Confocal Microscopy List

Phase contrast image segmentation

Classic

List

Threaded

6 messages Options

Zdenek Svindrych-2

Phase contrast image segmentation

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Dear Listers,

not exactly a confocal question, but here it goes: Do you know of a simple,
reliable, or even barely useable way to segment transmitted light images of
yeast or bacteria, such as this one?:
https://drive.google.com/file/d/15ke38DZsADu8ahL9m1WzBEiIxIsBzj8O
Preferably in ImageJ, but I'm happy to try other platforms.

Thanks!

Best, zdenek
--
--
Zdenek Svindrych, Ph.D.
Research Scientist - Microscopy Imaging Specialist
Department of Biochemistry and Cell Biology
Geisel School of Medicine at Dartmouth

haesleinhuepf

Re: Phase contrast image segmentation

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Dear Zdenek,

I recently gave an online lecture introducing students to basic ImageJ Macro where we process very similar images:

https://youtu.be/q4ISEWq6q4E

Also check out image.sc for alternate approaches:

https://forum.image.sc/t/cell-segmentation-in-phase-contrast/943/14<https://forum.image.sc/t/cell-segmentation-in-phase-contrast/943/14?u=haesleinhuepf>

I'm sure there are more sophisticated plugins and tools for processing that kind of data, especially on python side, but maybe that's a starting point.

Cheers,
Robert

--
Robert Haase, Dr. rer. medic.
Post-doc, Myers Lab

Center for Systems Biology Dresden
Max Planck Institute of Molecular Cell Biology and Genetics
Pfotenhauerstr. 108
01307 Dresden

________________________________
From: Zdenek Svindrych <[hidden email]> on behalf of Zdenek Svindrych <[hidden email]>
Sent: Tuesday, 15 September 2020, 23:35
To: [hidden email]
Subject: Phase contrast image segmentation

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Dear Listers,

not exactly a confocal question, but here it goes: Do you know of a simple,
reliable, or even barely useable way to segment transmitted light images of
yeast or bacteria, such as this one?:
https://drive.google.com/file/d/15ke38DZsADu8ahL9m1WzBEiIxIsBzj8O
Preferably in ImageJ, but I'm happy to try other platforms.

Thanks!

Best, zdenek
--
--
Zdenek Svindrych, Ph.D.
Research Scientist - Microscopy Imaging Specialist
Department of Biochemistry and Cell Biology
Geisel School of Medicine at Dartmouth

Montero Llopis, Paula De La Milagrosa

Re: Phase contrast image segmentation

In reply to this post by Zdenek Svindrych-2

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Zdenek,

There are several open source tools you can you to do image segmentation of bacterial cells, in most cases using phase contrast and/or fluorescence, but you may be able to use brightfiled/DIC. In most of these tools you can extract a lot of information and get fancy if needed. In many cases they have great documentation and tutorials to get to set up and, in some cases, there is online help. Here are links to a few of them (not comprehensive). The article link even has a figure that compares some of the tools.

https://www.microbej.com/ This one is Image J based, so you may be interested in it
https://oufti.org/ Standalone and/or MATLAB based if you wanna get fancy
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7027861/
https://drescherlab.org/data/bacstalk/docs/ This one is mostly for stalk measurements

I hope this helps. Bacteria can be tricky to segment!

Thank you.

Best regards,

Paula

Paula Montero Llopis, PhD

MicRoN Director

77 Avenue Louis Pasteur

NRB1032

Office 617 432 0177

Cell 203 927 1468

micron.hms.harvard.edu

________________________________
From: Confocal Microscopy List <[hidden email]> on behalf of Zdenek Svindrych <[hidden email]>
Sent: Tuesday, September 15, 2020 5:33 PM
To: [hidden email] <[hidden email]>
Subject: Phase contrast image segmentation

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Dear Listers,

not exactly a confocal question, but here it goes: Do you know of a simple,
reliable, or even barely useable way to segment transmitted light images of
yeast or bacteria, such as this one?:
https://drive.google.com/file/d/15ke38DZsADu8ahL9m1WzBEiIxIsBzj8O
Preferably in ImageJ, but I'm happy to try other platforms.

