← Back
Home › Systems › Healthcare

medical-image-reconstruction

MRI/CT/PET image reconstruction on the scanner

GE AIR Recon DL, Siemens Deep Resolve, Gadgetron

SystemTechniques · 12Primitives · 29 medical-image-reconstru… Fpga dataflow pipelineshardware-os · medium: Fully pipelined fixed-latency FPGA logic: initiation interval, HLS, BRAM and DSP budgetingFiltered back projectionmedia · medium: Reconstruct a tomographic image by ramp-filtering projections and back-projecting them along ray anglesImage alignment registrationmedia · medium: Align frames via feature matching or pyramidal optical flow for burst merging and stitchingImage denoisingmedia · medium: Bilateral, non-local means and temporal noise reduction for sensor imagesMl em tomographic reconstructionmedia · medium: Iterative maximum-likelihood image estimation (MLEM, ordered subsets OSEM) for Poisson projection data using forward and back projectionRadiometric calibrationmedia · medium: Convert sensor counts to reflectance: radiometric and atmospheric correctionNon uniform fftnumerics · medium: Fourier transform of off-grid samples by gridding with a convolution kernel, FFT on the grid and deapodisationSpline interpolationnumerics · low: Linear, bilinear/trilinear and cubic spline interpolation on grids and tablesDma transfersembedded · low: DMA controllers: descriptors, circular buffers, cache coherence for zero-CPU data movementFixed-rate control loopembedded · low: Run a controller at an exact period using absolute deadlines and a fixed dt, so jitter never accumulates and the math stays valid.›Voxel ray traversal ddagraphics · low: Amanatides-Woo grid DDA to step rays through voxels for picking, shadows and collisionCUDA and GPU languageslanguages · low: Programming GPUs: thousands of lockstep threads, bound by memory bandwidth; libraries first, Triton for fused kernels, CUDA C++ for peak speed.›Convolutional networksml-training · low: Layers that slide a small shared kernel over a grid, giving translation equivariance and few parameters; still the most compute-efficient choice for many vision tasks.›Dicom protocolnetworking · low: DICOM data elements, transfer syntaxes, DIMSE services (C-STORE, C-FIND) and DICOMwebConjugate gradientnumerics · low: Krylov iteration for symmetric positive definite Ax = b using one matrix-vector product and two dot products per step; converges in about √κ iterations.›Fast Fourier transformnumerics · low: Computes the DFT in O(n log n) by recursively splitting into even and odd halves; the basis of fast convolution and spectral methods.›LU and Cholesky factorizationnumerics · low: Factor A = PLU (any square matrix, partial pivoting) or A = LLᵀ (SPD) once in O(n³), then solve each right-hand side in O(n²).›PID controllerrobotics · low: Drive a measured value to a setpoint with three terms on the error: proportional, integral, derivative. The default feedback loop for motors, heaters, drones and gimbals.›Homogeneous transformsgeometry · low: 4x4 affine and projective matrices, homogeneous coordinates, view and projection, normal matricesImage pyramidsgraphics · low: Gaussian/Laplacian pyramids and tiled multi-resolution images for zoom and blendingImage warping remapgraphics · low: Resampling with lookup maps: resize, lens undistortion, homographies, bilinear/bicubicFeature detection orbrobotics · low: FAST corners with oriented BRIEF descriptors and Hamming matching for tracking and relocalizationRansacrobotics · low: Random sample consensus for robust model fitting with outliers (homographies, planes, poses)+ 5 more techniques+ 13 more primitives

Similar systems

Share the most building blocks with this one.

security-screening 17 %