Justin P. Haldar: Software

---

Software implementations of many of the methods and algorithms we have developed is publically distributed, and implementations of other methods may be available on request. Please see the links and descriptions below.

• BrainSuite and the BrainSuite Diffusion Pipeline

BrainSuite is a collection of image analysis tools designed to process magnetic resonance images (MRI) of the human brain. BrainSuite13 provides an automatic sequence to extract cortical surface mesh models from the MRI, tools to register these to a labeled atlas to define anatomical regions of interest, and tools for processing diffusion imaging data (including distortion correction, coregistration to anatomical data, ODF and tensor fitting, and tractography). BrainSuite also contains visualization tools for exploring this kind of data, and can produce interactive maps of regional white matter connectivity. The software package includes an implementation of the Funk-Radon and Cosine Transform, distortion correction, and diffusion tensor estimation methods described respectively in:

• J. P. Haldar, R. M. Leahy.
  Linear Transforms for Fourier Data on the Sphere: Application to High Angular Resolution Diffusion MRI of the Brain.
  NeuroImage 71:233-247, 2013.

• C. Bhushan, J. P. Haldar, A. A. Joshi, R. M. Leahy.
  Correcting Susceptibility-Induced Distortion in Diffusion-Weighted MRI using Constrained Nonrigid Registration.
  Asia Pacific Signal and Information Processing Association (APSIPA) Annual Summit and Conference, Hollywood, 2012.

• J. H. Kim, J. Haldar, Z.-P. Liang, S.-K. Song.
  Diffusion Tensor Imaging of Mouse Brain Stem and Cervical Spinal Cord.
  Journal of Neuroscience Methods 176:186-191, 2009.

• LORAKS 2.0: Sample Code and Examples

An implementation of the low-rank signal reconstruction methods described in:

• J. P. Haldar
  Low-Rank Modeling of Local k-Space Neighborhoods (LORAKS) for Constrained MRI.
  IEEE Transactions on Medical Imaging 33:668-681, 2014.

• T. H. Kim, K. Setsompop, J. P. Haldar.
  LORAKS Makes Better SENSE: Phase-Constrained Partial Fourier SENSE Reconstruction without Phase Calibration.
  Magnetic Resonance in Medicine 77:1021-1035, 2017.

• J. P. Haldar.
  Autocalibrated LORAKS for Fast Constrained MRI Reconstruction.
  IEEE International Symposium on Biomedical Imaging: From Nano to Macro, New York City, 2015, pp. 910-913.

and

• T. H. Kim, J. P. Haldar.
  LORAKS Software Version 2.0: Faster Implementation and Enhanced Capabilities.
  University of Southern California, Los Angeles, CA, Technical Report USC-SIPI-443, May 2018.

• Sample code for Low-Rank Modeling of Local k-Space Neighborhoods (LORAKS)

Outdated, but provided for legacy purposes. See LORAKS V2.0 above for a newer version. An implementation of the low-rank signal reconstruction method described in:

• J. P. Haldar
  Low-Rank Modeling of Local k-Space Neighborhoods (LORAKS) for Constrained MRI.
  IEEE Transactions on Medical Imaging 33:668-681, 2014.

and

• J. P. Haldar.
  Low-Rank Modeling of Local k-Space Neighborhoods (LORAKS): Implementation and Examples for Reproducible Research.
  University of Southern California, Los Angeles, CA, Technical Report USC-SIPI-414, April 2014.

• Nullspace-Based Sensitivity Map Estimation with PISCO

An implementation of the sensitivity map estimation methods described in:

• R. A. Lobos, C.-C. Chan, J. P. Haldar.
  New Theory and Faster Computations for Subspace-Based Sensitivity Map Estimation in Multichannel MRI.
  arXiv:2302.13431

and

• R. A. Lobos, C.-C. Chan, J. P. Haldar.
  PISCO Software Version 1.0.
  University of Southern California, Los Angeles, CA, Technical Report USC-SIPI-458, March 2023.

• Local Perturbation Responses: Code and Examples

An illustrative implementation of the image characterization approach described in:

• C.-C. Chan, J. P. Haldar.
  Local Perturbation Responses and Checkerboard Tests: Characterization tools for nonlinear MRI methods.
  Magnetic Resonance in Medicine. In Press.

• Sample code for the Fourier Radial Error Spectrum Plot (ESP)

An implementation of the new image quality metric described in:

• T. H. Kim, J. P. Haldar.
  The Fourier Radial Error Spectrum Plot: A more nuanced quantitative evaluation of image reconstruction quality.
  IEEE International Symposium on Biomedical Imaging, Washington, DC, 2018. In Press.

and

• T. H. Kim, J. P. Haldar.
  Assessing MR image reconstruction quality using the Fourier Radial Error Spectrum plot.
  Joint Annual Meeting ISMRM-ESMRMB, Paris, 2018.

