning approaches for image segmentation, including trainable Weka segmentation and deep-learning-based tools.

  5. Process large-scale microscopy datasets, including stitching and alignment of image stacks.

  6. Perform segmentation and quantitative analysis of 2D and 3D image data using specialized software tools.

  7. Use tools for 3D visualization and object tracking in microscopy datasets.

  8. Explain the principles of image deconvolution and apply deconvolution methods to microscopy data.

  9. Visualize and present bioimage analysis results using appropriate statistical and graphical tools.

  10. Use modern AI-based approaches for microscopy data analysis, such as image denoising and object segmentation methods.

  11. Design reproducible bioimage analysis workflows and critically evaluate the quality of the results.

Last update: Burdíková Zuzana, Mgr., Ph.D. (25.02.2026)
The course does not include work placement
 
Charles University | Information system of Charles University | http://www.cuni.cz/UKEN-329.html