Neuron Networks
Locked
 
|
Hi All,
I am looking for image processing software which can perform neuron network analyzing in a quantitative and automated manner. I was wondering whether image J/Fiji has such a plug in. I have series of phase-contrast images (attached one example) taken from growing neural cells, and I try to find a way to assess the differences in degree of cellular connectivity/networking between various conditions. I will be very glad if you could give me some advices on how/with which tools to do such analyses. thank you very much, Nurdan -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
B11.jpg (710K) |
|
Hi Nurdan,
maybe machine-learning would be a easier way to do it than develop an entire image processing pipeline. ilastik ( http://www.ilastik.org/) software could help you. It can classify your images in such an fast and easy way! You would have three classes: - axon - soma+dendrites... - other Once you would have classified your images you could use the results for further statistical analysis. Cheers, David On 08/08/2012 04:10 PM, Oezkucur, Nurdan wrote: > Hi All, > > I am looking for image processing software which can perform neuron network analyzing in a quantitative and automated manner. I was wondering whether image J/Fiji has such a plug in. I have series of phase-contrast images (attached one example) taken from growing neural cells, and I try to find a way to assess the differences in degree of cellular connectivity/networking between various conditions. I will be very glad if you could give me some advices on how/with which tools to do such analyses. > > thank you very much, > Nurdan > > > -- > ImageJ mailing list: http://imagej.nih.gov/ij/list.html -- Envoyé depuis un PC nul, dans une salle de réseau pourrie, sur un ordinateur de merde. -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
|
|
thanks David, it sounds very practical. I will look at that and see what/how I can do.
Nurdan >-----Original Message----- >From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of David >Rueda >Sent: Mittwoch, 8. August 2012 10:55 >To: [hidden email] >Subject: Re: Neuron Networks > >Hi Nurdan, > >maybe machine-learning would be a easier way to do it than develop an entire >image processing pipeline. ilastik ( http://www.ilastik.org/) software could >help you. It can classify your images in such an fast and easy way! You would >have three classes: > - axon > - soma+dendrites... > - other > >Once you would have classified your images you could use the results for >further statistical analysis. > >Cheers, >David > >On 08/08/2012 04:10 PM, Oezkucur, Nurdan wrote: >> Hi All, >> >> I am looking for image processing software which can perform neuron network >analyzing in a quantitative and automated manner. I was wondering whether >image J/Fiji has such a plug in. I have series of phase-contrast images >(attached one example) taken from growing neural cells, and I try to find a >way to assess the differences in degree of cellular connectivity/networking >between various conditions. I will be very glad if you could give me some >advices on how/with which tools to do such analyses. >> >> thank you very much, >> Nurdan >> >> >> -- >> ImageJ mailing list: http://imagej.nih.gov/ij/list.html > > >-- >Envoyé depuis un PC nul, >dans une salle de réseau pourrie, >sur un ordinateur de merde. > > >-- >ImageJ mailing list: http://imagej.nih.gov/ij/list.html -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
|
|
In reply to this post by David Rueda
Hi,
On Wed, 8 Aug 2012, David Rueda wrote: > maybe machine-learning would be a easier way to do it than develop an entire > image processing pipeline. ilastik ( http://www.ilastik.org/) software could > help you. It can classify your images in such an fast and easy way! You would > have three classes: > - axon > - soma+dendrites... > - other Similar to Ilastik is Fiji's Trainable Segmentation: http://fiji.sc/Trainable_Segmentation and its sibling Advanced Weka Segmentation: http://fiji.sc/Advanced_Weka_Segmentation The basic difference between the two is that the Fiji plugins try to integrate with ImageJ, hence have a much simpler user interface with less options. Ilastik is a special-purpose software to provide image segmentation by machine learning and has a pretty sophisticated user interface. So if you need to integrate your segmentation into a larger image processing/analysis workflow, I'd suggest to go with the Advanced Weka Segmentation, otherwise Ilastik seems to be a good choice. Ciao, Johannes -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
|
|
In reply to this post by Oezkucur, Nurdan
You have a very similar approach within Fiji:
