UCL Centre for Medical Image Computing


Software and Resources

Since its launch in 2005, the UCL Centre for Medical Image Computing has generated a wealth of open-source software and resources.

Please find below a sample of our currently most popular packages (you can test some of them online using NiftyWeb service). Follow the links to each package for information specific to a package. Please also refer to the Research Group section of our website if you are looking for packages created by a specific team of developers within our Centre.

brainageR - the brainageR software package implements the brain-age paradigm using structural neuroimaging data. Based on a trained model of over 3000 healthy adults, brainageR will generate a “brain-predicted age” value from a raw T1-weighted MRI scan. This uses a Gaussian Processes regression, implemented in R (https://github.com/james-cole/brainageR). Alongside a brain-predicted value, the software outputs brain tissue volume measurements (from SPM12) and snapshot images (from FSL slicesdir) to enable straightforward quality control. The brain-predicted age of an individual can then be compared to their chronological age to determine the brain-predicted age difference (brain-PAD) score, which can be used as an age-adjusted biomarker of brain health.

BrainPainter is a tool for colouring brain images using any user-defined input. For each brain region it takes values from a 0-1 (or 0-max), and colours the brain regions according to these numbers. Numbers could represent biomarkers, levels of atrophy, or absolutely anything. https://github.com/ucl-pond/brain-coloring

UCL Camino Diffusion MRI Toolkit - Camino is an open-source software toolkit for diffusion MRI processing. The toolkit implements standard techniques, such as diffusion tensor fitting, mapping fractional anisotropy and mean diffusivity, deterministic and probabilistic tractography. It also contains more specialized and cutting-edge techniques, such as Monte-Carlo diffusion simulation, multi-fibre and HARDI reconstruction techniques, multi-fibre PICo, compartment models, and axon density and diameter estimation. http://camino.cs.ucl.ac.uk

cid-X: XCAT DVF post processing and inverse computation - Consistent and invertible deformation vector fields for a breathing anthropomorphic phantom: a post-processing framework for the XCAT phantom. Project page: https://github.com/UCL/cid-X

CT-based imaging biomarkers - The RILD biomarkers project consists of set of image analysis tools to quantify radiation-induced lung damage (RILD) for academic research use. Project page: https://github.com/CMIC-RT/RILD_biomarkers

DeepReg is a freely available, community-supported, open-source toolkit for research and education in medical image registration using deep learning. The current version is implemented as a TensorFlow2 based framework and contains implementations for unsupervised and weakly-supervised algorithms with their combinations and variants. DeepReg has a practical focus on growing and diverse clinical applications, as seen in the provided examples - DeepReg Demos. https://deepreg.readthedocs.io/en/latest/

DTI-TK is a spatial normalization & atlas construction toolkit, designed from the ground up to support the manipulation of diffusion-tensor images (DTI) with special cares taken to respect the tensorial nature of the data. It implements a state-of-the-art registration algorithm that drives the alignment of white matter (WM) tracts by matching the orientation of the underlying fiber bundle at each voxel. The algorithm has been shown to both improve WM tract alignment and to enhance the power of statistical inference in clinical settings. http://dti-tk.sourceforge.net/pmwiki/pmwiki.php

Image Quality Transfer (IQT) aims to bridge the technological gap that exists between bespoke and expensive experimental systems such as the Human Connectome Project (HCP) scanner and accessible commercial clinical systems using machine learning (ML). The technique learns mappings from low-quality (e.g. clinical) to high-quality (e.g. experimental) images exploiting the similarity of images across subjects, regions, modalities, and scales: image macro- and meso-structure is highly predictive of sub-voxel content. The mapping may then operate directly on low-quality images to estimate the corresponding high-quality images, or serve as a prior in an otherwise ill-posed image-reconstruction routine. https://github.com/ucl-mig/iqt

UCL Medical Software Quality Management System (QMS) provides a comprehensive environment for the development of medical software and software-based medical devices in a way that complies with the ISO-13485 standard.  It is a strategic initiative of researchers in the UCL Centre for Medical Image Computing and is now embedded within the UCL Wellcome-EPSRC Centre for Interventional & Surgical Sciences (WEISS). For more information of the resources available as part of the QMS and the projects it currently supports, please contact Sarina Hussain (sarina.hussain@ucl.ac.uk) or Prof. Dean Barratt (d.barratt@ucl.ac.uk).

