Ge Precision Healthcare LLC

Leland Stanford Junior University

Yirong Yang

Jiahua Fan

Brody DeSilva

Carrigan Hudgins

Systems/techniques that facilitate generation of image denoising training data via photon-count splitting are provided. In various embodiments, a system can access a set of sinograms generated by a photon-counting computed tomography scanner. In various aspects, the system can split the set of sinograms into a first reduced-photon-count set of sinograms and a second reduced-photon-count set of sinograms. In various instances, the system can convert, via image reconstruction, the first reduced-photon-count set of sinograms into at least one training input image and the second reduced-photon-count set of sinograms into at least one training output image. In various cases, the system can train a deep learning neural network based on the at least one training input image and the at least one training output image.

Generation of image denoising training data via photon-count splitting

Methods and systems are provided for reconstructing images within a photon-counting computed tomography (PCCT) system. In an example, a method comprises, during a scan of an imaging subject, obtaining photon counts from a detector element of a photon-counting detector of the PCCT system, the photon counts partitioned into a plurality of energy bins based on an energy imparted by each photon on the detector element; encoding the photon counts at the plurality of energy bins of the detector element into a single scalar output value, the single scalar output value representing a distribution of spectral information across the energy bins; and reconstructing an image from projection data acquired via the photon-counting detector, the projection data including the single scalar output value generated at the detector element; wherein a basis material decomposition process is not performed during image reconstruction.

Growing community of inventors

Stanford, CA, United States of America

Yirong Yang