Hello,

I´m trying to reconstruct a SPECT dataset with attenuation correction. I´ve successfully reconstructed images without attenuation correction but I have doubts with the attenuation map units. I have a CT derived volume with linear attenuation coefficients [cm-1] units, resampled to the SPECT matrix (64x64x64). However, reading the online help it says that the attenuation map should be adimensional. ¿Should I scale the map with exp(-mu*dx) where dx is the pixel size? Is it attenuation or transmission needed during the recon?

Thank you,
Mauro