QSMxT

A fast, BIDS-native Quantitative Susceptibility Mapping (QSM) pipeline and interactive TUI, built in Rust. QSMxT takes you from raw scanner DICOMs to quantitative susceptibility maps with a single self-contained binary — no environment to manage, no toolbox sprawl.

📖 Documentation: https://qsmxt.github.io/QSMxT/

The website has everything: installation, guides, the command reference, and algorithm details. This README is just a quick pointer.

Highlights

• End-to-end pipeline — masking, phase unwrapping, echo combination, background field removal, dipole inversion, and referencing.
• Interactive TUI — convert, configure, and run from the terminal; it also shows the equivalent CLI command as you go.
• DICOM → BIDS — built-in conversion with automatic series classification.
• 10 inversion algorithms, 8 background-field methods, flexible masking, plus SWI / T2* / R2* outputs.
• Built for scale — disk caching, memory-aware parallelism, and SLURM support.

All reconstruction algorithms are provided by QSM.rs.

Install

# Linux / macOS
curl -fsSL https://raw.githubusercontent.com/QSMxT/QSMxT/main/install.sh | sh

# Windows (PowerShell)
irm https://raw.githubusercontent.com/QSMxT/QSMxT/main/install.ps1 | iex

No install needed to experiment — try it in your browser. See the installation guide for manual downloads, building from source, and details.

Quick start

# Easiest: do everything interactively (convert, configure, run)
qsmxt tui

# Or on the command line:
qsmxt dicom-convert /path/to/dicoms study/bids   # DICOM → BIDS
qsmxt run study/bids                             # run the pipeline

Full guides and the complete command reference are on the documentation site (or run qsmxt --help).

Coming from QSMxT 8.x (Python)?

v9 is a ground-up rewrite in Rust focused on QSM reconstruction and DICOM/NIfTI → BIDS conversion. Some 8.x workflows (segmentation, template building, group analysis, NextQSM, the web UI) are not part of v9 — the Python line remains available on the python-legacy branch and the v8.3.2 tag. See what changed.

Citation

Stewart AW, Robinson SD, O'Brien K, et al. "QSMxT: Robust masking and artifact reduction for quantitative susceptibility mapping." Magnetic Resonance in Medicine 87.3 (2022): 1289–1300. https://doi.org/10.1002/mrm.29048

License

GPL-3.0-only.