| Resumen |
In this work, we propose a 3D convolutional neural network (CNN) for positron emission tomography (PET) image enhancement as an application of the artificial intelligence (AI) in the area of health. Our proposed network manages to increase the number of counts in the PET sinograms, thus, positively influencing the final quality of the reconstructed image. The enhanced sinogram, obtained by the network, is reconstructed using the ordered subset expectation maximization (OSEM) algorithm. The results show that the proposed network is able to increase the PSNR by 6% on average and the contrast almost twice. |
