Reconstruction From Projection TRIZ Cases

Controller updates computed tomography pixels based on luminance changes, reducing calculation time while maintaining image quality.

Blending scan data from multiple cardiac cycles using adaptive weighting factors reduces banding artifacts in reformatted CT cardiac images.

Phase-specific image reconstruction generates motion-corrected volume renderings to eliminate cardiac blurring artifacts.

Interactive rotation axis coordinate shifting corrects tomogram alignment, eliminating runout artifacts from high-magnification scans.

Reconstructs projection data at a resolution matching actual information to display diagnostic images without unsupported intensity variations.

A temporal compressive sensing system captures distinct linear combinations of radiation patterns to reconstruct time slice datasets.

Image processing apparatus determines optimal viewpoint and sight line direction for selected bone regions using pre-defined tables.

Correlation profiles between target and source patches extrapolate truncated CT projections, eliminating z-truncation artifacts.

Iterative reconstruction method uses structural image data to define inhomogeneous initial states for optical coefficient calculation.

Motion correction neural network generates corrected k-space data from corrupted inputs using estimated motion parameters.

A machine learning model calculates tube current values from scout images to determine optimal exposure settings.

A system selects the optimal ECG channel to synchronize cardiac scans.

Correlation-based motion estimation compensates for head movement during continuous scanning, maintaining image registration accuracy without repeated scans.

Photon-counting detectors resolve energy bins to improve contrast resolution while maintaining deep tissue penetration via photoacoustic signals.

Angle-dependent scatter model calculates radiation distribution to correct measurement data, reducing cone beam artifacts in dual-source CT systems.

Segmenting single and double scatter simulations balances computational efficiency with measurement precision, resolving trade-offs in TOF-PET imaging.

JE-MAP algorithm jointly estimates tissue types and linear attenuation coefficients using latent Markov Random Fields.

A photon counting CT apparatus estimates imaging energy ranges to reconstruct image data using primary photons.

Prior embedding neural networks reconstruct medical images from sparse data, eliminating large-scale training requirements while maintaining high accuracy.

Computational image reconstruction overcomes source size limits by fusing low and high resolution data for clearer X-ray shadowgraphs.

Geometric correction of optical coherence tomography data using morphological alignment to resolve sample movement distortions.

A computed tomography method segments scan data into multiple fields of view to reconstruct separate images with optimized resolution.

Scaling representative 2D point response functions by distance estimates 3D models with 1% accuracy, avoiding extensive spatial sampling.

A medical image analysis apparatus iteratively optimizes a single inverse function to output segmented datasets.

A center shift amount estimating apparatus specifies a region of interest to reconstruct temporarily corrected images for accurate deviation calculation.

Touch screen interfaces in IVUS systems recognize movements to pan and rotate longitudinal images, reducing time to locate features.

Segmenting optimization into subproblems resolves slow convergence and poor quality in hybrid CT systems with distinct system matrices.

Segmented task execution reduces local complexity while accelerating image reconstruction speed.

A computing facility ascertains position information from three-dimensional data sets to generate synthetic projections automatically.

A system adjusts acquisition time periods based on real-time curves to align multi-modality scan data.

A parallel fixed-point adder tree architecture computes discrete periodic Radon transforms using circular shift registers to eliminate external memory access.

A trained advance calculation function predicts imaging reconstruction durations using protocol and hardware parameters.

Segmented multi-band acquisition optimizes inferior and superior brain imaging, reducing total scan time while maintaining spatial and temporal SNR efficiency.

Segmenting rotational acquisitions into temporal phases separates arterial and venous signals, resolving spatial-temporal resolution trade-offs.

Segmented voxel projections enable tailored image reconstruction parameters for distinct object regions, resolving computational complexity trade-offs.

Segmented detector array overcomes inverse square law sensitivity loss to reconstruct gamma ray emission images.

Processor organizes projection data by angle and rotates images to align slices parallel to an axis for efficient kernel convolution.

A tomography apparatus processes X-ray image data to generate noise-free reference images for cross-sectional reconstruction.

A bias correction look-up table estimates post-log bias in computed tomography projection data to mitigate artifacts from positivity mapping.

Pre-calculating optimal k-space line assignments reduces gradient switching and eddy currents while maintaining uniform scan paths.

Weighted linear combinations of filtered back-projection density distributions reduce cross-talk artifacts from limited projection counts.

A neural network generates scatter estimates from CT projection data or uncorrected images using Monte Carlo training.

Pixon smoothing assigns local kernel functions to reduce noise in tomographic data while preserving lesion details for accurate medical diagnosis.

Reconstructs interior CT images without prior object information by combining Scout-View scan data with ROI measurements to eliminate artifacts.

Removing table contributions from projection data enables accurate spectral calibration without precise phantom centering.

Transforms high-intensity pixel values to generate scattered ray images, reducing calculation time and preventing excessive correction in real-time display.

Machine-learned models recover information from nuclear imaging data to generate diagnostically useful images.

A computed tomography system uses a fixed reference object to sample position data during spiral scans.

Iterative line integral solving reduces beam hardening artifacts in dual-energy CT, enabling accurate monochromatic image generation.

A scanner partitions reflected wave intensities into segments to reconstruct multi-layer structures using joint-layer hierarchical image recovery.