Team:TU Eindhoven/Modeling/Bacterial Cell Counter

From 2014.igem.org

(Difference between revisions)
Line 154: Line 154:
<h3>Bacterial Cell Detection</h3>
<h3>Bacterial Cell Detection</h3>
-
<p>The program is now at the point where it has to count the number of cells in the droplets. Using the droplets from the edgedetect picture, erosion with value 1 is performed. This makes sure the cells connected to the borders of the droplet are loose, the program deletes the border components so it only finds cells. The program uses EnclosingComponentCount instead of count now because a lot of cells became empty circles to get the results.
+
<p>The program is now at the point where it has to count the number of cells in the droplets. Using the droplets from the edgedetect picture, erosion with value 1 is performed. This makes sure the cells connected to the borders of the droplet are loose, the program deletes the border components so it only finds cells. The program uses <i>EnclosingComponentCount</i> instead of count now because a lot of cells became empty circles to get the results.
<br><br>
<br><br>
The results for this image are 9 false positives and 11 false negatives on a total of 93 cell in 126 droplets. Because some droplets had multiple false positives or false negatives the next step to improve this number is selecting to droplets better.
The results for this image are 9 false positives and 11 false negatives on a total of 93 cell in 126 droplets. Because some droplets had multiple false positives or false negatives the next step to improve this number is selecting to droplets better.

Revision as of 10:51, 11 October 2014

iGEM Team TU Eindhoven 2014

iGEM Team TU Eindhoven 2014

Figuur 1. Sample image used to show how the program works.

Bacterial Cell Counter

When photos of the microfluidics results are taken, the droplets and the cells in the droplets have to be counted. In order to have reliable results this has to be done on a large scale. To make it possible to analyses large numbers of photos a computer program has been made to do the counting. To give a better visualization of the process, a step by step analysis of an image is given below.

The program has the following steps: find and count the droplets, then find and count the cells in the droplets and finally create a histogram of the results (number of droplets with 0 cells, with 1 cell, with 2 et cetera). By adding up the histograms of multiple images you get a view of how the cells are divided over the droplets.

Figuur 2.

Droplet Detection

To detect the cells EdgeDetect (a function of Mathematica using gradient methods) is used followed by a dilation to make the edges clearer.

This results in clusters of pixels, all the pixels that are connected with only black pixels. The program then looks for the clusters bigger than 2000 pixels and smaller than 6000 pixels.

Parameters of Droplets Detection

The parameters of this function are chosen so they have a very low false positive, because a false positive means a non-existing droplet and thus false data. A false negative only lowers the sample size, which can be increased by analyzing more images.

Figuur 3.

Bacterial Cell Detection

The program is now at the point where it has to count the number of cells in the droplets. Using the droplets from the edgedetect picture, erosion with value 1 is performed. This makes sure the cells connected to the borders of the droplet are loose, the program deletes the border components so it only finds cells. The program uses EnclosingComponentCount instead of count now because a lot of cells became empty circles to get the results.

The results for this image are 9 false positives and 11 false negatives on a total of 93 cell in 126 droplets. Because some droplets had multiple false positives or false negatives the next step to improve this number is selecting to droplets better.

Bibliography

iGEM Team TU Eindhoven 2014