Reconstruction From Projection TRIZ Cases

Iterative reconstruction creates accurate volumes from simulated projection data to resolve soft tissue discrimination limits in abdominal radiation therapy.

Adaptive sampling selects regions of interest to reconstruct cone beam computed tomography images while reducing patient radiation dose.

Image domain differentiation generates synthetic line data to correct circular CT reconstruction without additional radiation exposure.

Corrects volumetric image accuracy by merging and aligning projection data sets with a reference topogram to compensate for cantilever deflection.

Weighted backprojection uses a priori attenuation data to eliminate streak artifacts and reduce x-ray dose during dynamic studies.

Extraction of a solitary adjustable parameter enables exhaustive fitting to reduce computational complexity and scan duration.

An adaptive prediction model selects reference pixels to generate precise intensity values for visual data.

Extended Field Iterative Reconstruction Technique dilutes detector noise by averaging successive reconstructions over an extended field larger than the region of interest.

A trained neural network detects metal artifacts in low-dose 3D scout scans to enable automatic image processing.

Masking fully sampled k-space center lines during iterative reconstruction eliminates banding artifacts while preserving image sharpness.

A cascaded neural network processes positron emission tomography and computed tomography images to generate attenuation-corrected results.

Pre-processes projection images using geometry-based normalization and filtering before direct reconstruction.

Navigator signals sort k-space data into motion-state bins for compressed sensing reconstruction.

A reconstruction system calculates multiple Mellin-Laplace transforms to analyze temporal signal distributions from scattering media.

A multi-frequency imaging system decomposes composite tomographic data into discrete material images to identify non-metallic contraband.

A cone beam CT system updates a projection matrix using digitally reconstructed radiograph matching to generate accurate three-dimensional volume images.

A signal processing system reconstructs phase contrast images using a dedicated phase variable for reference fluctuations.

MLEM reconstruction incorporates scattered coincidences to generate attenuation maps without transmission scans, reducing patient radiation exposure.

A neural network reconstructs medical images from sensor data patches using manifold learning techniques.

A Gaussian Mixture Markov Random Field model combines patch density and spatial structure to enhance image reconstruction accuracy.

Renormalization factors adjust decomposed bone and soft tissue images to match conventional radiograph scales, resolving scaling issues in display processing.

A hybrid x-ray detector merges energy-integrating and photon-counting arrays to acquire simultaneous imaging data.

A bone mineral information acquisition apparatus estimates body thickness and acquires pixel values to calculate bone mineral content.

Automated registration framework aligns ultrasound and pre-acquired 3D volumes using subject-specific shape information.

A system generates adaptive multi-resolution digitally reconstructed radiographs using dynamically selected rendering algorithms.

A cardiac CT processing method applies image parameters from a single reference 3D image to generate complete sequences.

A nuclear medicine diagnosis apparatus calculates detector-specific time-of-flight kernels using photon number information to match individual detection capabilities.

A study navigation system generates representative image series from aligned planar images to streamline medical review workflows.

Sinogram interpolation corrects metal artifacts from external ultrasound probes, improving radiotherapy dose delivery accuracy.

Segmented detector panels reconfigure dynamically to compensate for limited angular sampling, enabling artifact-free molecular imaging at the patient bedside.

A magnetic resonance imaging system selects spatial test regions based on position overview data scattering to reconstruct image data.

Dynamic regularization parameters adapt to voxel changes over time, resolving the contradiction between high temporal resolution and extended scan durations.

A tomography system acquires projection images using orthogonal megavoltage and kilovoltage x-ray sources to reconstruct three-dimensional volumes.

A hierarchical tomographic reconstruction method processes scan data through intermediate line integral representations to generate voxel values using deep learning.

Augments verification images with native projection data to remove streak artefacts and truncation errors during biopsy guidance.

A detection unit moves along a virtual circle to generate both X-ray CT and tomosynthesis images from identical fluorography data.

Hermetic Transform signal processing enhances image resolution in tomography devices.

A radiographic imaging system applies non-linear conversion to band limiting image signals before back projection.

Dynamic PET acquisition time adjustment using real-time count rate feedback to ensure uniform image quality across varying patient conditions.

Oversampling detection elements with vector decimation eliminates iterative approximation to resolve computational complexity in tomographic imaging.

An optical sensor detects object surface geometry to supplement missing projection data, resolving truncation artifacts in limited-field-of-view C-arm systems.

A tomographic image generation apparatus derives positional shift amounts from projection images to correct body movement artifacts during reconstruction.

Deep learning replaces linear beamforming to reduce speckle noise while preserving lateral resolution in ultrasound imaging.

Segmenting projection data into angular sectors with dynamic spectral filtering reduces limited-view artifacts while maintaining computational efficiency.

A tomosynthesis reconstruction method applies an exponential convolution kernel during filtered back projection to enhance edge visualization in 3D x-ray images.

Reconstructor detects divergent gradient locations across kVp-switched image scans to identify undersampling artifacts.

The fsHBMAP algorithm processes tomographic data using stochastic approximation to achieve high-quality image reconstruction.

A method determines image values in marked pixels using epipolar consistency conditions derived from Radon transform.