From samuel.garcia at cnrs.fr Tue Sep 4 10:38:43 2018 From: samuel.garcia at cnrs.fr (Samuel Garcia) Date: Tue, 4 Sep 2018 16:38:43 +0200 Subject: [Neuroimaging] Spike sorting with tridesclous : release 1.0.0 Message-ID: <75e0eb58-7c83-398a-1487-17d7cfd0383c@cnrs.fr> Dear all, I'm pleased to announce the v1.0.0 release of tridesclous . This tools could interest researcher that use multi electrode cellular electrophysiology recordings. Tridesclous is a tool for *spike sorting* written in python. Features: * can be used *offline* with several formats (binary raw, blackrock, neuralynx, plexon, tdt...) * can be used *online* with pyacq with several hardware (NI, MCC, blackrock, multichannel, ...) * methods based on mix of "legacy clustering" + "template matching", so it resolve spike collision * well design user interface (UI) with multi view. This allow fast and precise check of isolated neurons. * 2 level for users : script or UI * usable from several channel (tetrodes) to some hundreds of channels (dense arrays) * can use GPU (with opencl) for some step of processing, so it can be fast * quick start guide with open data * 100% open source Doc: https://tridesclous.readthedocs.io/en/latest/ Code: https://github.com/tridesclous/tridesclous Best, Samuel -------------- next part -------------- An HTML attachment was scrubbed... URL: From keschenb at uw.edu Tue Sep 4 14:11:29 2018 From: keschenb at uw.edu (Kristian M Eschenburg) Date: Tue, 4 Sep 2018 11:11:29 -0700 Subject: [Neuroimaging] using bedpostx diffusion model in dipy Message-ID: Hi All Question -- does dipy support the use of the bedpostx diffusion model for generating probabilistic tractography? I've run bedpostx on a large set of diffusion images and am interested in using some of the tracking methods contained in Dipy. Thanks! k -------------- next part -------------- An HTML attachment was scrubbed... URL: From alexandre.gramfort at inria.fr Tue Sep 4 14:57:43 2018 From: alexandre.gramfort at inria.fr (Alexandre Gramfort) Date: Tue, 4 Sep 2018 20:57:43 +0200 Subject: [Neuroimaging] Spike sorting with tridesclous : release 1.0.0 In-Reply-To: <75e0eb58-7c83-398a-1487-17d7cfd0383c@cnrs.fr> References: <75e0eb58-7c83-398a-1487-17d7cfd0383c@cnrs.fr> Message-ID: congrats Samuel ! Alex On Tue, Sep 4, 2018 at 4:42 PM Samuel Garcia wrote: > Dear all, > > I'm pleased to announce the v1.0.0 release of tridesclous > . > > This tools could interest researcher that use multi electrode cellular > electrophysiology recordings. > > Tridesclous is a tool for *spike sorting* written in python. > > > Features: > > - can be used *offline* with several formats (binary raw, blackrock, > neuralynx, plexon, tdt...) > - can be used *online* with pyacq with several hardware (NI, MCC, > blackrock, multichannel, ...) > - methods based on mix of "legacy clustering" + "template matching", > so it resolve spike collision > - well design user interface (UI) with multi view. This allow fast and > precise check of isolated neurons. > - 2 level for users : script or UI > - usable from several channel (tetrodes) to some hundreds of channels > (dense arrays) > - can use GPU (with opencl) for some step of processing, so it can be > fast > - quick start guide with open data > - 100% open source > > > Doc: https://tridesclous.readthedocs.io/en/latest/ > > Code: https://github.com/tridesclous/tridesclous > > > Best, > > > Samuel > _______________________________________________ > Neuroimaging mailing list > Neuroimaging at python.org > https://mail.python.org/mailman/listinfo/neuroimaging > -------------- next part -------------- An HTML attachment was scrubbed... URL: From elef at indiana.edu Tue Sep 4 19:48:03 2018 