Fundamentos de Análise de Imagens de Microscopia em Python

Fundamentals of microscopy image analysis in Python

About

This repository contains materials for the course "Fundamentos de Análise de Imagens de Microscopia em Python" taught by Prof. Alexandre Bruni-Cardoso for graduate students at the University of São Paulo (USP).

The course is designed for graduate students who want to learn computational tools for microscopy image analysis, with a focus on Python. It covers fundamental concepts, image processing, segmentation, and data visualization.

Course Objectives

To enable students to use computational tools, focusing on Python, for microscopy image analysis, addressing fundamental concepts, image processing, segmentation, and data visualization.

Prerequisites

• Personal laptop with minimum specifications: 4GB RAM, 10GB free storage, modern processor (e.g., Intel i5 or better), and internet connection
• Active Google account (for Google Colab)
• Pre-installation of FIJI/ImageJ
• Preferably, students should already use or plan to use bioimage analysis in their projects

Course Schedule

Day 1: Basic Concepts and Introduction to Python for Images

• Practical introduction based on active learning concepts
• Representing images as numbers: manipulation with NumPy and Scikit-image in Google Colab
• Image structure and characteristics (pixels, shape, channels)
• Python as a calculation tool, variables, functions, and packages
• Extracting numerical information: generating tables and histograms
• Image processing, segmentation, and measurements of morphology and fluorescence intensity

Materials: (notebooks day1_1 to day2_4 can be run directly in Google Colab. The remaining notebooks should be run locally after the installation of packages listed in bioimage_analyst_env.yml and bia_bob_environment.yml or in Google Colab after adjusting (Napari won't run in Colab))

• day1_1_Prepping_1_Google_Colab.pptx - Introduction to Google Colab
• day1_1_instant_gratification.ipynb - Quick start with Python for image analysis
• day1_2_course_overview.pptx - Course structure and expectations
• day1_2_python_basics_operations_variables.ipynb - Python fundamentals
• day1_3_pixels_numbers_LUT.ipynb - Understanding image data
• day1_4_functions_and_packages.ipynb - Working with Python functions and libraries
• day1_bioimage_analysis_workflow.excalidraw - Visual guide to the bioimage analysis process
• day1_3_Prepping_1_Google_Colab_Mounting_drive.pptx - Connecting Colab to Google Drive
• day1_5_image_processing_segmentation_quantification.ipynb - Basic image analysis workflow

Day 2: Deep Learning Tools for Segmentation and Data Visualization

• Introduction to Deep Learning applied to cell segmentation: using Stardist and Cellpose
• Setting up local Python environments with Miniconda
• Automating segmentation and image analysis
• Plotting simple graphs (fluorescence, morphology) with Matplotlib
• Introduction to Superplots with examples from scientific publications

Materials:

• day2_deep_learning_very_basic.excalidraw - Visual introduction to deep learning concepts
• day2_1_segmentation_with_DeepLearning_based_tools.ipynb - Using Cellpose and Stardist
• day2_2_best_practices_notebook.ipynb - Recommended practices for scientific computing
• day2_1_installing_miniconda_and_environment_management.pptx - Setting up local environments
• day2_3_loops_python.ipynb - Iteration and automation in Python
• day2_4_automation_segmentation_image_to_numbers.ipynb - Batch processing of images
• day2_5_superplots-tutorial.ipynb - Advanced data visualization for publication

Day 3: Cellpose Training and AI Tools

• Training custom models with Cellpose
• Exploring AI tools (LLMs and chats) applied to bioimage analysis
• Responsibilities and best practices for using AI to generate code
• Implementing tools like Gemini and Agentic AI for bioimage analysis workflows
• Final project

Materials:

• day3_1_cellpose_gui_train_your_own_model.pptx - Cellpose model training guide
• day3_1_run_your_own_cellpose_model.ipynb - Hands-on custom model training
• day3_2_AI_assisted_programming.pptx - Introduction to AI-assisted code generation
• day3_2_BiaBob_AI_assistant_for_bioimage_analysis.ipynb - Working with specialized AI tools
• day3_3_exercise_protein_translocation.pptx - Practical exercise
• day3_4_Wrapping_up.pptx - Course summary and next steps

Data Files

The repository includes various sample images for practical exercises:

• hela-cells-8bit.tif
• blobs.tif
• data_for_superplot_JCB_paper.csv

The images used in the exercises are downloaded directly from the internet, form imagej sample images ( https://imagej.net/ij/images/ ) or from the public repository at https://bbbc.broadinstitute.org/

Environment Files for

Windows:

• bia_bob_environment.yml - Environment configuration for BiaBob AI assistant
• bioimage_analyst_env.yml - General environment for bioimage analysis

Mac:

• bia_bob_macb.yml - Environment configuration for BiaBob AI assistant
• bioimage_analyst.yml- General environment for bioimage analysis

Methodology

The course combines lectures with practical activities in Jupyter Notebooks and exercises based on real cases.

Evaluation

Completion of practical exercises and submission of a final project.

Usage

1. Clone this repository or download the files
2. Set up your environment following instructions in day2_1_installing_miniconda_and_environment_management.pptx
3. Open notebooks in Jupyter or Google Colab
4. Follow the course materials in the order listed above

Contact

For questions about this course ant the material, please contact Prof. Alexandre Bruni Cardoso at brunicar@iq.usp.br

Citation

Bruni-Cardoso, A. (2025). Teaching material for microscopy image (bioimage) analysis with Python (QBQ5915 - 2025). Zenodo. https://doi.org/10.5281/zenodo.15089565

Most of the material in the course are explored in video tutorials in the YouTube channel

---

BSD 3-Clause License Copyright (c) 2024, Alexandre Bruni-Cardoso