Dear Anil,
Unfortunately I don't have any documentation on that. I think the only way to get this information is to look at the code. The link I sent you already is probably a good place to start. Best, Peter ________________________________ Von: Shetty, Anil <ashe...@bcm.edu> Gesendet: Dienstag, 29. Oktober 2019 16:32 An: Neher, Peter Betreff: RE: [mitk-users] IVIM module Good morning, Peter: I have used IVIM module in MITK diffusion for analyzing placenta. I have been able to compare 6 different ways provided in your drop down menu for “choose methods”. I was wondering if you would kindly provide me with methodology behind each of this method. In particular, I am interested in type of algorithm (Lavenberg-Marquardt) etc. that is being used and if any boundary conditions used in the code. The 6 ways that I have used are: 1. 3 parameter fit 2. Fit D & f with fixed D* value 3. Fit D & f (high b), then fit D* a. b > 80 b. b > 250 4. Linearly fit D and f (high b), the fit D* a. b > 80 b. b > 250 I need some write up on methods 1, 2, 3 and 4 and any conditions or boundary values set up in using those models. Greatly appreciate your help. Anil N. Shetty, Ph.D. Associate Professor Baylor College of Medicine Department of Obstetrics and Gynecology Pavilion for women Texas Children's Hospital 6651 Main Street/Suite F1040-18 Houston, TX 77030 832-826-7312 (Direct) 832-825-9351 (Fax) From: Neher, Peter [mailto:p.ne...@dkfz-heidelberg.de] Sent: Friday, September 27, 2019 2:40 AM To: Shetty, Anil <ashe...@bcm.edu> Subject: AW: [mitk-users] IVIM module There are no dedicated builds for Windows 7 but the Windows 10 installers should also work. Peter ________________________________ Von: Shetty, Anil <ashe...@bcm.edu<mailto:ashe...@bcm.edu>> Gesendet: Donnerstag, 26. September 2019 21:55 An: Neher, Peter Betreff: RE: [mitk-users] IVIM module Hi Peter: Do you have link for windows 7? The link you sent are for Windows 10. Thanks anil From: Neher, Peter [mailto:p.ne...@dkfz-heidelberg.de] Sent: Thursday, September 26, 2019 2:06 AM To: Shetty, Anil <ashe...@bcm.edu<mailto:ashe...@bcm.edu>> Subject: AW: [mitk-users] IVIM module This link is not working? ftp://ftp.dkfz-heidelberg.de/outgoing/MitkDiffusion/2019.09.26/<https://urldefense.proofpoint.com/v2/url?u=ftp-3A__ftp.dkfz-2Dheidelberg.de_outgoing_MitkDiffusion_2019.09.26_&d=DwMFAw&c=ZQs-KZ8oxEw0p81sqgiaRA&r=xunJOi6tH8YusoQF3t9EDA&m=ifEqsVfzliQFbnYQFSD4Q1MBaHnc_1Q8FQ3xOPhNS2I&s=nHa3qPMwJynHwv6NAjus3YxRWnqfTsEh1GMmSudJzaY&e=> ________________________________ Von: Shetty, Anil <ashe...@bcm.edu<mailto:ashe...@bcm.edu>> Gesendet: Mittwoch, 25. September 2019 23:31 An: Neher, Peter Betreff: RE: [mitk-users] IVIM module Hi Peter: I am trying to download your latest version for MITKDiffusion on windows 7 but not able to. Can you provide us with a good link to download the same please? Thank you, Anil From: Neher, Peter [mailto:p.ne...@dkfz-heidelberg.de] Sent: Tuesday, September 24, 2019 1:45 AM To: Shetty, Anil <ashe...@bcm.edu<mailto:ashe...@bcm.edu>>; mitk-users@lists.sourceforge.net<mailto:mitk-users@lists.sourceforge.net>; Maier-Hein, Klaus H. <k.maier-h...@dkfz-heidelberg.de<mailto:k.maier-h...