Thanks!

Best, zdenek
--
--
Zdenek Svindrych, Ph.D.
Research Scientist - Microscopy Imaging Specialist
Department of Biochemistry and Cell Biology
Geisel School of Medicine at Dartmouth

mmodel

Re: Phase contrast image segmentation

Benjamin Smith

Re: Phase contrast image segmentation

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Here is a very quick and dirty ImageJ macro to hopefully lay a foundation
for developing a more refined approach (see below). Each step in the macro
is commented to explain why that step was done. You also may get a much
better answer if you post this same question on the ImageJ forum:
https://forum.image.sc/ Note that this macro also makes use of one plugin,
Morphological Segmentation, so it requires the MorphoLibJ library.
Although, this is a good library to have anyway - lots of good tools in
there. The result of the macro should look like this:
https://drive.google.com/file/d/1680J9Y8zfTYsDBMLqxiC356Nkf2d40FV

Cheers,
Ben Smith

rename("test");
run("8-bit"); //Convert to gray scale run("Bandpass Filter...",
"filter_large=10 filter_small=1 suppress=None tolerance=5 autoscale
saturate"); //Remove background mottling
setAutoThreshold("Yen"); //Threshold dark border around bacteria
setOption("BlackBackground", true);
run("Convert to Mask");
run("Skeletonize"); //Reduce borders to single line (collapses background
to points)
run("Analyze Particles...", "size=50-Infinity show=Masks"); //Remove
background objects
//Not all of the bacteria borders are fully closed so use a watershed to
close them
run("Distance Map"); //Create a distance map
run("Invert");
run("Morphological Segmentation"); //Use the morphological segmentation
watershed, which lets you jump over small ridges
wait(1000);
selectWindow("Morphological Segmentation");
call("inra.ijpb.plugins.MorphologicalSegmentation.segment",
"tolerance=3.0", "calculateDams=true", "connectivity=4"); //Specify minimum
ridge height to 3
wait(1000);
call("inra.ijpb.plugins.MorphologicalSegmentation.setDisplayFormat",
"Watershed lines"); //Set output to watershed dams
wait(1000);
call("inra.ijpb.plugins.MorphologicalSegmentation.createResultImage");
//Generate the output image
selectWindow("Morphological Segmentation"); //Close the Morphological
segmentation tool
run("Close");
selectWindow("Mask of test-watershed-lines");
run("Erode"); //Solidify borders (needed for particle analyzer)
run("Analyze Particles...", "size=50-1000 show=[Count Masks] clear");
//Remove background objects
selectWindow("Count Masks of Mask of test-watershed-lines");
//Color code the count mask useing the random LUT
reds = newArray(0, 231, 25, 27, 112, 89, 53, 77, 155, 153, 91, 19, 25, 135,
187, 40, 52, 39, 147, 72, 92, 23, 5, 202, 52, 50, 108, 66, 238, 46, 136, 8,
162, 126, 75, 60, 135, 167, 136, 244, 161, 204, 158, 112, 192, 2, 98, 207,
233, 121, 191, 18, 210, 151, 252, 155, 139, 42, 188, 170, 127, 151, 38, 16,
227, 120, 137, 2, 72, 71, 214, 130, 194, 96, 102, 33, 92, 47, 211, 226,
237, 132, 244, 45, 82, 73, 239, 198, 118, 191, 170, 238, 177, 246, 66, 5,
44, 3, 18, 40, 226, 94, 130, 216, 68, 89, 150, 40, 96, 155, 95, 5, 132, 88,