From: elef at indiana.edu (Eleftherios Garyfallidis) Date: Tue, 4 Sep 2018 19:48:03 -0400 Subject: [Neuroimaging] using bedpostx diffusion model in dipy In-Reply-To: References: Message-ID: I do not think so. As far as I know we have never tried this. In theory should be doable with a bit of code (or more than a bit). On Tue, Sep 4, 2018 at 2:12 PM Kristian M Eschenburg wrote: > Hi All > > Question -- does dipy support the use of the bedpostx diffusion model for > generating probabilistic tractography? I've run bedpostx on a large set of > diffusion images and am interested in using some of the tracking methods > contained in Dipy. > > Thanks! > > > k > _______________________________________________ > Neuroimaging mailing list > Neuroimaging at python.org > https://mail.python.org/mailman/listinfo/neuroimaging > -------------- next part -------------- An HTML attachment was scrubbed... URL: From matthew.brett at gmail.com Thu Sep 6 05:32:44 2018 From: matthew.brett at gmail.com (Matthew Brett) Date: Thu, 6 Sep 2018 10:32:44 +0100 Subject: [Neuroimaging] Affine transformations In-Reply-To: References: <20180831145127.Horde.dEbME051yHvSIMhXHdsPte9@webmail.uni-jena.de> Message-ID: Oops, On Fri, Aug 31, 2018 at 2:15 PM, Matthew Brett wrote: [snip] > I think you'll find that the center doesn't matter for the zooms > because, if Z is a zooming affine, and hence of this form: > > Z = [[a, 0, 0, 0], > [0, b, 0, 0], > [0, 0, c, 0], > [0, 0, 0, 1]] > > then Tinv.dot(Z).dot(T) == Z for any translations in T. How silly of me. For some reason, as I woke up, I realized that's obviously not true, and the zooms do change the translations, as would have been obvious if I'd done the multiplication properly: In [22]: import sympy as sy In [23]: a, b, c, x, y, z = sy.symbols('a, b, c, x, y, z') In [24]: T = Matrix([[1, 0, 0, x], ...: [0, 1, 0, y], ...: [0, 0, 1, z], ...: [0, 0, 0, 1]]) ...: In [25]: Z = Matrix([[a, 0, 0, 0], ...: [0, b, 0, 0], ...: [0, 0, c, 0], ...: [0, 0, 0, 1]]) ...: In [26]: T.inv() * Z * T Out[26]: Matrix([ [a, 0, 0, a*x - x], [0, b, 0, b*y - y], [0, 0, c, c*z - z], [0, 0, 0, 1]]) So - yes - it does make a difference (to the origin) whether you zoom after a translation, or not. Sorry about that, Cheers, Matthew From stefan.hoffmann at uni-jena.de Thu Sep 6 10:17:20 2018 From: stefan.hoffmann at uni-jena.de (Stefan Hoffmann) Date: Thu, 06 Sep 2018 16:17:20 +0200 Subject: [Neuroimaging] Affine transformations In-Reply-To: References: <20180831145127.Horde.dEbME051yHvSIMhXHdsPte9@webmail.uni-jena.de> Message-ID: <20180906161720.Horde.OJtpaZyxdOn_wSD38p78aPS@webmail.uni-jena.de> No worries, thanks for still caring after all this time. And thanks for your help in first place! best regards, Stefan Zitat von Matthew Brett : > Oops, > > On Fri, Aug 31, 2018 at 2:15 PM, Matthew Brett > wrote: > [snip] >> I think you'll find that the center doesn't matter for the zooms >> because, if Z is a zooming affine, and hence of this form: >> >> Z = [[a, 0, 0, 0], >> [0, b, 0, 0], >> [0, 0, c, 0], >> [0, 0, 0, 1]] >> >> then Tinv.dot(Z).dot(T) == Z for any translations in T. > > How silly of me. For some reason, as I woke up, I realized that's > obviously not true, and the zooms do change the translations, as would > have been obvious if I'd done the multiplication properly: > > In [22]: import sympy as sy > In [23]: a, b, c, x, y, z = sy.symbols('a, b, c, x, y, z') > In [24]: T = Matrix([[1, 0, 0, x], > ...: [0, 1, 0, y], > ...: [0, 0, 1, z], > ...: [0, 0, 0, 1]]) > ...: > In [25]: Z = Matrix([[a, 0, 0, 0], > ...: [0, b, 0, 0], > ...: [0, 0, c, 0], > ...: [0, 0, 0, 1]]) > ...: > In [26]: T.inv() * Z * T > Out[26]: > Matrix([ > [a, 0, 0, a*x - x], > [0, b, 0, b*y - y], > [0, 0, c, c*z - z], > [0, 0, 0, 1]]) > > So - yes - it does make a difference (to the origin) whether you zoom > after a translation, or not. > > Sorry about that, > > Cheers, > > Matthew > _______________________________________________ > Neuroimaging mailing list > Neuroimaging at python.org > https://mail.python.org/mailman/listinfo/neuroimaging From maxwell4444 at hotmail.com Fri Sep 7 09:40:52 2018 From: maxwell4444 at hotmail.com (Y X) Date: Fri, 7 Sep 2018 13:40:52 +0000 Subject: [Neuroimaging] Question about DIPY MAP-MRI model Message-ID: Dear DIPY developer, I am a user of DIPY and find MAP-MRI model is very useful. During my testing, I found a potential error here for non-gaussian (NG) parameter. The NG and NG perpendicular maps looks not correct as a low intensity was observed at WM area, and the NG parallel maps looks like a NG contrast. Here the data I test is using a 2 shell DWI data (b 0, 1000, 2000, 64 dir each shell), a DSI data was also test with similar result with Q4 half setting. I like the DIPY tools and MAP-MRI model here, can you help with the problem? Bests, Xu -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Figure_1.png Type: image/png Size: 103832 bytes Desc: Figure_1.png URL: From fick.rutger at gmail.com Mon Sep 10 06:45:52 2018 From: fick.rutger at gmail.com (Rutger Fick) Date: Mon, 10 Sep 2018 12:45:52 +0200 Subject: [Neuroimaging] Question about DIPY MAP-MRI model In-Reply-To: References: Message-ID: Hi Xu, Can you make the plots with the corresponding colorbars next to them? It will be easier to judge the values then. Are you following the same steps as in the dipy mapmri example to generate these plots? Rutger On Fri, 7 Sep 2018 at 15:42, Y X wrote: > Dear DIPY developer, > > I am a user of DIPY and find MAP-MRI model is very useful. During my > testing, I found a potential error here for non-gaussian (NG) parameter. > The NG and NG perpendicular maps looks not correct as a low intensity was > observed at WM area, and the NG parallel maps looks like a NG contrast. > > > > Here the data I test is using a 2 shell DWI data (b 0, 1000, 2000, 64 dir > each shell), a DSI data was also test with similar result with Q4 half > setting. > > > > I like the DIPY tools and MAP-MRI model here, can you help with the > problem? > > > > Bests, > > Xu > > > > > _______________________________________________ > Neuroimaging mailing list > Neuroimaging at python.org > https://mail.python.org/mailman/listinfo/neuroimaging > -------------- next part -------------- An HTML attachment was scrubbed... URL: From xzhen3 at wisc.edu Mon Sep 10 12:39:03 2018 From: xzhen3 at wisc.edu (Xingjian Zhen) Date: Mon, 10 Sep 2018 11:39:03 -0500 Subject: [Neuroimaging] [DIPY] Do you have tools using DTI to track the fibers? Message-ID: <0C85D84B-57E6-4295-96AA-A33FC310C257@wisc.edu> Hi dear officers, Do you have tutorials about using DTI ( shape: m x n x p x 3 x 3 ) to trace the fibers with DIPY? Thank you so much. ? Best wishes. Xingjian Zhen Computer Science, UW-Madison Medical Sciences Center, University Ave, Madison, WI 53706 xzhen3 at wisc.edu +1(608)335-9054 From Kai.Wang-1 at Colorado.EDU Tue Sep 11 14:51:42 2018 From: Kai.Wang-1 at Colorado.EDU (Kai Wang) Date: Tue, 11 Sep 2018 18:51:42 +0000 Subject: [Neuroimaging] [PySurfer] how to customize the colormap(LUT)? Message-ID: Dear PySurfer experts, I'm trying to plot 17 masks (a parcellation) on the brain surfer. Not finding the pre-defined colormaps ideal, I would like to customize the colors. I only need one color for each mask, not a range of colors. Can I specify RGBA values to use for each mask? Thank you very much, Kai Wang, Ph.D. Postdoctoral