BioInfoSummer: Image Analysis with Fiji/ImageJ

Example 1: Fluorescent stain measurement

Fluorescent wide field image of three fluorescent markers/stains:

  • Blue: DAPI nuclear marker
  • Red: Marker for wounding
  • Green: Cytoplasmic marker for healthy cells.


Aim

Analyse the area covered and/or intensity of marker for wounding.

1. Open "[...]/Images/Widefield/Fluorescence Measurement/Fluoro 01.tif"


2. Calibrate the image:

Image → Properties
Unit of length: µm
Pixel width: 0.6
Pixel height: 0.6
Voxel depth: 1

Calibration are usually encoded in the image metadata in modern microscopes


3. Split channels so we can do measurements on a specific channel.

Image → Colour → Split Channels


Filtering

4. Apply a Gaussian Blur with a sigma of 1 px to reduce noise.

Process → Filters → Gaussian Blur...

A Gaussian blur filter helps to smooth noise that is typically present in images acquired with highly sensitive camera sensors or detectors.


Segmentation

5. Threshold based segmentation:

Image → Adjust → Threshold...

Thresholding segments the images based on pixel intensity. Thresholds can be set manually or automatically via algorithms that examine the image histogram.


Measurement

6. Set measurement types:

Analyze → Set Measurements...
Tick: Area, Min & max grey value, Mean grey value, Area Fraction, Limit to Threshold, Display label


7. Measure:

Analyze → Measure

Here we are measuring the cumulative area of all stained parts. In the next example, we will utilise connected components to analyse the characteristics of individual segmented regions.