k-Wave

1. Irsyadul
   

   Member
   Joined: Nov '16
   Posts: 2

   Hi Dr. Cox and Dr. Treeby
   I am studying MAT-CI (Magnetoacoustic Tomography - Current Injection). I would like to know if K-Wave can be used to reconstruct the image from MAT-CI experimental data, where the experimental data consists of acoustic-pressure and acquired time as shown in the file at the following link?

   https://www.dropbox.com/s/66gydtwdneqddzd/acq0002.txt?dl=0
   https://www.dropbox.com/s/91r1f9asdkj431o/Data%20Plotting.jpg?dl=0

   I use the following equipment:

   • A copper ring as a sample with 3cm outer diameter and 2cm inner diameter
   • Unfocused ultrasonic transducer with 13mm element size and 1 MHz central frequency (but can be focused with distance to target 0.6 - 0.8 inches)
   • Radius transducer 15cm with 1.8 degree each step (so 200 points of measurement around the sample).

   I have seen examples on K-Wave (example_pr_2D_TR_circular_sensor). If the example can be used can you provide a guide how to modify it, especially to construct the time matrix and the corresponding data sensor?

   Your reply will be appreciated! thanks a lot!

   Posted 7 years ago #

2. bencox
   

   Developer
   Joined: Mar '10
   Posts: 292

   Hi Irsyadul,

   The image reconstruction examples in the k-Wave toolbox are for photoacoustic tomography (PAT) in which an impulsive (very short) excitation pulse is used and the detectors are assumed to be broadband and omnidirectional. The image reconstruction problem then becomes an inverse initial value problem, and techniques based on 'time reversal' work well. I've just looked through this paper on MAT-CI: Liu, G., Huang, X., Xia, H. et al. Chin. Sci. Bull. (2013) 58: 3600, and the formulae there suggest that MAT-CI is very similar to PAT in its formulation. However, there are several issues you need to think about.

   If you don't have a short pulsed source, but you know it's temporal shape, you could try deconvolving it from the measured time series before reconstructing. Also, you may also have to take into account the impulse response function of your detectors. Finally, your detector elements seem quite large, so I suspect their response is quite directional. In fact, they may be so directional that you can get away with backprojecting along straight lines normal to the detector elements.

   Best wishes,
   Ben

   Posted 7 years ago #

3. Irsyadul
   

   Member
   Joined: Nov '16
   Posts: 2

   Thanks for your reply Ben,

   Yes, I inspired by that paper, but I use a different approach, where in my experiment I use continuous DC signal as excitation and the experiment conducted in free air, not in distilled water like in that paper.

   For comparative study, if I use very short excitation pulse in the experiment and do it in distilled water like in that paper, Can I use K-Wave to reconstruct the image or still I must deconvolve it first then use K-Wave or I can't use it at all??

   For your assistance, thank you.
   Irsyadul

   Posted 7 years ago #

RSS feed for this topic

Reply

You must log in to post.