Hi Divya,
> Is there any plugin on tool for ImageJ that I can use to automatically
> select the lesions, as they stand out quite nicely from the normal
> skin of the patient, and measure the area of these lesions?
There are two different approaches discussed on the Fiji wiki: the "easy"
way and the "flexible" way. See this page:
http://fiji.sc/SegmentationThe easy way uses the Trainable Weka Segmentation plugin, which is a
machine learning approach. You would paint squiggly lines over the lesions
and add them to a "lesion" class, and paint more lines over the background
and add to a "background" class. Then let the computer "train" from those
manually painted examples to extrapolate over the entire dataset.
The flexible way involves crafting a macro with three broad steps:
1) preprocessing
2) threshold + mask manipulation
3) analyzing the data
Then you can repeatedly execute the macro over your whole dataset.
See the wiki page for more information.
If you can't figure out a good workflow, feel free to post a sample image
somewhere and people may have more specific suggestions.
Regards,
Curtis
On Wed, May 28, 2014 at 10:20 AM, DivyaThomas <
[hidden email]>wrote:
> For a project I need to measure the area of lesions on patients via their
> pictures. The lesions aren't usually connected and do not have nice edges
> so using the draw tool is out of the question unless I want to get
> inaccurate results. Is there any plugin on tool for ImageJ that I can use
> to automatically select the lesions, as they stand out quite nicely from
> the
> normal skin of the patient, and measure the area of these lesions?
>
>
>
> --
> View this message in context:
>
http://imagej.1557.x6.nabble.com/Measure-skin-lesions-tp5007953.html> Sent from the ImageJ mailing list archive at Nabble.com.
>
> --
> ImageJ mailing list:
http://imagej.nih.gov/ij/list.html>
--
ImageJ mailing list:
http://imagej.nih.gov/ij/list.html
Locked
