/Poisson_Likelihood_Loss_pytorch

This is a PyTorch implementation of the Poisson Likelihood loss function used for NM image reconstruction.

Primary LanguagePythonMIT LicenseMIT

Poisson Likelihood loss function

This is a PyTorch implementation of the Poisson Likelihood loss function used for NM image reconstruction [1].

, where n is the number of bins, g bar represents the mean (target), and g is the observation.

[1] Bruyant, P. P. (2002). Analytic and iterative reconstruction algorithms in SPECT. Journal of Nuclear Medicine, 43(10), 1343–1358.