@dkfz-heidelberg.de>> Subject: AW: [mitk-users] IVIM module I'm not sure about that. I know that quite some papers were published using the IVIM methods implemented in MITK, but I don't know about a comparative study of the different fits. Do you know about anything like that, Klaus? Peter ________________________________ Von: Shetty, Anil <ashe...@bcm.edu<mailto:ashe...@bcm.edu>> Gesendet: Montag, 23. September 2019 17:15 An: Neher, Peter; mitk-users@lists.sourceforge.net<mailto:mitk-users@lists.sourceforge.net> Betreff: RE: [mitk-users] IVIM module Thank you very much Peter, for clarification. I do appreciate your timely response. Has there been any study done to compare different methods you propose in IVIM analysis? How do you determine which method provides the best estimation of D/D*/f? I mean between (3 parameter fit vs. fixed D* then D and f etc,.) Thank you for your help Regards, Anil N. Shetty, Ph.D. Associate Professor Baylor College of Medicine Department of Obstetrics and Gynecology Pavilion for women Texas Children's Hospital 6651 Main Street/Suite F1040-18 Houston, TX 77030 832-826-7312 (Direct) 832-825-9351 (Fax) From: Neher, Peter [mailto:p.ne...@dkfz-heidelberg.de] Sent: Monday, September 23, 2019 2:05 AM To: Shetty, Anil <ashe...@bcm.edu<mailto:ashe...@bcm.edu>>; mitk-users@lists.sourceforge.net<mailto:mitk-users@lists.sourceforge.net> Subject: AW: [mitk-users] IVIM module ***CAUTION:*** This email is not from a BCM Source. Only click links or open attachments you know are safe. ________________________________ Dear Anil, only the weighted measurements are plotted, i.e. b>0. But anyway, only the scale, so the positions on the y-axis, is logarithmic, no the shown values. So you plot ln(s/s0), but the axis label still shows s/s0. Just a matter of visualization. Peter ________________________________ Von: Shetty, Anil <ashe...@bcm.edu<mailto:ashe...@bcm.edu>> Gesendet: Samstag, 21. September 2019 12:59 An: Neher, Peter Betreff: Re: [mitk-users] IVIM module Hi Peter: Thank you for your help with IVIM module. BTW, I have slight confusion here. If the signal is plotted with "Ln(S/S0)"on the y-axis, the first point at b =0 should be =0, since S/S0 =1 at b =0. How do you reverse it to show max on y-axis =1 at b =0 in your plot? S = S0exp(-bD) mono-exponential at low b-values. Ln(S/S0) = -bD >From plot at b =0, L(s/S0) = 0 Am I missing something? Thanks Anil ________________________________ From: Neher, Peter <p.ne...@dkfz-heidelberg.de<mailto:p.ne...@dkfz-heidelberg.de>> Sent: Friday, September 20, 2019 02:14 AM To: Shetty, Anil <ashe...@bcm.edu<mailto:ashe...@bcm.edu>>; mitk-users@lists.sourceforge.net<mailto:mitk-users@lists.sourceforge.net> <mitk-users@lists.sourceforge.net<mailto:mitk-users@lists.sourceforge.net>> Subject: AW: [mitk-users] IVIM module Or rather S/S_0, but the scale is logarithmic. ________________________________________ Von: Neher, Peter <p.ne...@dkfz-heidelberg.de<mailto:p.ne...@dkfz-heidelberg.de>> Gesendet: Freitag, 20. September 2019 08:56 An: Shetty, Anil; mitk-users@lists.sourceforge.net<mailto:mitk-users@lists.sourceforge.net> Betreff: Re: [mitk-users] IVIM module ln(s) I will see to it that this will be indicated in the plot. Peter ________________________________________ Von: Shetty, Anil <ashe...@bcm.edu<mailto:ashe...@bcm.edu>> Gesendet: Donnerstag, 19. September 2019 17:27 An: Neher, Peter Betreff: RE: [mitk-users] IVIM module Hi Peter: One more thing. The plot in figure, is that "Signal vs b" or "Ln(Signal) vs b" ? Thanks -Anil -----Original Message----- From: Neher, Peter [mailto:p.ne...@dkfz-heidelberg.de] Sent: Thursday, September 19, 2019 8:47 AM To: Shetty, Anil <ashe...@bcm.edu<mailto:ashe...@bcm.edu>>; mitk-users@lists.sourceforge.net<mailto:mitk-users@lists.sourceforge.net>; Götz, Michael <m.go...@dkfz-heidelberg.de<mailto:m.go...@dkfz-heidelberg.de>>; Jäger, Paul <p.jae...@dkfz-heidelberg.de<mailto:p.jae...@dkfz-heidelberg.de>> Subject: AW: [mitk-users] IVIM module Dear Anil, If I look at the code, the method actually does a 3-parameter LM fit: https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_MIC-2DDKFZ_MITK-2DDiffusion_blob_master_Modules_DiffusionCore_Algorithms_Reconstruction_itkDiffusionIntravoxelIncoherentMotionReconstructionImageFilter.cpp-23L447&d=DwIFAw&c=ZQs-KZ8oxEw0p81sqgiaRA&r=xunJOi6tH8YusoQF3t9EDA&m=RsPQxKza4BsCqlriISy6MlMZ1ZZNLO8-ysVRB2b_koY&s=7rs0ek8jQUqY_tKDw4FmWBioHKPqbVI2nz7tkl14xyM&e= I am not sure where the output on the clipboard originates from. Do you know anything about this, Michael & Paul? Peter ________________________________________ Von: ashetty <ashe...@bcm.edu<mailto:ashe...@bcm.edu>> Gesendet: Mittwoch, 18. September 2019 18:55 An: mitk-users@lists.sourceforge.net<mailto:mitk-users@lists.sourceforge.net> Betreff: [mitk-users] IVIM module I have been using IVIM module to determine f, D and D* using 3-parameter fit. However, when I copy measurement points to excel, I see that it does first linear fit of D and f at high b, and then, fits for D*. Should this not do 3 parameters simultaneously using L-M method when I choose option of 3-parameter fit? Appreciate your answer. _anil -- Sent from: https://urldefense.proofpoint.com/v2/url?u=http-3A__mitk-2Dusers.1123740.n5.nabble.com_&d=DwIFAw&c=ZQs-KZ8oxEw0p81sqgiaRA&r=xunJOi6tH8YusoQF3t9EDA&m=RsPQxKza4BsCqlriISy6MlMZ1ZZNLO8-ysVRB2b_koY&s=tRN1QFquy_aNi79SNJJDlv8XtYG0IK5q3-k1LYQnzU8&e= _______________________________________________ mitk-users mailing list mitk-users@lists.sourceforge.net<mailto:mitk-users@lists.sourceforge.net> https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.sourceforge.net_lists_listinfo_mitk-2Dusers&d=DwIFAw&c=ZQs-KZ8oxEw0p81sqgiaRA&r=xunJOi6tH8YusoQF3t9EDA&m=RsPQxKza4BsCqlriISy6MlMZ1ZZNLO8-ysVRB2b_koY&s=8d-vo07FfYJigUbg-VF3z8WO1PTdEs1EifBqZ_xf17Q&e= _______________________________________________ mitk-users mailing list mitk-users@lists.sourceforge.net<mailto:mitk-users@lists.sourceforge.net> https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.sourceforge.net_lists_listinfo_mitk-2Dusers&d=DwIFAw&c=ZQs-KZ8oxEw0p81sqgiaRA&r=xunJOi6tH8YusoQF3t9EDA&m=WM7OiABjV744a2ejcTzW2Pt0Gmq5WD9nHV3cdqDDLyk&s=G7JXaFm6cQSZxS-18uPRBs0jnIqcc__a6vvLb77tpco&e=
_______________________________________________ mitk-users mailing list mitk-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mitk-users
