Reconstruction From Projection TRIZ Cases

A computed tomography method blends partial images using gradient weights to reduce noise while preserving edge sharpness.

Segmenting CT data subsets isolates moving regions to update image reconstruction, reducing motion artifacts while avoiding mechanical complexity limits.

A classification unit adapts thresholds using multi-energy intensity data to separate materials.

Prior image constrained compressed sensing mitigates motion artifacts in cardiac CT by minimizing an objective function with sparsity constraints.

Iterative multi-material correction reduces residual beam hardening artifacts by performing multiple decomposition and re-projection iterations.

A forward projection apparatus varies ray spacing and interpolation kernel width based on effective image element spacing.

Prescan-based correction resolves intensity variations and phase singularities caused by geometric coil compression.

A transfer function modifies forward projection data in iterative CT reconstruction algorithms to remove beam hardening and scatter artifacts.

An X-ray CT apparatus selects image quality improvement processes based on irradiation dose ratios to maintain consistent reconstruction results.

A virtual PET system copies real crystal response characteristics to determine count loss correction factors for continuous incremental scanning.

Radial k-space sampling accelerates acquisition to resolve dynamic stiffness changes without sacrificing spatial resolution.

A controller moves the X-ray emitter and detector based on object location data, resolving manual positioning complexity while improving imaging efficiency.

Matches and splices projection data from opposing trajectories to generate non-truncated datasets, reducing cone-beam artifacts in CT imaging.

Multi-directional likelihood search reduces calculation load by segmenting iterative parameter estimation into preliminary and connecting steps.

A geometry-dependent weighting function corrects projection data for cone beam scanners.

Dynamic projection data rebinning generates high-resolution images from photon counting detectors.

Parallel data-acquisition circuits average digital signals from X-ray detection elements to improve image quality.

A medical image processing technique reconstructs three-dimensional moving images to identify characteristic regions within the body.

A four-dimensional reconstruction framework generates volumetric digital subtraction angiography datasets from time-resolved projection images.

A CT imaging apparatus generates additional projections based on real-time evaluation of previous data to improve image reconstruction.

A tomographic imaging system solves electromagnetic inverse problems using differential equations to reconstruct spatial distributions of object features.

Tomographic imaging system suppresses bone artifacts using learned models to enhance lesion visibility in lung scans.

Segmenting data acquisition into virtual frames allows concurrent processing, reducing scan latency and computing bottlenecks in time-of-flight PET imaging.

A computed tomography system acquires dual-energy datasets to generate base material images for tissue characterization.

A weighting coefficient map controls prior knowledge application per pixel in iterative X-ray CT reconstruction.

Modular detector heads on a universal gantry adapt to specific scan requirements, reducing system complexity and reconfiguration costs.

Gridless point-vector modeling replaces fixed grids with dynamic vector structures, resolving scalability constraints while maintaining high precision.

Auxiliary channel of embedded display port transmits control signals to tune antenna resonance frequency across multiple bands.

Geometry templates pre-fetch projection data to resolve bandwidth constraints in CT image reconstruction.

A multidimensional image rendering module generates real-time three and four dimensional visualizations for magnetic resonance imaging workflows.

Neural network generates synthetic reference data to correct artifacts in measured k-space, accelerating acquisition while maintaining image quality.

A cascaded recurrent neural network processes sub-sampled k-space data to reconstruct magnetic resonance images.

A mobile device captures map images and projects specific data onto surfaces for enhanced visualization.

A medical imaging system registers external photographs with internal scan data to create a unified 3D visualization.

Processing circuit extracts body motion regions from respiratory-gated nuclear medicine images and controls display to highlight these areas.

A compact line-field swept source OCT system employs a cat's-eye laser architecture to generate tunable optical signals for biological monitoring.

Detects low-speed fluctuation periods in heart motion to synchronize tube positions and combine projection data, reducing image blurring during volume scans.

A medical imaging system projects 3D tomosynthesis data into a synthetic 2D view for precise lesion localization.

A tomographic image reconstruction method uses spatially variant hyper-parameters to optimize data fidelity and penalty terms iteratively.

A radiographic apparatus limits optimization parameters to accelerate calculation speed while maintaining convergence accuracy.

Neural network reconstructs magnetic resonance images using pre-computed sensitivity maps from extended field of view calibration scans.

Image display apparatus generates tomographic images using slab data to reduce storage requirements.

Streak tube imaging system captures ultrashort laser pulse returns to generate two-dimensional target images for lidar detection.

Reduces reconstruction time by segmenting datasets and applying high-fidelity algorithms only to clinical regions of interest.

A GPU reconstructs tomographic images using trilinear interpolation and 3D blurring kernels for efficient parallel processing.

A computer-implemented method segments volumetric scans into classified parts to generate adaptive 3D wireframe models.

Switches photon-counting detector signals from macro-pixels to micro-pixels for material decomposition.

A shared image reconstruction apparatus processes magnetic resonance imaging tasks to optimize resource allocation across multiple systems.

A scattering tomography device reconstructs internal images using a pre-set asymptotic equation for the reconstruction function.

Residual deep learning network reconstructs high-fidelity 3D volumes from sparse projections, reducing imaging time and radiation dose.