Deep learning-based image reconstruction and post-processing methods in positron emission tomography for low-dose imaging and resolution enhancement.

Image processing plays a crucial role in maximising diagnostic quality of positron emission tomography (PET) images. Recently, deep learning methods developed across many fields have shown tremendous potential when applied to medical image enhancement, resulting in a rich and rapidly advancing literature surrounding this subject. This review encapsulates methods for integrating deep learning into PET image reconstruction and post-processing for low-dose imaging and resolution enhancement. A brief introduction to conventional image processing techniques in PET is firstly presented. We then review methods which integrate deep learning into the image reconstruction framework as either deep learning-based regularisation or as a fully data-driven mapping from measured signal to images. Deep learning-based post-processing methods for low-dose imaging, temporal resolution enhancement and spatial resolution enhancement are also reviewed. Finally, the challenges associated with applying deep learning to enhance PET images in the clinical setting are discussed and future research directions to address these challenges are presented.

Monte Carlo calculated calibration settings for commercially available nuclear medicine ionization chambers.

PURPOSE: A method for calculating nuclear medicine ionization chamber (NMIC) calibration settings with a Monte Carlo model is presented and validated against physical measurements. This work provides Monte Carlo-calculated calibration settings for select isotopes with no current manufacturer recommendations and a method by which NMIC manufacturers or standards laboratories can utilize highly detailed specifications to calculate comprehensive lists of calibration settings for general isotopes. METHODS: A Monte Carlo model of a Capintec PET series NMIC was developed and used to calculate the chamber response to relevant radioactive decay products over an energy range relevant to nuclear medicine. The photon detection efficiency (PDE) of a high purity germanium (HPGe) detector was modeled and physically validated to facilitate measurements of NMIC calibration settings with HPGe detector spectroscopy. Modeled NMIC response to various isotopes was compared against spectroscopic measurements and National Institute of Standards and Technology (NIST)-validated calibration settings to validate the Monte Carlo-calculated NMIC calibration settings. RESULTS: HPGe detector PDE was validated against the physical measurements to within 3.3 % $3.3\%$ at 95 % $95\%$ confidence and used to measure calibration settings, which produced activity readings 0.7 % $0.7\%$ , 1.6 % $1.6\%$ , 0.8 % $0.8\%$ , and 1.0 % $1.0\%$ different than those validated by NIST for 11 $^{11}$ C, 18 $^{18}$ F, 68 $^{68}$ Ga, and 64 $^{64}$ Cu respectively. The Monte Carlo model of the NMIC reproduced measured calibration settings to within 7 % $7\%$ at 95 % $95\%$ confidence for isotopes with a sufficiently small yield of low energy photons. CONCLUSIONS: A method of calculating NMIC calibration settings with Monte Carlo modeling has been developed and validated against HPGe detector spectroscopy. NMIC manufacturers or standards laboratories can use more detailed specifications of the chamber geometries to extend the applicability of this method to a wider range of isotopes.