ABSTRACT
Nuclear medicine uses radioactively labeled pharmaceuticals to diagnose and assess disease in the human body by single photon emission computed tomography (SPECT) and positron emission tomography (PET). The image reconstruction consists in retrieving the 3-D spatial distribution of the radiopharmaceutical from the projection data acquired under different angles. Monte Carlo methods are statistical simulations using a sequence of random numbers to model different physical processes. The Monte Carlo method has been applied in nuclear medicine for detector design, analyzing quantification issues, correction methods for image degradations, and detection tasks. With the market of affordable miniclusters rapidly gaining importance, there has been a renewed interest in using Monte Carlo simulations for image reconstruction. This method allows for a patient-specific simulation of the data acquisition, resulting in a more accurate quantification when incorporated in an iterative reconstruction algorithm. Quantitative image reconstruction ultimately enhances diagnosis, dosimetry, and treatment planning.
