Starting with Fiji

[T Purushotham]

Respected to all,
Greetings of the day. I have lot of case images of fetal echo with different types like four chamber view, LVOT, RVOT, Three vessel view and Three vessel arch view.Now, I want to develop a macro or algorithm in such a way that, whenever I upload the image  it can automatically identifies the image whether it belongs to which type in above mentioned five cases.
                 Can  anyone help me to move forward in this. Also, how far this ImageJ/Fijji useful for me to overcome my problem. Mention any new Machine Learning strategies to solve this.

Thanks and Regards,
T Purushotham,
 IIT- Madras.


--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[TimFeinstein]

Hello,

It sounds like your images will likely have many variables – intensity, rotation angle, exact size and position of the section, etc.  This will make it difficult to ‘forward’ engineer a macro by brute force estimation of variables.  This sounds like a problem for a trainable machine learning classification algorithm.  Fiji has a few options, most notably Trainable Weka Segmentation.  If those do not do the job there are some more dedicated machine learning classification programs out there, including free ones like CellProfiler Analyst and AdvancedCellClassifier.  As their names suggest they are meant for classifying cells, but the machine learning algorithms may be adaptable to your problem as well.  

Good luck!


Tim

Timothy Feinstein, Ph.D.
Research Scientist
Department of Developmental Biology
University of Pittsburgh

 


On 9/19/17, 12:54 AM, "ImageJ Interest Group on behalf of T Purushotham" <[hidden email] on behalf of [hidden email]> wrote:

    Respected to all,
    Greetings of the day. I have lot of case images of fetal echo with different types like four chamber view, LVOT, RVOT, Three vessel view and Three vessel arch view.Now, I want to develop a macro or algorithm in such a way that, whenever I upload the image  it can automatically identifies the image whether it belongs to which type in above mentioned five cases.
                     Can  anyone help me to move forward in this. Also, how far this ImageJ/Fijji useful for me to overcome my problem. Mention any new Machine Learning strategies to solve this.
   
    Thanks and Regards,
    T Purushotham,
     IIT- Madras.
   
   
    --
    ImageJ mailing list: https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fimagej.nih.gov%2Fij%2Flist.html&data=01%7C01%7Ctnf8%40pitt.edu%7C1761090b5c0b4eea3baa08d4ff1abac8%7C9ef9f489e0a04eeb87cc3a526112fd0d%7C1&sdata=%2BhdteZdG39anzO8vCozaeC8tGjkxJPvn%2BO0mJucTgXs%3D&reserved=0
   


--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[anusuya pal]

Hello,

If I understand your problem correctly, you can use Biovoxxel tool in Fiji.
This tool will separate the particles in terms of many parameters. And the
best thing is you can manually set the parameters according to your image.

I hope it helps.

Thanks
Anu

On 20-Sep-2017 7:56 AM, "Feinstein, Timothy N" <[hidden email]> wrote:

> Hello,
>
> It sounds like your images will likely have many variables – intensity,
> rotation angle, exact size and position of the section, etc.  This will
> make it difficult to ‘forward’ engineer a macro by brute force estimation
> of variables.  This sounds like a problem for a trainable machine learning
> classification algorithm.  Fiji has a few options, most notably Trainable
> Weka Segmentation.  If those do not do the job there are some more
> dedicated machine learning classification programs out there, including
> free ones like CellProfiler Analyst and AdvancedCellClassifier.  As their
> names suggest they are meant for classifying cells, but the machine
> learning algorithms may be adaptable to your problem as well.
>
> Good luck!
>
>
> Tim
>
> Timothy Feinstein, Ph.D.
> Research Scientist
> Department of Developmental Biology
> University of Pittsburgh
>
>
>
>
> On 9/19/17, 12:54 AM, "ImageJ Interest Group on behalf of T Purushotham" <
> [hidden email] on behalf of [hidden email]> wrote:
>
>     Respected to all,
>     Greetings of the day. I have lot of case images of fetal echo with
> different types like four chamber view, LVOT, RVOT, Three vessel view and
> Three vessel arch view.Now, I want to develop a macro or algorithm in such
> a way that, whenever I upload the image  it can automatically identifies
> the image whether it belongs to which type in above mentioned five cases.
>                      Can  anyone help me to move forward in this. Also,
> how far this ImageJ/Fijji useful for me to overcome my problem. Mention any
> new Machine Learning strategies to solve this.
>
>     Thanks and Regards,
>     T Purushotham,
>      IIT- Madras.
>
>
>     --
>     ImageJ mailing list: https://na01.safelinks.
> protection.outlook.com/?url=http%3A%2F%2Fimagej.nih.gov%
> 2Fij%2Flist.html&data=01%7C01%7Ctnf8%40pitt.edu%
> 7C1761090b5c0b4eea3baa08d4ff1abac8%7C9ef9f489e0a04eeb87cc3a526112
> fd0d%7C1&sdata=%2BhdteZdG39anzO8vCozaeC8tGjkxJ
> Pvn%2BO0mJucTgXs%3D&reserved=0
>
>
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html