From zeph.phillips at gmail.com Thu Apr 12 23:01:26 2018 From: zeph.phillips at gmail.com (Zeph Phillips) Date: Fri, 13 Apr 2018 12:01:26 +0900 Subject: [Nirfast] Inversed Values for Jacobian Matrix Message-ID: When I try to construct the Jacobian matrix for continuous wave DOT, I find that my values are inversed than what I expect, according to literature and other figures (if we assume that the each value is the log amplitude change/log ua changes). Thus to get the correct DOT image, I have to multiply these values by a negative one value. Has anyone run into this issue? Perhaps I am constructing the Jacobian matrix inappropriately or interpreting it wrong? Jacobian Construction: J = jacobian(mesh,0,850); Optical Properties: val.mua=0.018; val.mus=0.69; val.ri=1.3; -- ______________________ Kind Regards, *Zeph Phillips* *zeph.phillips at gmail.com* -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.png Type: image/png Size: 8193 bytes Desc: not available URL: From h.dehghani at cs.bham.ac.uk Fri Apr 13 07:03:15 2018 From: h.dehghani at cs.bham.ac.uk (Hamid Dehghani) Date: Fri, 13 Apr 2018 12:03:15 +0100 Subject: [Nirfast] Inversed Values for Jacobian Matrix In-Reply-To: References: Message-ID: <66e33c27-17d8-fe59-aff4-0c609bdc991f@cs.bham.ac.uk> Dear Zeph Are you using a spectral mesh? If so, then the Jacobian is in fact Log amplitude/ua and not Log amplitude change/log ua If you are using a single wavelength mesh (standard) than when you use the Jacobian function, it gives you two values for J J.complex = is the amplitude/mua or kappa J.complete = is the log amplitude/mua or kappa The plot you show, appears correct to me. Hamid On 13/04/2018 04:01, zeph.phillips at gmail.com wrote: > When I try to construct the Jacobian matrix for continuous wave DOT, I > find that my values are inversed than what I expect, according to > literature and other figures (if we assume that the each value is the > log amplitude change/log ua changes). Thus to get the correct DOT image, > I have to multiply these values by a negative one value. > > Has anyone run into this issue? Perhaps I am constructing the Jacobian > matrix inappropriately or interpreting it wrong? > > Jacobian Construction: > J = jacobian(mesh,0,850); > > Optical Properties: > val.mua=0.018; > val.mus=0.69; > val.ri=1.3; > > > > > -- > ______________________ > Kind?Regards, > *Zeph Phillips* > /zeph.phillips at gmail.com/ > > > _______________________________________________ > Nirfast mailing list > Nirfast at public.kitware.com > https://public.kitware.com/mailman/listinfo/nirfast > -- -------------------------- Hamid Dehghani, PhD, MIPEM Professor of Medical Imaging School of Computer Science Co-Director, Physical Sciences for Health Doctoral Training Centre University of Birmingham B15 2TT 0121 414 8728 https://www.cs.bham.ac.uk/~dehghanh/mi-lab.html