Hi, sorry to bother you. Do you happen to know how to calculate the current absorbed by the electrodes and the impedance?

I ran the simulation again. This time it ran completely. Physical results the same but without any errors, i.e. the analysis section could indeed connect ports. So is there a random seed thing and a voxeling issue after all?

Please select the overall field and select the E field, then look under "Field Data Tools". Please "Go to the HTML manual" from "Help" and search for NaN filter. I would recommend using the latest version of Sim4Life v9.0. [image: 1753361215835-bc62f2dd-580a-4b6b-b11d-728454d14416-sim4life_xhh8cgbkwm.png]

You could probably export (XPostProcessor.VtkExporter or XPostProcessor.VtkXmlExporter, not sure it is in s4l_v1) it as a VTK file and modify your neural network DataLoader to support VTK files, e.g., via the meshio package or the much larger vtk package, or many others. Or you could write a little Python script to grab the field data and save it to npy. To access the field, you would get the output of the extractor or pipeline entity in the analysis, say a FloatFieldData object. It has a Field(self, index: int) -> np.ndarray method to get the field as a numpy array, which you could save using e.g., np.save("filename.npy", fd.Field(0))

Hi, Sorry to bother You. IN ATTACHED VIDEO SHOWS WHAT HAPPENS AFTER CLICKING THE OPTION "CREATE PLOT" AND THE SYSTEM CRASHES. The workstation and memory have been tested, all without errors. Can one of the service engineers help? Kind regards, Thank you, Andrei Andrei Churakov MD,PhD, Associate Professor of the Department of Electronic Engineering and Technology BSUIR E-mail: anchurakov@bsuir.by

This question is quite old. Maybe you found a solution? Otherwise I guess you could use one of these options: you can crop the field, e.g. via the bounding box (or the extent if you have an E-field on a rectilinear grid) you could use the interpolation filter: define your little box grids, and interpolate from the field sensor field to your box grids you could define small field sensors in the simulation setup (not sure it is scalable to hundreds of sensors)

Hi @parsley, Thank you for your feedback and for reporting the issue you experienced. The MATCH tool has already been successfully used/validated by many users and offers several useful features. For example, it allows users to add loading or matching circuits the S-matrix of a simulation and obtain updated scattering parameters without re‑running the full simulation. Have you experienced any problems with this particular workflow? From the information you provided, the issue appears to be related to only the Initial Matching function—the option that generates a matching circuit for a given S-matrix of a simulation at a chosen target frequency. In your case, you reported that it worked with the dipole antenna example but not with your specific antenna model. We would be happy to discuss the details so we can reproduce the problem and resolve it quickly. We will contact you to find out a convenient time to follow up. Thank you again for bringing this to our attention.

[image: 1736961843959-327c9ace-0099-4e59-b102-4a34c31e609c-sim4life_fxfwepw8sg.gif] to extract the inner normal vector to the cortical surface mask the field (set NaN outside GM) using the Mask Filter available under "Field Data Tools" in the top ribbion. add the Grey Matter Surface in the post-pro interpolate the (masked) field to the grey matter surface using the "Interpolator" which is active when the masked field and surface is selected. get the normal component via the NormalFieldEvaluator under the "Field Data Tools"

@bryn @gc00 @halder Thank you so much, You've been very helpful !

@ofli Can the conductivity map after head segmentation be exported?

Hi, you can get additional information by enabling the "Far Field" sensor and then choosing the "Radiation Report" evaluator (see for example the Phased Array Patch Antenna tutorial). It will not solve your problem, but might help you get a clearer picture. From what you describe, it's possible you have a very low mismatch efficiency (most of the energy is reflected). This could be addressed by improving the way you are modeling your source, or by just ignoring it if it is not important (i.e. if you are not actually simulating the real feed and are only interested in the antenna itself). You would do that by considering the Radiation Efficiency (ratio of Radiated Power to Accepted Power) instead.

Hello! Thank you for your quick answer :) I'll do that in the meantime