124, 241, 143, 15, 1, 220, 46, 220, 216, 155, 190, 4, 98, 33, 142, 153,
185, 61, 153, 160, 164, 102, 75, 8, 26, 231, 160, 245, 119, 69, 95, 137,
232, 249, 89, 220, 54, 3, 43, 134, 172, 31, 106, 234, 132, 1, 29, 65, 12,
107, 52, 196, 95, 165, 4, 101, 124, 240, 203, 50, 142, 231, 5, 23, 43, 198,
41, 80, 66, 232, 86, 151, 126, 51, 122, 111, 162, 192, 148, 210, 5, 251,
40, 176, 32, 2, 22, 91, 180, 31, 218, 9, 10, 178, 27, 115, 54, 199, 163,
10, 180, 222, 102, 198, 202, 220, 96, 145, 237, 127, 172, 51, 64, 47, 84,
37, 196, 147, 244, 206, 244, 43, 170, 186, 162, 157, 125, 230, 212, 140,
186, 127, 12, 48, 0, 92, 168, 73, 228, 121, 248, 255);
greens = newArray(0, 217, 131, 18, 66, 198, 59, 170, 108, 8, 167, 207, 248,
138, 215, 204, 95, 249, 102, 66, 1, 138, 71, 33, 209, 80, 72, 87, 129, 24,
200, 98, 16, 160, 47, 1, 224, 49, 126, 41, 235, 205, 84, 249, 222, 125, 5,
58, 90, 137, 73, 98, 109, 167, 91, 106, 68, 23, 10, 241, 51, 35, 77, 112,
243, 235, 53, 41, 21, 134, 106, 60, 231, 107, 124, 96, 131, 62, 213, 220,
78, 96, 44, 106, 128, 96, 129, 164, 59, 86, 45, 87, 237, 192, 171, 93, 223,
206, 158, 42, 255, 175, 142, 177, 48, 67, 221, 123, 125, 182, 135, 31, 163,
211, 220, 28, 81, 113, 235, 236, 30, 104, 238, 227, 18, 134, 193, 15, 122,
151, 6, 132, 0, 128, 88, 20, 130, 87, 127, 55, 13, 76, 15, 124, 251, 160,
177, 179, 14, 7, 189, 49, 210, 77, 63, 132, 17, 84, 214, 63, 206, 252, 184,
168, 109, 129, 89, 70, 66, 151, 151, 100, 27, 123, 35, 210, 112, 183, 135,
97, 19, 7, 159, 221, 46, 253, 42, 229, 101, 187, 10, 33, 152, 138, 236,
208, 117, 230, 98, 200, 139, 200, 255, 19, 219, 183, 181, 3, 43, 143, 130,
75, 21, 228, 121, 208, 51, 82, 41, 233, 56, 220, 190, 176, 136, 108, 88,
118, 228, 206, 11, 170, 236, 232, 204, 241, 241, 69, 71, 28, 143, 207, 52,
188, 183, 80, 4, 222, 162, 30, 213, 228, 119, 142, 10, 255);
blues = newArray(0, 43, 203, 219, 253, 158, 5, 190, 72, 11, 23, 233, 220,
246, 53, 10, 90, 49, 215, 182, 74, 50, 27, 107, 39, 48, 192, 134, 247, 89,
14, 3, 67, 65, 30, 136, 78, 129, 178, 138, 186, 204, 5, 160, 18, 103, 255,
162, 42, 128, 213, 204, 49, 80, 181, 130, 60, 185, 31, 203, 184, 89, 108,
190, 109, 157, 231, 62, 128, 96, 150, 153, 160, 54, 24, 143, 90, 216, 128,
120, 87, 244, 15, 213, 235, 142, 140, 33, 98, 164, 202, 38, 84, 63, 229,
163, 28, 239, 210, 131, 79, 83, 168, 79, 89, 170, 26, 168, 149, 217, 31,
20, 11, 141, 121, 139, 21, 58, 194, 75, 110, 234, 16, 126, 24, 41, 62, 88,
232, 38, 243, 83, 195, 58, 84, 106, 151, 32, 146, 24, 140, 217, 176, 186,
174, 170, 250, 1, 48, 141, 99, 213, 127, 46, 97, 12, 143, 237, 153, 185,
72, 170, 23, 222, 216, 23, 92, 232, 54, 64, 72, 128, 182, 38, 192, 138,
115, 162, 231, 2, 143, 117, 167, 196, 4, 182, 220, 52, 252, 34, 2, 233,
132, 135, 230, 36, 199, 228, 222, 43, 119, 7, 148, 59, 10, 35, 158, 237,
17, 116, 110, 129, 172, 233, 172, 47, 114, 26, 115, 212, 177, 157, 74, 183,
102, 132, 151, 18, 242, 242, 12, 138, 21, 159, 165, 213, 230, 147, 174,
206, 116, 9, 242, 233, 202, 205, 116, 169, 80, 53, 161, 239, 211, 73, 96,
255);
setLut(reds, greens, blues);
close("\\Others");

