Project EMCqMRI

EMCqMRI is a PyTorch-based, open-source framework.

The EMCqMRI package was designed with the goal to accelerate initial developments in the field of quantitative MRI (qMRI).

It provides base classes for the implementation of task-specific custom modules and offers a framework for training Deep Learning inference models and prediction of quantitative maps.

Validated qMRI methods, including signal models, likelihood functions and inference methods, are also included within this package.

• Project status: Development

Getting started

The codebase is currently under active development

Main features: - flexible creation of custom modules; - compositional & portable APIs for ease of integration in existing workflows; - domain-specific implementations for networks, losses, evaluation metrics and more;

The EMCqMRI package is distributed with PyPI. Details on environment preparation, PyTorch installation, and external dependencies can be found in Installation.

Technical documentation is available at emcqmri.readthedocs.io.

Features

• Feature highlights

APIs

• Signal Module
• Dataset Module
• Inference Module
• Likelihood Module

Installation

• Installation guide

Setting up a new project

• Tutorial: Setting up a new project with EMCqMRI

Configuration files and binding options

• Configuration files
• The EMCqMRI Configuration Object
• Linking and overriding configurations via command line

Indices and tables

• Index

• Module Index