Skip to content

DCEPrep

A preprocessing and analysis pipeline for Dynamic Contrast-Enhanced (DCE) MRI data.

DCEPrep handles the full workflow for measuring permeability (Ktrans) from DCE-MRI acquisitions — from raw DICOM to publication-ready QC reports.

Overview of DCEPrep processing steps

What Does DCEPrep Do?

DCEPrep takes BIDS-organized MRI data and runs a two-phase pipeline:

Phase 1 — Preprocessing (preprocess_all.sh) Prepares raw DCE and VFA images for kinetic modeling. Steps include brain extraction, image registration, bias field correction, z-axis normalization, T1 mapping, and arterial input function (AIF) selection.

Phase 2 — Analysis (DCE_all.sh) Runs pharmacokinetic modeling via ROCKETSHIP to produce Ktrans permeability maps, then generates per-case and population-level HTML/PDF QC reports.

Key Features

  • Fully scriptable — driven by two shell scripts with straightforward CLI flags
  • BIDS-native — expects and produces BIDS-compliant data structures
  • Docker-first — a single container bundles FSL, ANTs, FreeSurfer, MATLAB, and ROCKETSHIP
  • Automated AIF — optional neural-network-based arterial input function detection
  • Variance Reduction — z-axis normalization and bias correction designed to reduce scanner variability
  • Automated QC — per-case and population HTML reports with outlier detection
  • GPU acceleration — Several steps utilize optional CUDA-based acceleration for faster processing

How DCEPrep Fits In

DCEPrep is part of the petmri processing ecosystem. It sits downstream of DICOM conversion and upstream of group-level statistical analysis:

flowchart TD
    A[DICOM acquisition]
    B["BIDS conversion\ndce2bids (or sort_dicom.py)"]
    G[Group-level statistics]

    subgraph DCEPrep["DCEPrep"]
        PRE["preprocess_all.sh\n─────────────────\n○  Bias field correction\n○  Image Registration\n○  Intensity Normalization\n○  Motion Correction\n○  T1 mapping\n○  AIF selection\n○  QC reports"]
        ANA["DCE_all.sh\n─────────────────\n○  Ktrans fitting w/ ROCKETSHIP\n○  QC reports\n○  Population-level summaries"]
        PRE --> ANA
    end

    A --> B --> PRE
    ANA --> G

    style A fill:#757575,stroke:#555,color:#fff
    style B fill:#F9A825,stroke:#F57F17,color:#000
    style G fill:#757575,stroke:#555,color:#fff
    style PRE fill:#C8DCEE,stroke:#7AAAC8,color:#1a2a3a
    style ANA fill:#C8DCEE,stroke:#7AAAC8,color:#1a2a3a
Color Meaning
Blue Handled by DCEPrep
Amber Partially supported (dce2bids, or the bundled sort_dicom.py)
Gray Outside DCEPrep scope

See ROCKETSHIP Integration for details on the MATLAB dependency.

The petmri Software Family

DCEPrep is one of several tools maintained by the petmri group for DCE-MRI processing, including ROCKETSHIP, AutoAIF, and dce2bids. See petmri.github.io for an overview of the full software family.

Citation

If you use DCEPrep in your research, please cite:

Barnes S, et al. Automated DCE-MRI processing with DCEPrep for Blood-Brain Barrier permeability in a multi-site aging study. Pending. 2026.