http://fiji.sc/wiki/index.php/Advanced_Weka_Segmentation ignacio On Wed, Aug 8, 2012 at 11:30 AM, Oezkucur, Nurdan <[hidden email] > wrote: > thanks David, it sounds very practical. I will look at that and see > what/how I can do. > > Nurdan > > >-----Original Message----- > >From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of > David > >Rueda > >Sent: Mittwoch, 8. August 2012 10:55 > >To: [hidden email] > >Subject: Re: Neuron Networks > > > >Hi Nurdan, > > > >maybe machine-learning would be a easier way to do it than develop an > entire > >image processing pipeline. ilastik ( http://www.ilastik.org/) software > could > >help you. It can classify your images in such an fast and easy way! You > would > >have three classes: > > - axon > > - soma+dendrites... > > - other > > > >Once you would have classified your images you could use the results for > >further statistical analysis. > > > >Cheers, > >David > > > >On 08/08/2012 04:10 PM, Oezkucur, Nurdan wrote: > >> Hi All, > >> > >> I am looking for image processing software which can perform neuron > network > >analyzing in a quantitative and automated manner. I was wondering whether > >image J/Fiji has such a plug in. I have series of phase-contrast images > >(attached one example) taken from growing neural cells, and I try to find > a > >way to assess the differences in degree of cellular > connectivity/networking > >between various conditions. I will be very glad if you could give me some > >advices on how/with which tools to do such analyses. > >> > >> thank you very much, > >> Nurdan > >> > >> > >> -- > >> ImageJ mailing list: http://imagej.nih.gov/ij/list.html > > > > > >-- > >Envoyé depuis un PC nul, > >dans une salle de réseau pourrie, > >sur un ordinateur de merde. > > > > > >-- > >ImageJ mailing list: http://imagej.nih.gov/ij/list.html > > -- > ImageJ mailing list: http://imagej.nih.gov/ij/list.html > -- Ignacio Arganda-Carreras, Ph.D. Seung's lab, 46-5065 Department of Brain and Cognitive Sciences Massachusetts Institute of Technology 43 Vassar St. Cambridge, MA 02139 USA Phone: (001) 617-324-3747 Website: http://bioweb.cnb.csic.es/~iarganda/index_EN.html -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
|
|
In reply to this post by dscho
Hi Johannes,
Thank you very much for your suggestions, highly appreciated. I will check Fiji's too and see what works best for the images I want to analyze. cheers, Nurdan >-----Original Message----- >From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of >Johannes Schindelin >Sent: Mittwoch, 8. August 2012 11:50 >To: [hidden email] >Subject: Re: Neuron Networks > >Hi, > >On Wed, 8 Aug 2012, David Rueda wrote: > >> maybe machine-learning would be a easier way to do it than develop an >> entire image processing pipeline. ilastik ( http://www.ilastik.org/) >> software could help you. It can classify your images in such an fast >> and easy way! You would have three classes: >> - axon >> - soma+dendrites... >> - other > >Similar to Ilastik is Fiji's Trainable Segmentation: >http://fiji.sc/Trainable_Segmentation and its sibling Advanced Weka >Segmentation: http://fiji.sc/Advanced_Weka_Segmentation > >The basic difference between the two is that the Fiji plugins try to >integrate with ImageJ, hence have a much simpler user interface with less >options. Ilastik is a special-purpose software to provide image segmentation >by machine learning and has a pretty sophisticated user interface. > >So if you need to integrate your segmentation into a larger image >processing/analysis workflow, I'd suggest to go with the Advanced Weka >Segmentation, otherwise Ilastik seems to be a good choice. > >Ciao, >Johannes > >-- >ImageJ mailing list: http://imagej.nih.gov/ij/list.html -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
|
|
In reply to this post by Ignacio Arganda-Carreras
thanks Ignacio, I will also try Fiji. One question ahead; what exactly do the plots tell us about the images processed (segmented, having introduced objects like axons)?
nurdan On Aug 9, 2012, at 20:40, "Ignacio Arganda-Carreras" <[hidden email]> wrote: > You have a very similar approach within Fiji: > > http://fiji.sc/wiki/index.php/Advanced_Weka_Segmentation > > ignacio > > On Wed, Aug 8, 2012 at 11:30 AM, Oezkucur, Nurdan <[hidden email] >> wrote: > >> thanks David, it sounds very practical. I will look at that and see >> what/how I can do. >> >> Nurdan >> >>> -----Original Message----- >>> From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of >> David >>> Rueda >>> Sent: Mittwoch, 8. August 2012 10:55 >>> To: [hidden email] >>> Subject: Re: Neuron Networks >>> >>> Hi Nurdan, >>> >>> maybe machine-learning would be a easier way to do it than develop an >> entire >>> image processing pipeline. ilastik ( http://www.ilastik.org/) software >> could >>> help you. It can classify your images in such an fast and easy way! You >> would >>> have three classes: >>> - axon >>> - soma+dendrites... >>> - other >>> >>> Once you would have classified your images you could use the results for >>> further statistical