Deep learning transforms standard CT scans towards spectral images – Physics World – IAM Network

Top row: benchmark virtual monoenergetic (VM) images from dual-energy CT projection data reconstructed at 80 and 110 keV, respectively. Bottom row: corresponding VM images produced at the same energies from only 140-kVp images, using a deep learning approach. Conventional clinical CT scans generate a spectrally integrated attenuation image that shows tissue morphology, but does not directly provide any information regarding tissue composition. Dual-energy CT (DECT) systems, which acquire two spectrally distinct datasets, can reconstruct virtual monoenergetic (VM) and material-specific images that provide information about tissue composition. Compared with conventional CT, however, DECT is more expensive and complex, and often requires an increased radiation dose.