PETFit: A BIDS App for PET Kinetic Modelling

PETFit is a BIDS App for fitting kinetic models to PET time activity curve (TAC) data. It takes preprocessed PET data in BIDS format and runs a configurable kinetic modelling pipeline, producing parameter estimates and detailed HTML reports for quality control.

Note

PETFit is currently in active development (v0.1.3). If you encounter any bugs, please report them on the GitHub issues page — they are extremely valuable for making this pipeline robust.

How it works

PETFit can be used in two ways:

• Interactive mode: Use graphical web applications to configure your analysis step-by-step, preview results, and generate configuration files.

• Automatic mode: Run the full pipeline non-interactively using a pre-defined configuration file. This is ideal for re-running analysis on updated data, or HPC environments.

The typical workflow has two stages:

1. Region definition — Combine individual brain regions from PET preprocessing derivatives into analysis-ready TACs.

2. Kinetic modelling — Configure and run kinetic models on the combined TACs, with steps for data subsetting, weights calculation, delay fitting (plasma input) or reference TAC setup (reference tissue), and model fitting.

Three apps

PETFit provides three separate Shiny web applications for interactive mode:

• Region Definition App — Creates combined regional TACs from PET preprocessing derivative data.

• Modelling App with Plasma Input — Configures invasive kinetic models (1TCM, 2TCM, Logan, MA1, Patlak) that require blood input data.

• Modelling App with Reference Tissue — Configures non-invasive kinetic models (SRTM, refLogan, MRTM1, MRTM2) that use a reference region.

Getting started

Installation

Pull the Docker image, build an Apptainer container, or install the R package.

Installation

Quick start

A minimal end-to-end example to get you up and running.

Quick start

Supported models

Full reference for all invasive and non-invasive kinetic models.

Supported models

Folder structures

How PETFit organises region definitions, analyses, and ancillary workflows.

PETFit folder structures

Outputs

What files PETFit produces and how they are organised.

Outputs

Getting Started

• Installation
  • Docker
  • Apptainer
  • R package (for development)
• Quick start
  • Step 1: Define regions
  • Step 2: Run kinetic modelling
  • Step 3: Review reports
  • Key arguments

User Guide

• Usage guide
  • The two-step workflow
  • The three apps
  • Interactive vs automatic mode
  • Configuration files
  • Analysis folders
• Container usage
  • Docker
  • Apptainer
• Outputs
  • Directory structure
  • Region definition outputs
  • Modelling outputs
  • BIDS metadata integration
• Supported models
  • Plasma input models
  • Reference tissue models

Reference

• Troubleshooting
  • Data and file issues
  • Reference tissue issues
  • Blood data issues
  • Model fitting issues
  • Ancillary analysis issues
  • Performance
  • Docker issues
  • Apptainer / HPC issues
• API reference
  • Interactive apps
  • Automatic pipelines
  • Container command-line options
• Citation
  • kinfitr references
• Changelog
  • 0.1.5 (current)
  • 0.1.4
  • 0.1.3
  • 0.1.2
• License

Development

• Contributing
  • Development setup
  • Repository structure
  • Coding standards
  • Testing
  • Configuration management
  • Building documentation