On Tue, Sep 15, 2020 at 6:35 PM MODEL, MICHAEL <[hidden email]> wrote:

--
Benjamin E. Smith, Ph. D.
Imaging Specialist, Vision Science
University of California, Berkeley
195 Life Sciences Addition
Berkeley, CA 94720-3200
Tel (510) 642-9712
Fax (510) 643-6791
e-mail: [hidden email]
https://vision.berkeley.edu/faculty/core-grants-nei/core-grant-microscopic-imaging/

Straatman, Kees (Dr.)

Re: Phase contrast image segmentation

In reply to this post by Montero Llopis, Paula De La Milagrosa

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Dear Zdenek,

I get relatively good results using first Process > FFT > Bandpass Filter.. (I used default settings, did not test other options) and used the resulting image in the Trainable Weka Segmentation plugin. I used three
classes: 1 for background, 1 for bacterial wall and 1 for internal of bacteria. After a few rounds of training and finetuning of the classes the resulting image can be thresholded to get the bacteria. I used the internal selection, but bacterial wall can be used as well potentially in combination with the internal selection to separate the bacteria better.

Best wishes

Kees

Dr Ir K.R. Straatman
Advanced Imaging Facility

University of Leicester
www.le.ac.uk/advanced-imaging-facility<http://www.le.ac.uk/advanced-imaging-facility>

________________________________
From: Confocal Microscopy List <[hidden email]> on behalf of Montero Llopis, Paula De La Milagrosa <[hidden email]>
Sent: 16 September 2020 00:47
To: [hidden email] <[hidden email]>
Subject: Re: Phase contrast image segmentation

*****
To join, leave or search the confocal microscopy listserv, go to:
https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Flists.umn.edu%2Fcgi-bin%2Fwa%3FA0%3Dconfocalmicroscopy&data=02%7C01%7Ckrs5%40LEICESTER.AC.UK%7C9ad31e33e84d4106d36508d859d1f9cb%7Caebecd6a31d44b0195ce8274afe853d9%7C0%7C0%7C637358105674191273&sdata=z84NLOrF16p0ZYUvOJQWQPduKVrIGKRI71lWsiKPEyc%3D&reserved=0
Post images on https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.imgur.com%2F&data=02%7C01%7Ckrs5%40LEICESTER.AC.UK%7C9ad31e33e84d4106d36508d859d1f9cb%7Caebecd6a31d44b0195ce8274afe853d9%7C0%7C0%7C637358105674191273&sdata=%2BIjcotlxzr1xI0x0%2FNw99mFer%2BC2PqXS%2BbOtKiENA5U%3D&reserved=0 and include the link in your posting.
*****

Hi Zdenek,

There are several open source tools you can you to do image segmentation of bacterial cells, in most cases using phase contrast and/or fluorescence, but you may be able to use brightfiled/DIC. In most of these tools you can extract a lot of information and get fancy if needed. In many cases they have great documentation and tutorials to get to set up and, in some cases, there is online help. Here are links to a few of them (not comprehensive). The article link even has a figure that compares some of the tools.

