Reconstruction From Projection TRIZ Cases

Oppositely directed translational motion of the x-ray source and detector array reduces radiation dose while maintaining high in-plane spatial resolution.

A reconstruction algorithm applies a chi-square-gamma statistic to determine iteration stops in nuclear imaging.

A machine learning algorithm trains on sub-sampled data using segmented subsets to enforce data consistency and define loss functions.

Collimation overlap between axial scans and intermediary filtering reduce shading artifacts in 3D CT image reconstruction.

A prior image constrained compressed sensing method reconstructs medical images using averaged data to maintain high signal-to-noise ratios.

Extrapolating truncated projections via a profile function reduces artifacts in interventional C-arm X-ray systems.

A progressive generative adversarial network generates high-resolution medical images by growing layers from low resolution.

An image displaying apparatus generates maximum intensity projection images of divided breast areas.

Monte-Carlo ray-tracing calculates multiple paths to render realistic ultrasound images, resolving computational bottlenecks in real-time simulation.

Deep learning reconstruction model with learnable positional and imaging parameter embeddings for multi-slice magnetic resonance imaging.

Segmenting X-ray projections by cardiac phase similarity enables faster convergence while minimizing motion artifacts in cross-sectional heart images.

Segmenting frequency space reduces cone-beam artifacts and accelerates processing for accurate diagnostic imaging.

Image processing element sets weighting factors for nearest neighbor and bilinear interpolation based on pixel value differences.

Phase-correlated image generation processes motion-extracted projections to synthesize high-quality multi-phase CT data.

Adapting gantry speed and projection density across respiratory phases reduces streak artefacts in 4DCBCT images.

Variable pixel pitch combined with deep learning reduces communication bandwidth while maintaining high spatial resolution.

A PET device moves detectors to sample data from multiple positions, enabling ultra-high resolution image reconstruction.

Iterative mask reconstruction extends the field of view in truncated computed tomography scans, resolving artifacts from limited scan ranges.

Movable blocking elements determine detector position to correct scattered radiation, resolving the trade-off between image quality and patient dose.

Periodic modulation encodes optical information into temporal structures, reducing scan time and improving imaging accuracy.

Invalidity matrix deweights marker particles in dental tomosynthesis projections, removing artifacts from high contrast anatomy.

A channelized preconditioner decomposes image processing into independent frequency sub-bands to update estimates during iterative reconstruction.

A photon counting detector system replaces zero counts with non-zero values to maintain spatial resolution.

Applies local quality principles to resolve spatial resolution versus contrast sensitivity trade-offs by using geometry-dependent filtering strategies.

Computing depth derivatives of OCT signals to enhance morphological feature visualization in dense tissues.

A body tissue image generator applies filtering to measurement images for candidate pixel identification.

A trained model infers high-quality virtual monochromatic X-ray images from second energy data to improve diagnostic clarity.

Iterative keyhole reconstruction replaces full field projection data with targeted high-resolution scans to refine cardiac imaging.

Segmenting projection data into view subsets enables parallel processing that reduces reconstruction time while maintaining low-dose imaging accuracy.

Neural networks localize and correct scatter distributions in sinograms, eliminating mechanical gratings that reduce detective quantum efficiency.

A PET image processing method calculates noise standard deviation from count numbers to configure NLM filter parameters automatically.

A CT image processing device calculates region of interest synchronization signals from projection data to track periodic motion.

Spectral decomposition separates contrast agent data from non-contrast agent data to enable virtual image simulation at reduced doses.

A continuous kilovoltage beam acquisition method synchronizes radiation source emission with area detector readout to maximize duty cycle.

Orthogonal computed tomography scans align data to generate artifact-reduced voxel records, resolving beam hardening in metallic objects.

Periodic tube voltage variation collects multi-energy scan data for tissue discrimination, avoiding high hardware requirements of dual-source systems.

A controller and projector system projects visual depth indicators onto the patient to guide slice image placement.

Segmenting raw PET data with time-of-flight vectors corrects non-rigid motion artifacts without increasing reconstruction times.

An x-ray source translates along a 60-120 degree arc to reconstruct multiple 3D volumes, eliminating separate breast compressions and reducing radiation dose.

Sliding window reconstruction removes oldest projection images to insert new data, resolving imaging speed delays during medical monitoring.

A digital breast tomosynthesis system applies lesion-type specific reconstruction parameters to selected image regions for high-resolution visualization.

Forward projection feedback validates deep learning image reconstruction against measured data, preventing algorithmic hallucinations in medical diagnostics.

A spectral density reconstruction algorithm reduces dimensionality using optical path symmetry to process multiplexed imaging data.

Exact axial inverse rebinning in Fourier space reduces computational time while maintaining image resolution and accuracy.

Virtual scan path reprojection fills missing cone-beam CT data to eliminate artifacts and restore diagnostic accuracy.

A tolerance error estimating apparatus calculates rotation axis deviation using X-ray projection images.

A CT detector system unlogs clipped data to estimate mean values for bias correction.

CT system detects projection truncation via geometric calculation to mark truncated regions, preventing misdiagnosis from unreal image presentations.

A position determination unit computes transducer location using time of flight and beamforming signal intensity.

Monte Carlo scaling adjusts single-scatter simulation to resolve cascade gamma contamination and improve quantitative accuracy.