analysis. >>> >>> Cheers, >>> David >>> >>> On 08/08/2012 04:10 PM, Oezkucur, Nurdan wrote: >>>> Hi All, >>>> >>>> I am looking for image processing software which can perform neuron >> network >>> analyzing in a quantitative and automated manner. I was wondering whether >>> image J/Fiji has such a plug in. I have series of phase-contrast images >>> (attached one example) taken from growing neural cells, and I try to find >> a >>> way to assess the differences in degree of cellular >> connectivity/networking >>> between various conditions. I will be very glad if you could give me some >>> advices on how/with which tools to do such analyses. >>>> >>>> thank you very much, >>>> Nurdan >>>> >>>> >>>> -- >>>> ImageJ mailing list: http://imagej.nih.gov/ij/list.html >>> >>> >>> -- >>> Envoyé depuis un PC nul, >>> dans une salle de réseau pourrie, >>> sur un ordinateur de merde. >>> >>> >>> -- >>> ImageJ mailing list: http://imagej.nih.gov/ij/list.html >> >> -- >> ImageJ mailing list: http://imagej.nih.gov/ij/list.html >> > > > > -- > Ignacio Arganda-Carreras, Ph.D. > Seung's lab, 46-5065 > Department of Brain and Cognitive Sciences > Massachusetts Institute of Technology > 43 Vassar St. > Cambridge, MA 02139 > USA > > Phone: (001) 617-324-3747 > Website: http://bioweb.cnb.csic.es/~iarganda/index_EN.html > > -- > ImageJ mailing list: http://imagej.nih.gov/ij/list.html -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
|
|
Do you mean the "Plot result" option? That has to do with the accuracy
of the result. It's a way of visualizing how well the method performed based on true positive ratio vs false positive ratio. ignacio On Thu, Aug 9, 2012 at 10:20 PM, Oezkucur, Nurdan <[hidden email]> wrote: > thanks Ignacio, I will also try Fiji. One question ahead; what exactly do the plots tell us about the images processed (segmented, having introduced objects like axons)? > > nurdan > > On Aug 9, 2012, at 20:40, "Ignacio Arganda-Carreras" <[hidden email]> wrote: > >> You have a very similar approach within Fiji: >> >> http://fiji.sc/wiki/index.php/Advanced_Weka_Segmentation >> >> ignacio >> >> On Wed, Aug 8, 2012 at 11:30 AM, Oezkucur, Nurdan <[hidden email] >>> wrote: >> >>> thanks David, it sounds very practical. I will look at that and see >>> what/how I can do. >>> >>> Nurdan >>> >>>> -----Original Message----- >>>> From: ImageJ Interest Group [mailto:[hidden email]] On Behalf Of >>> David >>>> Rueda >>>> Sent: Mittwoch, 8. August 2012 10:55 >>>> To: [hidden email] >>>> Subject: Re: Neuron Networks >>>> >>>> Hi Nurdan, >>>> >>>> maybe machine-learning would be a easier way to do it than develop an >>> entire >>>> image processing pipeline. ilastik ( http://www.ilastik.org/) software >>> could >>>> help you. It can classify your images in such an fast and easy way! You >>> would >>>> have three classes: >>>> - axon >>>> - soma+dendrites... >>>> - other >>>> >>>> Once you would have classified your images you could use the results for >>>> further statistical analysis. >>>> >>>> Cheers, >>>> David >>>> >>>> On 08/08/2012 04:10 PM, Oezkucur, Nurdan wrote: >>>>> Hi All, >>>>> >>>>> I am looking for image processing software which can perform neuron >>> network >>>> analyzing in a quantitative and automated manner. I was wondering whether >>>> image J/Fiji has such a plug in. I have series of phase-contrast images >>>> (attached one example) taken from growing neural cells, and I try to find >>> a >>>> way to assess the differences in degree of cellular >>> connectivity/networking >>>> between various conditions. I will be very glad if you could give me some >>>> advices on how/with which tools to do such analyses. >>>>> >>>>> thank you very much, >>>>> Nurdan >>>>> >>>>> >>>>> -- >>>>> ImageJ mailing list: http://imagej.nih.gov/ij/list.html >>>> >>>> >>>> -- >>>> Envoyé depuis un PC nul, >>>> dans une salle de réseau pourrie, >>>> sur un ordinateur de merde. >>>> >>>> >>>> -- >>>> ImageJ mailing list: http://imagej.nih.gov/ij/list.html >>> >>> -- >>> ImageJ mailing list: http://imagej.nih.gov/ij/list.html >>> >> >> >> >> -- >> Ignacio Arganda-Carreras, Ph.D. >> Seung's lab, 46-5065 >> Department of Brain and Cognitive Sciences >> Massachusetts Institute of Technology >> 43 Vassar St. >> Cambridge, MA 02139 >> USA >> >> Phone: (001) 617-324-3747 >> Website: http://bioweb.cnb.csic.es/~iarganda/index_EN.html >> >> -- >> ImageJ mailing list: http://imagej.nih.gov/ij/list.html > > -- > ImageJ mailing list: http://imagej.nih.gov/ij/list.html -- Ignacio Arganda-Carreras, Ph.D. Seung's lab, 46-5065 Department of Brain and Cognitive Sciences Massachusetts Institute of Technology 43 Vassar St. Cambridge, MA 02139 USA Phone: (001) 617-324-3747 Website: http://bioweb.cnb.csic.es/~iarganda/index_EN.html -- ImageJ mailing list: http://imagej.nih.gov/ij/list.html |
«
Return to ImageJ
|
1 view|%1 views
| Free forum by Nabble | Edit this page |