Canon Medical Systems Corporation

Ruoqiao Zhang

Zhou Yu

Jian Zhou

Xiaohui Zhan

Yan Liu

Xiaofeng Niu

Qiulin Tang

Ilmar Hein

An X-ray scanner system is provided. The system includes a photon-counting detector, a memory, and processing circuitry. The detector has a plurality of detector pixels in a channel direction. The memory stores a detector response forward model of the photon-counting detector. The detector response forward model is to be used during image reconstruction of an imaging object. The processing circuitry estimates an attenuation profile of the imaging object, determine, with respect to each of the plurality of detector pixels, a set of spatial weights, based on the estimated attenuation profile, and update, based on the determined set of spatial weights, the detector response forward model stored in the memory.

Detector response calibaration data weight optimization method for a photon counting x-ray imaging system

A photon counting computed tomography (CT) method including receiving a first forward model including a set of first parameters and a set of second parameters corresponding to a plurality of materials and path lengths by scanning a slab at plural tube voltages and plural tube currents of an X-ray tube; evaluating an image quality of a material decomposition image reconstructed by the set of first parameters and the set of second parameters; and updating at least one second parameters from the set of second parameters if the image quality of the material decomposition image does not satisfy a predetermined threshold, wherein the updating is achieved by updating the at least one second parameter from the set of second parameters to an energy dependent parameter from a constant value.

Iterative method for material decomposition calibration in a full size photon counting computed tomography system

Data acquired from a scan of an object can be decomposed into frequency components. The frequency components can be input into a trained model to obtain processed frequency components. These processed frequency components can be composed and used to generate a final image. The trained model can be trained, independently or dependently, using frequency components covering the same frequencies as the to-be-processed frequency components. In addition, organ specific processing can be enabled by training the trained model using image and/or projection datasets of the specific organ.

Projection based deep learning with frequency splitting for computed tomography

A deep learning (DL) network corrects/performs sinogram completion in computed tomography (CT) images based on complementary high- and low-kV projection data generated from a sparse (or fast) kilo-voltage (kV)-switching CT scan. The DL network is trained using inputs and targets, which respectively generated with and without kV switching. Another DL network can be trained to correct sinogram-completion errors in the projection data after a basis/material decomposition. A third DL network can be trained to correct sinogram-completion errors in reconstructed images based on the kV-switching projection data. Performance of the DL network can be improved by dividing a 3D convolutional neural network (CNN) into two steps performed by respective 2D CNNs. Further, the projection data and DLL can be divided into high- and low-frequency components to improve performance.

Growing community of inventors

Arlington Heights, IL, United States of America

Ruoqiao Zhang