Research associate Institute of Cognitive Science, University of Colorado Boulder 1777 Exposition Dr., Boulder, CO, 80301 kai.wang-1 at colorado.edu -------------- next part -------------- An HTML attachment was scrubbed... URL: From mwaskom at nyu.edu Tue Sep 11 15:01:02 2018 From: mwaskom at nyu.edu (Michael Waskom) Date: Tue, 11 Sep 2018 15:01:02 -0400 Subject: [Neuroimaging] [PySurfer] how to customize the colormap(LUT)? In-Reply-To: References: Message-ID: Hi Kai, `Brain.add_data` accepts a list or array of colors for the colormap parameter: http://pysurfer.github.io/generated/surfer.Brain.html#surfer.Brain.add_data Best, Michael On Tue, Sep 11, 2018 at 2:51 PM, Kai Wang wrote: > Dear PySurfer experts, > > > > I?m trying to plot 17 masks (a parcellation) on the brain surfer. > Not finding the pre-defined colormaps ideal, I would like to customize the > colors. I only need one color for each mask, not a range of colors. Can I > specify RGBA values to use for each mask? > > > > > > Thank you very much, > > Kai Wang, Ph.D. > > > > Postdoctoral Research associate > > Institute of Cognitive Science, University of Colorado Boulder > > 1777 Exposition Dr., Boulder, CO, 80301 > > kai.wang-1 at colorado.edu > > > > _______________________________________________ > Neuroimaging mailing list > Neuroimaging at python.org > https://mail.python.org/mailman/listinfo/neuroimaging > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From gabriele.arnulfo at gmail.com Wed Sep 12 04:27:22 2018 From: gabriele.arnulfo at gmail.com (Gabriele Arnulfo) Date: Wed, 12 Sep 2018 10:27:22 +0200 Subject: [Neuroimaging] [PySurfer] how to customize the colormap(LUT)? In-Reply-To: References: Message-ID: Dear Kai Wang, You can pass matplotlib colormaps to Brain command. Il Mar 11 Set 2018, 20:51 Kai Wang ha scritto: > Dear PySurfer experts, > > > > I?m trying to plot 17 masks (a parcellation) on the brain surfer. > Not finding the pre-defined colormaps ideal, I would like to customize the > colors. I only need one color for each mask, not a range of colors. Can I > specify RGBA values to use for each mask? > > > > > > Thank you very much, > > Kai Wang, Ph.D. > > > > Postdoctoral Research associate > > Institute of Cognitive Science, University of Colorado Boulder > > 1777 Exposition Dr., Boulder, CO, 80301 > > kai.wang-1 at colorado.edu > > > _______________________________________________ > Neuroimaging mailing list > Neuroimaging at python.org > https://mail.python.org/mailman/listinfo/neuroimaging > -------------- next part -------------- An HTML attachment was scrubbed... URL: From anders.eklund at liu.se Sat Sep 15 07:25:30 2018 From: anders.eklund at liu.se (Anders Eklund) Date: Sat, 15 Sep 2018 11:25:30 +0000 Subject: [Neuroimaging] =?iso-8859-1?q?Postdoc_position_in_deep_learning_?= =?iso-8859-1?q?at_Link=F6ping_University=2C_Sweden?= Message-ID: Hello, I'm looking for a postdoc to work on medical image processing and deep learning of neuroimaging data (structural MRI, diffusion MRI (especially q-space trajectory imaging, QTI) and functional MRI). I'm looking for someone with experience in CNNs (convolutional neural networks) and GANs (generative adversarial networks), or someone who has a strong interest in data analysis and deep learning. Starting salary (before tax) is approximately 3500 EUR at the current EUR / SEK exchange rate. The position is for 2 years, and can be extended for 1 more year. For more information, see (apply through the ad) https://liu.se/en/work-at-liu/vacancies?rmpage=job&rmjob=9259&rmlang=UK - Anders -------------- next part -------------- An HTML attachment was scrubbed... URL: From jingweili.sjtu.nus at gmail.com Tue Sep 18 