https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.microbej.com%2F&data=02%7C01%7Ckrs5%40LEICESTER.AC.UK%7C9ad31e33e84d4106d36508d859d1f9cb%7Caebecd6a31d44b0195ce8274afe853d9%7C0%7C0%7C637358105674191273&sdata=4vvbbn45ZAMzz2fsbPQWXwLDu89noMBXwood8lZYZ6c%3D&reserved=0 This one is Image J based, so you may be interested in it
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Foufti.org%2F&data=02%7C01%7Ckrs5%40LEICESTER.AC.UK%7C9ad31e33e84d4106d36508d859d1f9cb%7Caebecd6a31d44b0195ce8274afe853d9%7C0%7C0%7C637358105674191273&sdata=MR4yd4YPR3lxZaS%2BKrZbD0TvQcG6nwbJC8m0lIbxYDA%3D&reserved=0 Standalone and/or MATLAB based if you wanna get fancy
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.ncbi.nlm.nih.gov%2Fpmc%2Farticles%2FPMC7027861%2F&data=02%7C01%7Ckrs5%40LEICESTER.AC.UK%7C9ad31e33e84d4106d36508d859d1f9cb%7Caebecd6a31d44b0195ce8274afe853d9%7C0%7C0%7C637358105674191273&sdata=897JXS01XPehasEOcK%2BPt4PyyN%2FAeiGeRBgXg4CKSyA%3D&reserved=0
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdrescherlab.org%2Fdata%2Fbacstalk%2Fdocs%2F&data=02%7C01%7Ckrs5%40LEICESTER.AC.UK%7C9ad31e33e84d4106d36508d859d1f9cb%7Caebecd6a31d44b0195ce8274afe853d9%7C0%7C0%7C637358105674191273&sdata=J7QPlm7Ch3SvF%2BsHlCe5U8JiOUJgGkupp%2FE8yCWQzA8%3D&reserved=0 This one is mostly for stalk measurements

I hope this helps. Bacteria can be tricky to segment!

Thank you.

Best regards,

Paula

Paula Montero Llopis, PhD

MicRoN Director

77 Avenue Louis Pasteur

NRB1032

Office 617 432 0177

Cell 203 927 1468

micron.hms.harvard.edu

________________________________
From: Confocal Microscopy List <[hidden email]> on behalf of Zdenek Svindrych <[hidden email]>
Sent: Tuesday, September 15, 2020 5:33 PM
To: [hidden email] <[hidden email]>
Subject: Phase contrast image segmentation

*****
To join, leave or search the confocal microscopy listserv, go to:
https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Flists.umn.edu%2Fcgi-bin%2Fwa%3FA0%3Dconfocalmicroscopy&data=02%7C01%7Ckrs5%40LEICESTER.AC.UK%7C9ad31e33e84d4106d36508d859d1f9cb%7Caebecd6a31d44b0195ce8274afe853d9%7C0%7C0%7C637358105674191273&sdata=z84NLOrF16p0ZYUvOJQWQPduKVrIGKRI71lWsiKPEyc%3D&reserved=0
Post images on https://eur03.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.imgur.com%2F&data=02%7C01%7Ckrs5%40LEICESTER.AC.UK%7C9ad31e33e84d4106d36508d859d1f9cb%7Caebecd6a31d44b0195ce8274afe853d9%7C0%7C0%7C637358105674201268&sdata=F4dEVlxzjGhdOzcAjAcESyWWHobFGzSo7Z9Kz%2F5BqHk%3D&reserved=0 and include the link in your posting.
*****

Dear Listers,

not exactly a confocal question, but here it goes: Do you know of a simple,
reliable, or even barely useable way to segment transmitted light images of
yeast or bacteria, such as this one?:
https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdrive.google.com%2Ffile%2Fd%2F15ke38DZsADu8ahL9m1WzBEiIxIsBzj8O&data=02%7C01%7Ckrs5%40LEICESTER.AC.UK%7C9ad31e33e84d4106d36508d859d1f9cb%7Caebecd6a31d44b0195ce8274afe853d9%7C0%7C0%7C637358105674201268&sdata=rGyTKKEb7OXCLXbX7WEzBlYQR%2FQDhYbGyGjvla65EM4%3D&reserved=0
Preferably in ImageJ, but I'm happy to try other platforms.

Thanks!

Best, zdenek
--
--
Zdenek Svindrych, Ph.D.
Research Scientist - Microscopy Imaging Specialist
Department of Biochemistry and Cell Biology
Geisel School of Medicine at Dartmouth