MISST - Microstructure Imaging Sequence Simulation ToolBox - Microstructure Imaging Sequence Simulation Toolbox (MISST) is a practical diffusion MRI simulator for development, testing, and optimisation of novel MR pulse sequences for microstructure imaging. MISST is based on a matrix method approach and simulates the signal for a large variety of pulse sequences and tissue models. Its key purpose is to provide a deep understanding of the restricted diffusion MRI signal for a wide range of realistic, fully flexible scanner acquisition protocols in practical computational time. http://mig.cs.ucl.ac.uk/index.php?n=Tutorial.MISST

NifTK package - A suit containing image processing tools, biomarkers like BSI and viewers like NiftyView, NiftyMITK, NiftyIGI and the open source part of NiftyMIDAS. This repository is open-sourced under BSD. https://cmiclab.cs.ucl.ac.uk/CMIC/NifTK - http://niftk.org

NODDI (Practical in vivo neurite orientation dispersion and density imaging of the human brain) is a practical diffusion MRI technique for estimating the microstructural complexity of dendrites and axons in vivo on clinical MRI scanners.https://www.nitrc.org/projects/noddi_toolbox

POSSUM (Physics-Oriented Simulated Scanner for Understanding MRI) is a software tool that produces realistic simulated MRI and FMRI images or time series. POSSUM is part of FSL (FMRIB's Software Library). POSSUM has an easy-to-use graphical user interface (GUI) and its component programs can also be run from the command line. POSSUM includes tools for the pulse sequence generation, signal generation, noise addition and image reconstruction.https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/POSSUM

SciKit-Surgery is a collection of open-source libraries for surgical navigation. The libraries are largely written in Python, and their modular nature allows the researcher to rapidly assemble their own clinical application. https://www.software.ac.uk/blog/2020-07-13-scikit-surgery-compact-libraries-surgical-navigation.

SnappySonic is part of the SciKit-Surgery project. SnappySonic is an ultrasound acquisition replay simulator. The output from a tracking system (NDI or ArUco tags) is used to select a frame of recorded video to show. A suitable video of ultrasound data is included in the data directory. However, different videos could be used. https://openresearchsoftware.metajnl.com/articles/10.5334/jors.289

The Spherical Mean Technique (SMT) is a clinically feasible method for microscopic diffusion anisotropy imaging. The purpose is to map microscopic features unconfounded by the effects of fibre crossings and orientation dispersion, which are ubiquitous in the brain. This technique requires only an off-the-shelf diffusion sequence with two (or more) b-shells achievable on any standard MRI scanner, thus facilitating its widespread use in neuroscience research and clinical neurology.https://github.com/ekaden/smt

Subtype and Stage Inference (SuStaIn) infers disease progression subtypes from cross-sectional datasets. https://github.com/ucl-pond/pysustainIf you use this code, please cite: https://doi.org/10.1038/s41467-018-05892-0

SuPReMo implements the unified motion modelling and image registration framework as proposed by Jamie McClelland et al. and was developed by Bjoern Eiben and the Radiotherapy Image Computing Group at the Centre for Medical Image Computing, University College London, UK. The source code is available on github and the documentation is hosted on the corresponding github-pages.

TOAST is an open-source software library for diffuse optical tomography and related modalities. It contains a forward solvers using the finite element method for simulating the propagation of light in highly scattering, inhomogeneous biological tissues and inverse solvers for reconstruction. The library is written in C++ with bindings to Matlab and Python. http://web4.cs.ucl.ac.uk/research/vis/toast/

You may also want to find out more about the following:

STIR is open-source software for tomographic image reconstruction [1]. Its aim is to provide a Multi-Platform Object-Oriented framework for all data manipulations in tomographic imaging. Currently, the emphasis is on (iterative) image reconstruction in PET and SPECT, but other application areas and imaging modalities can be added. (http://stir.sourceforge.net/)
More on software developed by our colleagues at UCL Institute of Nuclear Medicine

UCL XNAT Service is an open-source imaging informatics platform developed by the Neuroinformatics Research Group at Washington University. It facilitates management, archiving, sharing and analysing medical imaging data and associated files. (https://www.ucl.ac.uk/isd/services/research-it/ucl-xnat-service)

More on the software services offered by our colleagues in the Medical Imaging Research Software Group here:

Fusilli is a collection of deep-learning based models for multimodal data fusion, which is the process of combining different data modalities together for a common predictive task. For example, combining brain MRI with clinical data to predict disease progression. Fusilli is a complete pipeline for training, evaluating, and comparing different multimodal data fusion methods, and no expertise in deep learning is required. The models and training specifics in Fusilli are completely customisable.

Here is the documentation: https://fusilli.readthedocs.io/en/latest/ and the GitHub: https://github.com/florencejt/fusilli