ImageRefiner

Image Refinement for medical images using MATLAB

Aim of this software

This software was written to improve the MATLAB script published in "Gli1+ Pericyte Loss Induces Capillary Rarefaction and Proximal Tubular Injury" by Kramann, R. et al.

Set up Guide

Clone Git-Repository to a place of your choosing on your hard drive
Change Variables IMPORT_PATH and EXPORT_PATH to the appropriate values for your setup inside Settings.m
Inside MATLAB add repository folder to path (with subfolders)
The ImageRefiner requires a file ImageRefinerLog.log inside the folder target in the ImageRefiner base folder in order to log Messages and programm messages used for debugging.

Working with ImageRefiner

create a instance of the ImageRefiner: imageRefinerTask = Tasks.ImageRefinerTask('C:\\path\to\folder\with\data', 'ExportDirectory', 'C:\\path\to\export\folder', 'ExportFormat', 'excel', 'ColumnsToExport', {'Area', 'Perimeter', 'Centroid'}, 'VerificationMode', false) Please make sure to replace the folder paths to fit your local setup. You can omit or add all the name value pairs in the call to Task.ImageRefinerTask.
run the refinement process: imageRefinerTask.execute()

Please use the task to run the image refinement process instead of directly calling the different parts of the refiner as the task also includes functionality to export the generated results.

A full list of supported Settings which may be used during the analysis process are shown below:

Parameters	Description	Type	Default Value
ImportDirectory	Path where data is located	char	none
ImageExtension	Image Extension - please add '.'	char	'.tif'
ExportDirectory	Path to directory where exports should export to	char	target directory in git repository
ExportFormat	what format should the exports have	char	'excel'
ColumnsToExport	which columns need to be included in export	cell	{'Centroid', 'Area', 'Perimeter', 'MinFeretDiameter'}
VerificationMode	perform verification on all images in directory	logical	false
ColorChannel	Ending of split color channels	char	'CH2'
AreaBound	Upper/lower limit of the detected areas [10^-6 m^2]	double	[4.7, 100]
Connectivity	Connectivity parameter for analysis	double	8
PixelSizeOfArtifacts	Pixel size used to identify artifacts	double	15
ThresholdBlackWhiteImage	Threshold for black/white images	double	50

Data

The supplied data needs to adhere to this scheme:

  data
  |----191_19
  |       |----Cortex
  |       |       |----191-19_GF01
  |       |       |       |-----<Name>_CH1.tif
  |       |       |       |-----<Name>_CH2.tif
  |       |       |       |-----<Name>_CH3.tif
  |       |       |       |-----<Name>_CH4.tif
  |       |       |----191-19_GF02
  |       |       |       |-----<Name>_CH1.tif
  |       |       |       |-----<Name>_CH2.tif
  |       |       |       |-----<Name>_CH3.tif
  |       |       |       |-----<Name>_CH4.tif
  |       |----Medulla
  |       |       |----191-19_GF01
  |       |       |       |-----<Name>_CH1.tif
  |       |       |       |-----<Name>_CH2.tif
  |       |       |       |-----<Name>_CH3.tif
  |       |       |       |-----<Name>_CH4.tif
  ...

If you do not stick to this scheme the code will not work correctly!

Developer

Developed by Robert Goellinger, RWTH Aachen University. Issues or requests can be submitted by opening a new issue on github. Please make sure to include your log-file for analysis of the underlying problems.

Cite as:

Robert Goellinger (2021), ImageRefiner - A Matlab Toolbox for refining nephrological images, RWTH Aachen University

Resources

Winslow, L.; log4m - A powerful and simple logger for matlab (https://www.mathworks.com/matlabcentral/fileexchange/37701-log4m-a-powerful-and-simple-logger-for-matlab), MATLAB Central File Exchange. Retrieved January 7, 2021.
Kramann, R.; Wongboonsin, J.; Chang-Panesso, M.; Machado, F.; Humphreys, B.; Gli1+ Pericyte Loss Induces Capillary Rarefaction and Proximal Tubular Injury; Journal of the American Society of Nephrology;

RobertGoellinger/ImageRefiner