Photon-counting computed tomography (PCCT) represents a paradigm shift in computed tomography imaging by enabling direct detection and energy discrimination of individual X-ray photons, as opposed to energy integration in conventional computed tomography systems. This review provides a comprehensive overview of the technical principles, advantages, limitations, and clinical applications of photon-counting computed tomography in modern radiology. Technically, photon-counting computed tomography employs semiconductor detectors to directly convert photons into electrical signals and categorize them based on predefined energy thresholds, enabling continuous spectral imaging and material-specific characterization, including K-edge imaging. Compared with conventional computed tomography and dual-energy computed tomography, photon-counting computed tomography offers superior spatial resolution, improved dose efficiency, reduced electronic noise, and enhanced material decomposition capabilities. Clinically, photoncounting computed tomography has demonstrated significant advantages across cardiovascular, pulmonary, neurological, musculoskeletal, and oncologic imaging, particularly in plaque characterization, interstitial lung disease assessment, and iodine quantification. Despite its promising performance, challenges such as high system cost, technical limitations, and lack of standardization remain. Future developments are expected to establish photon-counting computed tomography as a cornerstone of precision imaging, particularly with the integration of artificial intelligence and advanced multi-contrast agents.