Weighted momentum transfer DCS jacobian

29 views
Skip to first unread message

IRENE

unread,
Apr 20, 2026, 6:15:31 AMApr 20
to mcx-users
Dear Dr.Fang, 

I am doing diffuse correlation tomography (DCT) with mcxlab. I am trying to compute the DCS Jacobian by using the weighted momentum transfer Jacobian with the MCX 'replay' mode. 

Can I ask what the meaning is of 'weighted' here? 

Best wishes,
Xingmin 

Qianqian Fang

unread,
Apr 24, 2026, 10:24:48 PMApr 24
to mcx-...@googlegroups.com, IRENE

hi Irene,

the DCS replay was my previous attempt of extending the replay-based Jacobian towards momentum transfer and DCT back in 2018 - despite at the time, and still, I don't have experience with DCT. 

Frankly, I don't really know if this saved quantity could be used for DCT reconstruction at all, so I am happy that you paid attention to it, perhaps you can let me know if this is in the right direction.

what otDCT output actually saves is w_p*(1-cos(theta)) after every scattering event of the replayed photon, where 

w_p is the detected weight of the p-th replayed photon, and

theta is the zenith/pole angle of the scattering; (1-cos(theta)) is the momentum transfer of the scattering event.


this formula attempts to mimic the replay-Jacobian of mua, i.e. Eq. 8 of the replay paper

https://opg.optica.org/boe/fulltext.cfm?uri=boe-9-10-4588

where J_mua = -\bar{L} = -sum(w_p * L_p,v)/sum(w_p); when building J_mua, the replay phase accumulates w_p*L in each voxel - a weighted back-projection of the derivative to absorption at the detector to the volume. 

similarly, I imagined that by accumulating w_p*(1-cos(theta))  during the replay, I could also create a "back-projection" like operator to the voxels where scattering event happens. However, I did not do any derivation nor did I specify what quantity was the derivative taken against.

please let me know if this quantity has any potential use for what you want to do, if not, what would be a sensible quantity to output?

Qianqian

--
You received this message because you are subscribed to the Google Groups "mcx-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mcx-users+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/mcx-users/c2bbee46-84b6-46cd-99d4-f7aa0b157185n%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages