Summary

• Page properties (2 modified, 0 added, 0 removed)

Details

Page properties

Author

...	...	@@ -1,1 +1,1 @@
1		-XWiki.puchades
	1	+XWiki.sharoncy

Content

...	...	@@ -1,13 +1,19 @@
1		-== [[image:ilastik_logo.PNG||style="float:right"]](% style="color:#c0392b" %)Analysis approach for series of rodent brain section image(%%) ==
	1	+== [[image:ilastik_logo.PNG||style="float:right"]] ==
2	2
	3	+
	4	+(% class="wikigeneratedid" %)
	5	+== (% style="color:#c0392b" %)Analysis approach for series of rodent brain section image(%%) ==
	6	+
	7	+Ilastik is a versatile image analysis tool specifically designed for the classification, segmentation and analysis of biological images based on supervised machine learning algorithms.
	8	+
3	3	There are two main approaches for the analysis of rodent brain section images.
4	4
5	5	1. Pixel classification only (with two or more classes)
6		-1. Pixel classification with two classes (//immunoreactivity// and //background//), followed by object classification with two classes (//objects-of-interest// and //artefact//).
	12	+1. Pixel classification with two classes (//immunoreactivity// and //background//), followed by object classification with two classes (//objects of interest// and //artefact//).
7	7
8	8	**Which approach is best for my dataset?**
9	9
10		-As a general rule, pixel classification is suitable for images in which there are clear differences in the colour, intensity and/ or texture of the feature-of-interest versus the background and other structures.  If there is non-specific labelling in the image that is very similar in appearance to the labelling-of-interest, object classification may allow the non-specific labelling to be filtered out based on object level features such as size and shape. The best approach is determined by trial and error.
	16	+As a general rule, pixel classification is suitable for images in which there are clear differences in the colour, intensity and/ or texture of the feature of interest (labelling) versus the background and other structures.  If there is non-specific labelling in the image that is very similar in appearance to the labelling of interest, object classification may allow the non-specific labelling to be filtered out based on object level features such as size and shape. The best approach is determined by trial and error.
11	11
12	12	=== (% style="color:#c0392b" %)Pixel classification workflow(%%) ===
13	13
...	...	@@ -19,9 +19,9 @@
19	19
20	20	-Apply the classifier to the rest of the images (batch processing)
21	21
22		--Export the probability maps in HDH5 format, and simple_segmentation images in .png format, with the default settings.
	28	+-Export the probability maps in HDH5 format, and simple_segmentation images in PNG format with the default settings.
23	23
24		--Review our results.
	30	+-Review the results.
25	25
26	26	=== (% style="color:#c0392b" %)Object classification workflow(%%) ===
27	27
...	...	@@ -29,12 +29,12 @@
29	29
30	30	-Save the object classification file in the same folder as the raw images for analysis.  If the images are moved after the ilastik file is created, the link between the ilastik file and the images may be lost, resulting in a corrupted file.
31	31
32		--In the **Input Data** applet, upload the original images and their respective probability maps in HDH5 format (output from the pixel classification).
	38	+-In the **Input Data** applet, upload the original images and their respective probability maps in HDH5 format (output from the Pixel Classification).
33	33
34		--Train the classifier with two classes (labeling and artefacts)
	40	+-Train the classifier with two classes (labelling and artefacts)
35	35
36	36	(% class="wikigeneratedid" id="H" %)
37	37	-In the **Object Information Export** applet, export “Object Predictions” in PNG format.  Do not change the default export location.
38	38
39	39	(% class="wikigeneratedid" %)
40		--Review our results.
	46	+-Review the results.