• Implementations of Greedy Algorithms for Estimating Nonnegative and Simultaneously Sparse Signals

Implementations of the greedy algorithms described in:

• D. Kim, J. P. Haldar.
  Greedy Algorithms for Nonnegativity-Constrained Simultaneous Sparse Recovery.
  Signal Processing 125:274-289, 2016.

• Code for Joint Denoising of Diffusion MRI Magnitude Images

An implementation of a modified version of the regularized statistical denoising method described in:

• J. P. Haldar, V. J. Wedeen, M. Nezamzadeh, G. Dai, M. W. Weiner, N. Schuff, Z.-P. Liang.
  Improved Diffusion Imaging through SNR-Enhancing Joint Reconstruction.
  Magnetic Resonance in Medicine 69:277-289, 2013.

Specifically, the algorithm has been modified to handle magnitude images, as described in:

• D. Varadarajan, J. P. Haldar.
  A Majorize-Minimize Framework for Rician and Non-Central Chi MR Images.
  IEEE Transactions on Medical Imaging 2015. In Press.

• Sample code for Incremented Rank PowerFactorization (IRPF)

An implementation example for a variation on the IRPF matrix recovery algorithm described in:

• J. P. Haldar, D. Hernando.
  Rank-Constrained Solutions to Linear Matrix Equations using PowerFactorization.
  IEEE Signal Processing Letters 16:584-587, 2009.

This IRPF code can be used to implement variations on the algorithms presented in:

• J. P. Haldar, Z.-P. Liang.
  Spatiotemporal Imaging With Partially Separable Functions: A Matrix Recovery Approach.
  IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Rotterdam, 2010, pp. 716-719.
  Recipient of an IEEE ISBI 2010 Best Student Paper Award.

• B. Zhao, J. P. Haldar, C. Brinegar, Z.-P. Liang.
  Low Rank Matrix Recovery for Real-Time Cardiac MRI.
  IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Rotterdam, 2010, pp. 996-999.

• J. P. Haldar, Z.-P. Liang.
  Low-Rank Approximations for Dynamic Imaging.
  IEEE International Symposium on Biomedical Imaging: From Nano to Macro, Chicago, 2011, pp. 1052-1055.

• Sample code for Distortion Correction in Diffusion MRI with Interlaced q-Space Sampling

An implementation example for the reconstruction algorithm described in:

• C. Bhushan, A. A. Joshi, R. M. Leahy, J. P. Haldar.
  Improved B₀-distortion correction in diffusion MRI using interlaced q-space sampling and constrained reconstruction.
  Magnetic Resonance in Medicine 72:1218-1232, 2014.

This code not only enables accelerated distortion-corrected MRI using the novel interlaced phase encoding direction scheme, but also supports the popular reversed gradient approach to distortion correction.

• Illinois Massively Parallel Acceleration Toolkit for Image reconstruction with ENhanced Throughput in MRI (IMPATIENT MRI)

The IMPATIENT MRI toolkit is an open-source CUDA-based GPU implementation of iterative MR image reconstruction. IMPATIENT MRI is very flexible, and enables efficient solution of MRI reconstruction problems involving non-Cartesian data acquisition, field inhomogeneity modeling, and regularization. Our GPU implementation work is described in journal papers (listed below) and several conference papers (which can be found on the Publications page):

• J. Gai, N. Obeid, J. L. Holtrop, X.-L. Wu, F. Lam, M. Fu, J. P. Haldar, W.-m. W. Hwu, Z.-P. Liang, B. P. Sutton.
  More IMPATIENT: A Gridding-Accelerated Toeplitz-based Strategy for Non-Cartesian High-Resolution 3D MRI on GPUs.
  Journal of Parallel and Distributed Computing 2013, In Press.

• S. S. Stone, J. P. Haldar, S. C. Tsao, W.-m. W. Hwu, B. P. Sutton, Z.-P. Liang.
  Accelerating Advanced MRI Reconstructions on GPUs.
  Journal of Parallel and Distributed Computing 68:1307-1318, 2008.

The IMPATIENT MRI toolkit can be used to implement many of our published methods, including methods used to produce some of the results shown the following incomplete list of journal papers:

• J. P. Haldar, V. J. Wedeen, M. Nezamzadeh, G. Dai, M. W. Weiner, N. Schuff, Z.-P. Liang.
  Improved Diffusion Imaging through SNR-Enhancing Joint Reconstruction.
  Magnetic Resonance in Medicine 69:277-289, 2013.

• J. P. Haldar, D. Hernando, Z.-P. Liang.
  Compressed-Sensing MRI with Random Encoding.
  IEEE Transactions on Medical Imaging 30:893-903, 2011.

• J. P. Haldar, D. Hernando, S.-K. Song, Z.-P. Liang.
  Anatomically Constrained Reconstruction from Noisy Data.
  Magnetic Resonance in Medicine 59:810-818, 2008.

---