This is basic work on Image segmentation in MATLAB using OTSU method. In order to run the code, you have download the exp_data_2
folder and copy the directory address or the path in the code. The code should be able to run in any versions of MATLAB but to be specific, I ran this code in MATLAB2023b
version. You must also install the Image Processing Toolbox
from the MATLAB app store. Sample results are given in file names pos3 - new exp 2
to pos15 - new exp 2
. To find which position in the microfluidic device has healthy cells, check the microsoft word file 2021_03_13 2iL.doc
in exp_data_2/ON_2ILIF
.
Now let's dive into the workflow or the function of this code:
- The main code to run is named
MAIN_OFFLINE_run.m
- The aim to study from the image is based on clusted cell and not tracking single cell. So, the
tracking_flag = 0;
will always be same throughout the runtime of this code. - The channel numbers represent the microscope channel input but the number is
-1
from the microscope. - The GFP is the studied fluorescence here. All the calculated results can be found in
outputArg1.mat
file from the MATLAB workspace. - In this particular experiment, even though Sulphorodamine channel is present but no Sulphorodamine was used in the experiment.
Now let's explore the OFFLINE_function.m
function:
- The parameters you have entered in the MAIN section are passed here in this function.
- To note: we are not using anti-masking here, so we always select our background as Dial-A-Wave junction or any other junction available in the experiment channel 1.
- The Phase Contrast part of this code has parameters exported from the Image Segmentation app where the Cell shapes are identified as circles, dilated and eroded accordingly.
cb = 3
is the caliberation phase which is usually 3hrs of the experiment.
This code closely follows the original ChipSeg code from here.