Photon counting CT is a new technology that enables us to improve the quality of images by a conventional CT, in which the detection of transmitted photons is conducted by an integration of photon energies. This paper describes the features and advantages of the photon counting detection compared to the energy integration detection. They are (1) reduction of electrical noise and improvement of the signal-to-noise ratio, (2) improvement of the image contrast using the weighting factors to images acquired with energy bins, (3) k-edge imaging by setting two energy bins at the k-edge of some contrast media such as gadopenteto megruminacid and gold-nanoparticles, and (4) material decomposition using the data acquired with multiple energy windows. For the material decomposition, the principal component analysis, singular value decomposition method, and the application of the artificial neural network are described. The photon counting CT technique has a potential to improve the diagnostic accuracy and introduce new clinical methods, however, much efforts are required to use this technology in the clinical situation.