Question: Any methods to reverse the EIT reconstruction ?

[ChabaneAmaury]

The question is self-explanatory, I would ideally find the voltage measurements back based on the calculation of the Gauss-newtonian reconstruction.

I am not sure whether it is possible or not, so if not I will try and find a workaround.

[liubenyuan]

About GN reconstructions, are you refer to one-step GN (time difference imaging)? Find $x$ based on $y$ where $y=Hx$. Or you mean that find x from an EIT image?

[ChabaneAmaury]

I mean find $x$ based on $y=Hx$

[liubenyuan]

H is the imaging matrix, x and x0 are the measurement vector at timestamp now and ref, dv=(x-x0)/x0 is the normalization method we use on time-difference imaging. y is the tdEIT image. y = H*dv, pinv(H) is the pseudo inverse of H.

[ChabaneAmaury]

Ok thank you very much I will try later to see if I can do something about this

[ChabaneAmaury]

So I ran some tests, and it turns out the proposed losses doesn't work....they are too small to have a significant impact, unfortunately. I think I will just stick with my regular MSE.

[liubenyuan]

ok!

[ChabaneAmaury]

Can you mark this discussion as solved?