Skip to content

Image processing algorithm to identify structure of tumour spheroids with cell cycle labelling

License

Notifications You must be signed in to change notification settings

ap-browning/SpheroidsImageProcessing

Repository files navigation

SpheroidsImageProcessing

DOI

This repository contains codes used to batch-process confocal images of spheroids encoded with FUCCI.

Requirements

  1. MATLAB R2021a or later with the image processing toolbox
  2. FIJI.

Components

  1. MATLAB app AdjustRawImages.mlapp used to manually adjust contrast raw tif images and embed necessary metadata.
  2. Script ProcessImages.m used to perform the batch processing on a folder containing the output of component 1.
  3. Folder functions containing necessary subroutines.
  4. Folder Demo containing a demo script to demonstrate the algorothm.
  5. Folder DemoImages containing raw microscope images (oir), raw tif images and adjusted images,
  6. Document SpheroidsImageProcessing.pdf containing further details. Run Demo/Demo.m to reproduce Figure 1.

Pipeline

  1. Raw oir files are batch converted to tif using FIJI.
  2. The FIJI macro ExportInfoMacro.ijm is run on the output folder from step 2 to produce a txt file for each image containing the metadata to be extracted by MATLAB.
  3. The AdjustRawImages.mlapp is run (right click and select run) on the folder containing the raw tif and txt files. This will create two subfolders: Adjusted and Original. To demo, select DemoImages/Original in the prompt after running the app.
  4. The ProcessImages.m script is run on the Adjusted folder from step 3. To demo, select DemoImages/Adjusted in the prompt after running the scipt.
  5. After processing is complete, the images will display one-by-one along with an input prompt.
  6. To keep the processed image, hit y then Enter (or just Enter). To reject, hit n then Enter.
  7. Type a filename to save the data or press n. Press Enter.
  8. Press y to filter the data (i.e., remove rejected images), else press n. Press Enter.
  9. Type a folder name to save the processed images or press n. Press Enter.
  10. The script will always save the processed data (including rejected images) to a Processed-date-time.csv file in the directory above the Adjusted folder selected.

Authors

Alexander P Browning and Ryan J Murphy
School of Mathematical Sciences
Queensland University of Technology
Brisbane, Australia