04:07:32 2018 From: jingweili.sjtu.nus at gmail.com (Jingwei Li) Date: Tue, 18 Sep 2018 16:07:32 +0800 Subject: [Neuroimaging] [fmriprep] Pick middle point of ribbon Message-ID: Dear fmriprep experts, I found that when fmriprep does projection from EPI to fsaverage space, it equally samples 6 points between pial surface and white-matter surface and averages them. If I want to only pick the middle point of the ribbon, is there a way to modify fmriprep code to achieve this? Thanks! Regards, Jingwei -------------- next part -------------- An HTML attachment was scrubbed... URL: From markiewicz at stanford.edu Tue Sep 18 08:47:12 2018 From: markiewicz at stanford.edu (Christopher Markiewicz) Date: Tue, 18 Sep 2018 12:47:12 +0000 Subject: [Neuroimaging] [fmriprep] Pick middle point of ribbon In-Reply-To: References: Message-ID: Hi Jingwei, To keep the number of channels constrained, fMRIPrep support is offered on Neurostars [0] and the GitHub issues page [1]. Best, Chris [0] https://neurostars.org/tags/fmriprep [1] https://github.com/poldracklab/fmriprep/issues/ ________________________________ From: Neuroimaging on behalf of Jingwei Li Sent: Tuesday, September 18, 2018 4:07:32 AM To: neuroimaging at python.org Subject: [Neuroimaging] [fmriprep] Pick middle point of ribbon Dear fmriprep experts, I found that when fmriprep does projection from EPI to fsaverage space, it equally samples 6 points between pial surface and white-matter surface and averages them. If I want to only pick the middle point of the ribbon, is there a way to modify fmriprep code to achieve this? Thanks! Regards, Jingwei -------------- next part -------------- An HTML attachment was scrubbed... URL: From romain.valabregue at upmc.fr Wed Sep 19 09:19:12 2018 From: romain.valabregue at upmc.fr (valabregue) Date: Wed, 19 Sep 2018 15:19:12 +0200 Subject: [Neuroimaging] non linear deformation Message-ID: Dear all Can anyone point me out some python library to apply non linear deformation to the image ? This is for data augmentation so I do not need to do a non linear registration, but just a way to apply a random smooth deformation field. Is there in nibabel support for reading deformation field (from spm or ants) and then to apply it ? Many thanks Romain From fbriend at hotmail.fr Wed Sep 19 09:36:24 2018 From: fbriend at hotmail.fr (=?iso-8859-1?Q?Fr=E9d=E9ric_Briend?=) Date: Wed, 19 Sep 2018 13:36:24 +0000 Subject: [Neuroimaging] non linear deformation In-Reply-To: References: Message-ID: Hi Romain, It seems to me that you could use the elastix registration toolbox of 3D slicer (link). Especially, you will find the python script here. Best, Frederic ________________________________ De : Neuroimaging de la part de valabregue Envoy? : mercredi 19 septembre 2018 15:19:12 ? : Neuroimaging analysis in Python Objet : [Neuroimaging] non linear deformation Dear all Can anyone point me out some python library to apply non linear deformation to the image ? This is for data augmentation so I do not need to do a non linear registration, but just a way to apply a random smooth deformation field. Is there in nibabel support for reading deformation field (from spm or ants) and then to apply it ? Many thanks Romain _______________________________________________ Neuroimaging mailing list Neuroimaging at python.org https://mail.python.org/mailman/listinfo/neuroimaging -------------- next part -------------- An HTML attachment was scrubbed... URL: From matthew.m.mccormick at gmail.com Wed Sep 19 09:39:31 2018 From: matthew.m.mccormick at gmail.com (Matthew McCormick) Date: Wed, 19 Sep 2018 09:39:31 -0400 Subject: [Neuroimaging] non linear deformation In-Reply-To: References: Message-ID: Hi Romain, ITK can help with this task: python -m pip install --upgrade pip python -m pip install itk or conda -c conda-forge install itk There are many transforms available, described here: https://itk.org/ITKSoftwareGuide/html/Book2/ITKSoftwareGuide-Book2ch3.html#x26-1170003.9 HTH, Matt On Wed, Sep 19, 2018 at 9:26 AM valabregue wrote: > Dear all > > Can anyone point me out some python library to apply non linear > deformation to the image ? > > This is for data augmentation so I do not need to do a non linear > registration, but just a way to apply a random smooth deformation field. > > Is there in nibabel support for reading deformation field (from spm or > ants) and then to apply it ? > > > Many thanks > > Romain > > _______________________________________________ > Neuroimaging mailing list > Neuroimaging at python.org > https://mail.python.org/mailman/listinfo/neuroimaging > -------------- next part -------------- An HTML attachment was scrubbed... URL: From alexis.roche at gmail.com Wed Sep 19 09:51:33 2018 From: alexis.roche at gmail.com (Alexis Roche) Date: Wed, 19 Sep 2018 14:51:33 +0100 Subject: [Neuroimaging] non linear deformation In-Reply-To: References: Message-ID: Hi Romain, If it?s just to apply a deformation field, one easy way to go is to use scipy.ndimage.map_coordinates Cheers On Wed 19 Sep 2018 at 14:26, valabregue wrote: > Dear all > > Can anyone point me out some python library to apply non linear > deformation to the image ? > > This is for data augmentation so I do not need to do a non linear > registration, but just a way to apply a random smooth deformation field. > > Is there in nibabel support for reading deformation field (from spm or > ants) and then to apply it ? > > > Many thanks > > Romain > > _______________________________________________ > Neuroimaging mailing list > Neuroimaging at python.org > https://mail.python.org/mailman/listinfo/neuroimaging > -- https://sites.google.com/site/alexisroche -------------- next part -------------- An HTML attachment was scrubbed... URL: From arokem at gmail.com Sat Sep 22 23:18:10 2018 From: arokem at gmail.com (Ariel Rokem) Date: Sat, 22 Sep 2018 20:18:10 -0700 Subject: [Neuroimaging] non linear deformation In-Reply-To: References: Message-ID: Hi Romain, Dipy has a full implementation of the symmetric diffeomorphic registration: http://nipy.org/dipy/examples_built/syn_registration_3d.html If you already have your deformation field, you can use that to initialize a dipy.align.imwarp.DiffeomorphicMap object and apply the transformation to your input image. Hope that helps, Ariel On Wed, Sep 19, 2018 at 6:51 AM Alexis Roche wrote: > Hi Romain, > > If it?s just to apply a deformation field, one easy way to go is to use > scipy.ndimage.map_coordinates > > Cheers > > On Wed 19 Sep 2018 at 14:26, valabregue wrote: > >> Dear all >> >> Can anyone point me out some python library to apply non linear >> deformation to the image ? >> >> This is for data augmentation so I do not need to do a non linear >> registration, but just a way to apply a random smooth deformation field. >> >> Is there in nibabel support for reading deformation field (from spm or >> ants) and then to apply it ? >> >> >> Many thanks >> >> Romain >> >> _______________________________________________ >> Neuroimaging mailing list >> Neuroimaging at python.org >> https://mail.python.org/mailman/listinfo/neuroimaging >> > -- > https://sites.google.com/site/alexisroche > _______________________________________________ > Neuroimaging mailing list > Neuroimaging at python.org > https://mail.python.org/mailman/listinfo/neuroimaging > -------------- next part -------------- An HTML attachment was scrubbed... URL: