k-Wave

1. Alexander
   

   Member
   Joined: Jun '17
   Posts: 19

   Hi,

   I wanted to clarify if the implementation of time reversal that k-wave uses can compensate for the affects of heterogeneous speed of sound? I have a back projection algorithm for PAT that can compensated for that and was looking to compare it to time reversal.

   When I did a quick test, which just involved defining a heterogeneous speed of sound medium and homogeneous density. It appears that the time reversal algorithm doesn't compensate for sos, unless I am missing a flag or that density should also be heterogeneous.

   If that is true presumably there is a term c0 (speed of sound) when you solve the differential equations so when we solve the equation at a given point could we use the associated speed of sound of that point to compensate for its affects? Perhaps that doesn't give a perfect solution, but would it at least be a good 1st order approximation?

   Posted 4 years ago #

2. Bradley Treeby
   

   Developer
   Joined: Mar '10
   Posts: 1,640

   Hi Alexander,

   You can perform time reversal with a heterogeneous sound speed. However, you need to consider the effects of artefact trapping.

   Hope that helps,

   Brad.

   Posted 4 years ago #

3. Alexander
   

   Member
   Joined: Jun '17
   Posts: 19

   Thanks Dr. Treeby, I'll take a look at the attached paper.

   But time reversal is able to compensate already then? Or it just take the mean value of the heterogeneous map? If it is compensating then I'll take a look at my scripts from before and see if it was an error on my part.

   Thanks,
   Alex

   Posted 4 years ago #

4. Bradley Treeby
   

   Developer
   Joined: Mar '10
   Posts: 1,640

   Hi Alex,

   Yes, time reversal can take a heterogeneous sound speed map. Essentially, it runs the same code as the forward model, but flips the sensor data in time and uses source.p_mode = 'dirichlet'.

   Brad.

   Posted 4 years ago #

5. Lin
   

   Member
   Joined: Jan '21
   Posts: 2

   Hi Alexander,

   I'm new in photoacoustic tomography, and I'm learning how to use back projection algorithm to compensate for the affects of heterogeneous speed of sound, could you please share your code?

   Thanks!

   Posted 3 years ago #

RSS feed for this topic

Reply

You must log in to post.