version 0.9.00

Changes:

• new tools: nlmeans, tensorflow, nufftbase, calc, psf
• calc: access some mathematical functions from the command-line
• nlmean: non-local means filter
• tensorflow: load and apply Tensorflow graph
• nufftbase: compute Fourier transform of basis used in nuFFT
• psf: compute PSF
• nufft: various new options to control behavior
• nufft: basis and weights
• nufft: zero and low memory modes
• nufft: use -x for image dimensions
• sim: Bloch-McConnell simulations
• traj: rotation of Cartesian trajectories
• traj: RAGA trajectories
• reconet: MPI support
• reconet: tensorflow graph support
• reconet: export graph for visualization
• pics: mpi and multi-GPU support
• pics: add shared-img-dims option to pics
• pics: add option to specify FISTA
• pics: preconditioning
• nlinv: add option to specifiy non-Cart image size
• phantom: BRAIN geometry
• phantom: 64ch sensitivities
• phantom: ellipsoid simulation
• phantom: support for generic geometries
• mobafit: add basis to mobafit
• mobafit: add IR and IR LL to mobafit
• mobafit: magnitude fitting
• mobafit: scaling and initial values
• mobafit: minimum and maximum bounds
• mobafit: GPU support
• nrmse: add option for scientific notation
• moba: Export and ksp coils in moba
• moba: Various new options
• estdelay: support for looping
• version: show tensorflow and MPI support
• signal: Spin-Echo signal model
• scripts/shell: kspace_precond.sh: Script to compute k-space preconditioner
• scripts/python: bart_tf.py: Export tensorflow graphs for use with BART
• scripts/python: bart.py: Print stdout and stderr output from bart call
• scripts/python: splines_from_svg.py: extract spline control points from SVG files
• library/geom: Brain geometry
• library/misc: better backtraces
• library/num: md_ravel_index, md_select_strides, md_zsmin2, md_unravel
• library/num: library for complex-valued Gaussians
• library/num: cuda kernels: zdot, zfsq2, zfftmod_3d, zsmin, compress, decompress
• library/num: cuda batch kernels: xpay, axpy, dot_bat
• library/num: cuda kernels: (z)add_unfold, (z)mul(c)_unfold, (z)fmac(c)_unfold
• library/num: md_nlmeans(2), md_nlmeans_distance(2)
• library/num: library for MPI sync, broadcast, reductions
• library/num: double precision bessel function
• library/num: special functions: sinc, sincf, jinc
• library/num: library for virtual pointers
• library/num: time-ordered matrix exponentials: mat_to_exp
• library/num: functions for adjoint sensitivity analysis
• library/num: functions for quadratures: trapezoidal, simpson
• library/num: configure fftw wisdom/measure via environment variable
• library/simu: Bloch McConnell simulations
• library/simu: Spin Echo signal model
• library/linops: linop: flip
• library/nlops: nlops: zphsr and zabs
• library/nn: nlops: ardioid, siglog, igaussian
• library/network: tensorflow networks
• library/noncart: nufft inverse gridding on GPU
• library/noncart: RAGA trajectories
• library/iter: weighted least-squares prox function
• library/iter: batched conjgrad
• library/iter: multiple regularizations for primal-dual
• general: webasm support
• general: BART_ prefix to environment variables
• general: generic loop interface to bart
• general: basic MPI support
• general: OpenMPI and GPU for gridding
• general: zero-overhead mode for nuFFT
• general: performance improvements
• general: many GPU improvements, unified GPU initialization
• built: FORTRAN and PNG options
• built: LTO integration test
• built: OMP=0 CUDA=1 compilation
• built: more unit and integration tests
• many other bug fixes and improvements

version 0.8.00

Changes:

