• The GUI-based MATLAB toolbox including algorithms for magnetic susceptibility source separation based on convex optimization (χ-separation or chi-separation; H. Shin et al., Neuroimage, 2021) and deep learning-based reconstruction (χ-sepnet; M. Kim at al., 30th Annual Meeting of ISMRM, 2022). The toolbox also supports phase preprocessing (e.g. phase unwrapping and background removal) powered by MEDI and STI Suite toolboxs (see the manual for details) and QSM reconsturction using deep neural network (QSMnet; J. Yoon et al., Neuroimage, 2018).

• The χ-separation toolbox includes the following features:

  • DICOM/NIFTI/MATLAB data compatibility
  • QSMnet: Quantitative susceptibility mapping (QSM) reconstruction algorithm based on deep neural network (QSMnet; J. Yoon et al., Neuroimage, 2018)
  • χ-separation using R₂' (or R₂* ): Magnetic susceptibility source separation algorithms based on convex optimization (χ-separation; H. Shin et al., Neuroimage, 2021) that share similar contrasts and optimization parameters with either MEDI+0 (Liu et al., MRM, 2018) or iLSQR (Li et al., Neuroimage, 2015) algorithms. The toolbox also provides the option to use pseudo R₂ map if R₂ measurement is not availabe (using R₂' is reconmmanded for accurate estimation).
  • χ-sepnet using R₂' (or R₂* ): A U-Net-based neural network that reconstructs COSMOS-quality χ-separation using R₂' and phase (χ-sepnet; M. Kim at al., 30th Annual Meeting of ISMRM, 2022). In case R₂ is not measured, another neural network is trained to estimate χ-separation maps from R₂* and phase.

• Last update: May-26-2023 (Hyeong-Geol Shin, Jun-Hyeok Lee, Minjoon Kim, Kyeongseon Min)

• H. Shin, J. Lee, Y. H. Yun, S. H. Yoo, J. Jang, S.-H. Oh, Y. Nam, S. Jung, S. Kim, F. Masaki, W. Kim, H. J. Choi, J. Lee. χ-separation: Magnetic susceptibility source separation toward iron and myelin mapping in the brain. Neuroimage, 2021 Oct; 240:118371.
• M. Kim, H. Shin, C. Oh, H. Jeong, S. Ji, H. An, J. Kim, J. Jang, B. Bilgic, and J. Lee, "Chi-sepnet: Susceptibility source separation using deep neural network", 30th Annual Meeting of International Society of Magnetic Resonance in Medicine, 2022; 2464.
• J. Yoon, E. Gong, I. Chatnuntawech, B. Bilgic, J. Lee, W. Jung, J. Ko, H. Jung, K. Setsompop, G. Zaharchuk, E.Y. Kim, J. Pauly, J. Lee. Quantitative susceptibility mapping using deep neural network: QSMnet. Neuroimage. 2018;179:199-206

• Common

  • MATLAB (tested in R2019a-R2021a)

• Additional

  • For QSMnet and χ-sepnet, Deep Learning MATLAB Toolbox Converter for ONNX Model Format (https://www.mathworks.com/matlabcentral/fileexchange/67296-deep-learning-toolbox-converter-for-onnx-model-format)
  • For DICOM/NIFTI read and phase processing, see https://www.dropbox.com/sh/3zafav50bfnruuu/AABVVYpdsznsRXKy8YKK4ybla?dl=0

We provide software license for academic research purpose only and NOT for commercial or clinical use. To renew the license, please contact [email protected]. For commercial use of our software, contact us ([email protected]) with the following information:

Name:

Affiliation:

Software:

Purpose:

• [email protected]
• [email protected] (Hyeong-Geol Shin, PhD)