Image Processing and Analysis for Life Scientists

Basic image analysis for life scientists with a non-engineering background. The main goal is to teach how to address and solve scientific questions by state of the art image analysis strategies.

Description

Nowadays, image-based methods are indispensable for life scientists. Light microscopy especially, has evolved from sketched out observations by eye, to high throughput multi-plane, multi-channel, multi-position and multimode acquisitions that easily produce thousands of information-rich images that must be quantified somehow to answer biological questions.
 
This course will teach you core concepts from image acquisition to image filtering and segmentation, to help you tackle simple image analysis workflows on your own. All examples use open source solutions, in order to allow you to be independent from commercial solutions. Emphasis is made on good practices and typical pitfalls in image analysis. At the end of this course, you will be able to adapt and reuse workflows to suit your specific needs and be equipped with the tools and knowledge to adapt and seek advice from the ever-growing image analyst community of which you will be a part now
 
The course is taught by senior image analysts with longtime work experience in a service-oriented core facility.

What you will learn

• Recall digital image formation principles


• Understand human perception and color


• Distinguish between bit-depths


• Use lookup tables


• Perform mathematical operations on images


• Apply filtering to digital images


• Understand and use image segmentation techniques


• Create regions of interest and extract results from segmented images


• How to perform projections and reslicing on images for analysis


• Applying color deconvolution to brightfield images


• Understand the concepts of the ImageJ Macro language

Prerequisites

None.

Plan

Week 1: Digital Images
Introduction to digital image formation and how optical systems go from objects to images.
 
Week 2: Colors
Review of human visual perception and the RGB color model. Introduction to the concepts of image bit-depth and lookup tables.
 
Week3: Operating on Images
Introduction to image scaling, interpolation, and mathematical operations of images, and why certain bit-depths are more suitable than others.
 
Week4: Filtering
Using image filtering to enhance or suppress features in an image for easing subsequent analysis. We cover linear, nonlinear and Fourier filtering with emphasis on examples.
 
Week 5: Image Segmentation
Introduction to image segmentation and overview of available methods (thresholding, clustering, machine learning) and morphological operations.
 
Week 6: Regions of Interest
Going from analyzed objects to regions of interest and results tables. Emphasis is made on how to best obtain unbiased measurements and produce a reusable image analysis workflows
 
Week 7: Colors, and dimensionality reduction
Introduction to color models, and color deconvolution. Overview of the concept of dimensionality reduction through image projections and reslicing and application to measuring moving objects.
 
Extra Week: ImageJ Macro Programming Prime
Presentation of basic programming principles applied to the ImageJ Macro Language. Crash course on variables arrays, loops, conditionals, available macro functions and writing custom functions.