An x-ray imaging apparatus and associated methods are provided to receive measured projection data from a wide aperture scan of a wide axial region and a narrow aperture scan of a narrow axial region within the wide axial region and determine an estimated scatter in the wide axial region using an optimized scatter estimation technique. The optimized scatter estimation technique is based on the difference between the measured scatter in the narrow axial region and the estimated scatter in the narrow axial region. Kernel-based scatter estimation/correction techniques can be fitted to minimize the scatter difference in the narrow axial region and thereafter applying the fitted (optimized) kernel-based scatter estimation/correction to the wide axial region. Optimizations can occur in the projection data domain or the reconstruction domain. Iterative processes are also utilized.

Multimodal imaging apparatus and methods include a rotatable gantry system with multiple sources of radiation comprising different energy levels (for example, kV and MV). Fast slip-ring technology and helical scans allow data from multiple sources of radiation to be combined or utilized to generate improved images and workflows, including for IGRT. Features include increasing the precision of spatial registrations between respective image sets to allow more precise radiation treatment delivery, reducing image artifacts (e.g., scatter, metal and beam hardening, image blur, motion, etc.), and utilization of dual energy imaging (e.g., for material separation and quantitative imaging, patient setup, online adaptive IGRT, etc.).

An x-ray imaging apparatus and associated methods are provided to process projection data from an offset detector during a helical scan, including view completion. The detector may be offset in the channel and/or axial direction. Projection data measured from a current view is combined with projection data measured from at least one conjugate view to reconstruct a target image. A two-dimensional aperture weighting scheme is used to address data redundancy.

The invention relates to off-center detector 3D X-ray or proton radiography reconstruction. Redundancy weighting with a steep weighting function around the iso-axis typically leads to artifacts in the reconstruction, for example, if inconsistencies between two nominal redundant projections occur, e.g. due to slightly incorrect detector calibration or scatter correction, etc. With the present invention, an approach is presented for overcoming or mitigating these problems.

The disclosure relates to systems and methods for iterative reconstruction. Raw data detected from a plurality of angles by an imaging device may be obtained. A first seed image may be generated by performing a filtered back projection on the raw data. A first air mask may be determined by performing a minimum value back projection (BP) on the raw data. One or more images may be reconstructed by performing an iterative reconstruction based on the first seed image, the first air mask, and the raw data.

Methods, systems, and computer-readable media for generating a cross-section of a 3D model are disclosed. An example method includes determining a cross-section plane intersecting the 3D model, performing ray-tracing by passing each of a plurality of rays through a corresponding pixel of a viewing plane such that each ray intersects the cross-section plane, determining one or more rays that are within a threshold distance of the 3D model at their respective points of intersection with the cross section plane, and highlighting pixels corresponding to the determined rays.

A method relates to the use of deep learning techniques, which may be implemented using trained neural networks (50), to estimate various types of missing projection or other unreconstructed data. Similarly, the method may also be employed to replace or correct corrupted or erroneous projection data as opposed to estimating missing projection data.

When generating a measurement plan for measuring X-ray CT that performs X-ray irradiation while rotating a test object, and in doing so acquires projection image data, reconstructs volume data from the projection image data, and measures a targeted measurement location in the volume data, the present invention calculates required measurement accuracy and a measurement field of view range based on tolerance information included in CAD data of the test object and a measurement location on the test object defined by a measurement operator ahead of time, and automatically generates, from this information, an optimized measurement plan that minimizes the number of measurements.

Techniques for computed tomography (CT) image reconstruction are presented. The techniques can include acquiring, by a detector grid of a computed tomography system, detector signals for a location within an object of interest representing a voxel, where each detector signal of a plurality of the detector signals is obtained from an x-ray passing through the location at a different viewing angle; reconstructing a three-dimensional representation of at least the object of interest, the three-dimensional representation comprising the voxel, where the reconstructing comprises computationally perturbing a location of each detector signal of the plurality of detector signals within the detector grid, where the computationally perturbing corresponds to randomly perturbing a location of the x-ray within the voxel; and outputting the three-dimensional representation.

A tomography system having a central processing unit, a system memory communicatively connected to the central processing unit, and a hardware acceleration unit communicatively connected to the central processing unit and the system memory, the hardware accelerator configured to perform at least a portion of an MBIR process on computer tomography data. The hardware accelerator unit may include one or more voxel evaluation modules which evaluate an updated value of a voxel given a voxel location in a reconstructed volume. By processing voxel data for voxels in a voxel neighborhood, processing time is reduces.