• new tools: epg ictv sim reconet onehotenc measure mnist multicfl morphop fovshift
• epg: EPG simulations
• ictv: infimal convolution TV (experimental)
• sim: Bloch simulations
• reconet: deep learning reconstruction (MoDL, VarNET)
• onehotenc: transform onehotencoded data to integer encoded
• measure: compute MSE, SSIM, PSNR
• mnist: simple mnist network for demonstration
• multicfl: combine/split cfl files
• morphop: morphological operations
• fovshift: retrospectively shift the FOV
• pics: tensorflow loss
• pics: select wavelet type
• pics: TGV/ICTV regularization (experimental)
• moba: scaling parameters via command-line parameters
• moba: various numerical fixes
• moba: time timensions
• moba: support for ADMM
• moba: Bloch model recon with tests
• moba: multi-gpu option
• moba: scaling of TV derivatives
• phantom: NIST phantom and others
• phantom: rotation for TUBE and NIST phantom geometries
• phantom: rectangle geometry
• traj: flag for oversampled trajectory
• traj: 3D uniform (half-sphere) trajectory
• sim: output partial derivatives
• twixread: updates for VD/VE versions
• ismrmd: support for reading XML metadata
• estdelay: more generic regarding un-/centered trajectories
• tgv: for rof / tgv denoising
• rmfreq: support for contrast changes and coilwise contrast
• mobafit: T2 and diffusion fit
• ecalib: make number of iterations for orthiter configurable
• nufft: warn about incorrectly scaled trajectories
• nlinv: fix noncart ENLIVE
• pics: fix basis pursuit pics when using a sampling pattern (#285)
• fakeksp: fix output argument
• python interface: allow multiple files with the correct option string
• python interface: faster write_cfl
• library: New md functions: zacos, zsinh, zcosh, pdf_gauss, zmaxnorm, zcorr, tenmul
• library: framework for neural networks
• library: optimization algorithms: SGD, Adam, Adadelta, iPALM
• library: linear operators: scale, zconj, zreal, permute, padding, repmat, scaled_sum
• library: many new non-linear operators
• library: support for using different wavelet type
• library: nlop_attach, for attaching a random data pointer
• library: nlop reshape function
• library: allow forward nufft with toeplitz
• library: tgv/ictv + multiple penalties + 3D
• generic: radial DCF examples
• generic: Add support for multi cfl
• generic: POSIX shared memory files
• generic: improve determinism
• generic: improved parallelization and multi-gpu support
• generic: add better support for use as shared libray
• generic: bart for centos 7
• generic: Fix Fedora Packages
• generic: Windows support by MSYS2
• generic: support for linking with cudann
• generic: support for linking with tensorlfow
• generic: add pythontest to github action
• generic: LTO test builts
• many other bug fixes and improvements

version 0.7.00

Changes:

• new tools: signal mobafit roistat pol2mask conway
• signal: tool to simulate MRI signals
• mobafit: tool for point-wise fitting
• roistat: tool to compute ROI statistics
• pol2mask: tool to compute masks from polygons
• conway: Conway's game of life
• moba: model-based recon for multi-echo FLASH
• moba: model-based T2 mapping
• moba: make coil smoothness configurable
• moba: add command-line option for alpha_min0
• moba: add option for k-space filter
• moba: enable cartesian k-space data in moba
• wshfl: Add fista option
• estdelay: export to file
• traj: multi-echo trajectory
• traj: reading of gradient delays from file
• phantom: random tube phantom
• nlinv: non squared image sizes
• generic: additional tests and unit tests
• generic: continuous integration for gitlab
• generic: support for BART_COMPAT_VERSION
• generic: warning for dirty git versions
• generic: target for libismrm.a if ISMRMRD enabled
• generic: support for IBM POWER9 HPC ppc64le
• generic: MATLAB/Octave test
• library: add support for long options
• library: new functions: md_zslessequal
• library: linop_reshape_in, linop_reshape_out
• library: generalized linop_sum
• library: operator_stack along an arbitrary dimension
• library: new zsmax prox operator
• library: continuation for lsqr/itop
• library: warmstart for lsqr
• library: BLAS accelerated strided operations
• library: add more geometry functions
• library: more operator_p helper functions
• many other bug fixes and improvements

version 0.6.00

Changes:

• new tools: bin ssa upat rmfreq
• bin: tool for binning of dynamic data (Sebastian Rosenzweig)
• ssa: Singular Spectrum Analysis (SSA) and SSA-FARY method (Sebastian Rosenzweig)
• rmfreq: remove angle-dependent frequency in radial acquisitions (Sebastian Rosenzweig)
• upat: generate Cartesian undersampling patterns
• phantom: tubes and BART logo phantoms
• reshape: perform more general reshapes
• show: control precision (Max Litster)
• nufft: option to output over-sampled grid
• version: check for minimum BART version
• traj: problems introduced in 0.5.00 fixed (208, 209, ...)
• traj: support for asymmetric echos
• pics: allow data in COEFF_DIM if basis is not used
• nlinv: add direct support for non-Cartesian trajectories
• tests: bin, estdelay, ssa, nlinv, pics, nufft, phantom, reshape, upat
• generic: removed CMake build system
• generic: simplify building of libraries
• generic: support running valgrind on unit tests
• generic: better support and packages for Fedora and CentOS (Philip Schaten)
• library: new functions: md_reshape, md_pad, md_check_equal_dims, md_zcovar,
  md_zdiv_reg, md_exp, md_log, md_zsin, md_zcos, md_zatanr, md_zss2,
  md_zsmin, md_real, md_imag
• library: new line drawing functions required by view
• library: new functions: mat_pinv_left, mat_pinv_right
• library: new debug function to print bitmasks
• library: fix GPU fft to work with more complicated dimensions (Christian Holme)
• many other bug fixes and improvements

version 0.5.00

Changes:

• new tools: moba, looklocker, tgv, extract, fftrot
• moba: non-linear model-based recon for T1 mapping (Xiaoqing Wang)
• looklocker: post-processing for moba
• extract: extract ranges from multi-dim. arrays
• tgv: a tgv denoiser
• fftrot: perform rotations using FFTs
• join: append to non-existing file
• pics: low-mem flag (-U)
• pics: efficient basis functions (Sebastian Rosenzweig)
• pics: pattern for non-Cartesian trajectories
• nlinv: enlive on the gpu
• twixread: read multi-raid files
• twixread: mode for radial
• traj: custom angle to traj (Aurélien Trotier)
• traj: small golden angles (Jasper Schoormans)
• traj: option to rotate trajectory
• extract: multiple dims (Christian Holme)
• slice: multiple dims (Christian Holme)
• generic: NOEXEC_FLAG flag for non-executable stacks
• generic: automatic rebuilds for makefile changes
• generic: make it possible to run individual tests
• library: use long in gridding code to support larger grids
• library: zsmax functions for cpu and gpu
• library: function for setting random initial state in wavthresh
• library: lsqr/lad return operator_p to select regularization
• library: generalize cuda fft a bit for efficiency
• library: fix for CUDA strided memcpy
• many other bug fixes and improvements

version 0.4.04

Changes:

• new tools: index poly wshfl zexp
• index: generate indices
• poly: generate polynomial patterns (Siddharth Iyer)
• wshfl: wave-shuffling (Siddharth Iyer)
• zexp: replace zexpj with zexp for complex exponentials
• pics: support for temporal basis
• pics: overlapping blocks option for LLR (Jon Tamir)
• pics: make basis pursuit possible for noncartesian data
• pics: non-negative constraint (Jon Tamir)
• nlinv: introduce option for sms
• ecalib: fix bug in phase rotation
• ecalib: add debug_level argument
• estdelay: RING method (Sebastian Rosenzweig)
• traj: Half circle golden-ratio sampling (Sebastian Rosenzweig)
• traj: Option for asymmetric trajectories (Sebastian Rosenzweig)
• bench: add option to select benchmarks using bitmask
• phantom: add geometric object phantom (Sebastian Rosenzweig)
• phantom: fix inconsistency between img-phantom and k-phantom for rectangles (Sebastian Rosenzweig)
• copy: enhance copy command to do simple copies
• toimg: windowing and option to re-scale images to full dynamic range (Jon Tamir)
• generic: many new integration and unit tests
• generic: improved WSL support (Soumick Chatterjee)
• generic: add mkl support (Jon Tamir)
• generic: bart as a library (Damien Nguyen)
• generic: CMake improvements (Damien Nguyen)
• generic: Save FFTW wisdom when TOOLBOX path is set (Siddharth Iyer)
• generic: fixes for memory leaks
• library: optimized strided copy on GPU (Simon Yeung)
• library: parallelization to fftmod
• library: nufft: efficient use with temporal basis
• library: nufft: fix issue with over-sized trajectories
• library: add runga-kutta ode solver
• library: iter/italgos: alternating minimization (Christian Holme)
• library: irgnm: add minimum for alpha as parameter (Christian Holme)
• library: step size support in chambolle-pock
• library: add tenmul operator
• library: add md_zexp
• library: add zexp nonlinear operator
• library: add linop_null and linop_plus
• library: add md_zsum: (Complex) sum along specified dimensions (Sebastian Rosenzweig)
• library: fix bug in septrafo (broke cdf97 wavelets)
• many other bug fixes and improvements

version 0.4.03

Changes:

• nlinv: add ENLIVE support (Christian Holme)
• nlinv: add (hidden) option to specify coil weighting parameters (Christian Holme)
• nlinv: option for different pattern for each coil (Christian Holme)
• nlinv: sensitivity output also when using GPU (Christian Holme)
• nlinv: add timing output (Christian Holme)
• nlinv: add option for debug level (Christian Holme)
• nlinv: add option to provide initialization (Christian Holme)
• nlinv: make IRGNM reduction factor configurable (Christian Holme)
• nufft: add nufft according to conventional definition
• ecalib: add option to turn off phase rotation (Jonathan Tamir)
• homodyne: add option for uncentered ffts (Jonathan Tamir)
• bench: add benchmarks for fft, fftmod, md_fft
• library: admm: add parameter selection by residual balancing (Jonathan Tamir)
• library: add function to detect nontrivial strides
• library: fix bug for gpu-mode of cdiag operator
• library: irgnm: add callback (Christian Holme)
• general: add script for memory-economic use of espirit
• build: updates and fixes for CMake
• build: build fixes on Mac OS X
• many other bug fixes and improvements

version 0.4.02

Changes:

• new tools: std var vec wavepsf whiten
• std: compute standard deviation (Jonathan Tamir)
• var: compute variance (Jonathan Tamir)
• vec: create vectors from the command line
• wavepsf: create PSF for wave acquisition (Siddharth Iyer)
• whiten: compute/apply whitening matrix (Jonathan Tamir)
• pics: basis pursuit formulation (Jonathan Tamir)
• nlinv: support for simultaneous multi-slice (Sebastian Rosenzweig)
• nlinv: various enhancements and fixes (Christian Holme)
• pics: support for simultaneous multi-slice (Sebastian Rosenzweig)
• traj: radial simultaneous multi-slice trajectories (Sebastian Rosenzweig)
• fft: uncentered option (Jonathan Tamir)
• nufft: use Toepliz-mode by default
• nufft: add GPU option
• python 3 version for bartview (Siddharth Iyer)
• fix compilation for Cygwin on Windows (Johannes Töger)
• include relevant parts of LAPACKE in BART
• library: add NIHT algorithm (Sofia Dimoudi)
• library: add Chambolle-Pock primal dual algorithm for F(Ax) + G(x) (Jonathan Tamir)
• library: add md_zss function for sum-of-squares (Jonathan Tamir)
• library: improved parallelization (Michael Anderson)
• library: joint l1-wavelet regularization
• library: rename wavelet3 to wavelet
• library: Hamming and Hann windows (Jonathan Tamir)
• library: png write functions (Christian Holme)
• library: initial interface for nonlinear operators
• many other bug fixes and improvements

version 0.4.01

Changes:

• new tools: casorati mandelbrot wavelet estshift estdelay ccapply
• cc tool: new option to output compression matrix
• traj tool: new option for gradient-error corrected radial trajectories
• pics tool: fix for the use of weights for non-Cartesian trajectories
• pics tool: fix for the use of weights with higher (temporal) dimensions (Jon Tamir)
• pics tool: allow the use of higher dimensions in sensitivity maps (Jon Tamir)
• nlinv tool: various fixes (Christian Holme)
• threshold tool: fix for hard thresholding (Jon Tamir)
• cmake: integration testing (Damien Nguyen)
• cmake: CUDA support (Damien Nguyen)
• fixes for a couple of memory leaks (Damien Nguyen, Christian Holme)
• faster system tests, more unit tests and system tests
• library: improved precision of wavelet filters
• library: new functions for tensor multiplication (md_ztenmul, ...)
• library: function to compute sensitivities with Biot-Savart law
• library: functions for polynoms
• library: improved and extended functions for splines
• many other bug fixes and improvements

version 0.4.00

Changes:

• new tools: copy, squeeze, flatten, delta, zexpj, invert
• show tool: option to use custom number formatting
• poisson: option to specify random seed in poisson (David Smith)
• threshold: hard threshold option
• join: append mode
• nufft: nudft option
• nufft: randshift for nufft
• pics: batch mode
• pics: FT as sparsity transform
• nrmse: auto_scale option
• fmac tool: if only two args are specified then squash-and-sum
• remove rsense tool
• support for cmake (Hans Johnson)
• rename vars.(sh|m) to startup.(sh|m)
• more efficient Matlab interface (Tim Loderhose)
• write BART version and command line into created files
• change fftw to use 64-bit interface
• phase out old wavelet2 code
• library: phase cycling for nufft
• library: support weights in toeplitz mode
• library: fmac operator
• library: cabs: complex magnitude
• library: md_compare
• library: thomas algorithm
• library: inital spline code
• many other bug fixes and enhancements