A pipeline for whole heart segmentation and three-dimensional reconstruction from cine MRI, taking two-chamber, four-chamber and short-axis views.
Distributed as a Docker image, so the pipeline runs the same way on a laptop and on a cluster, with a wrapper script as the only file to download. It runs on CPU by default and on GPU when one is available.
A Python toolkit for building personalised whole heart models, covering segmentation handling, meshing and model post-processing in one reproducible workflow.
The pipelines are modular, so segmentation, meshing and model creation can be run separately or end to end, and external command line tools are driven through a common runner rather than by hand. It is used to produce the four-chamber meshes the group publishes.
A platform with custom image processing and computer vision toolkits for applying statistical, machine learning and simulation approaches to cardiovascular data.
An interactive medical imaging application built on MITK, with image processing and computer vision toolkits for cardiovascular research. It is free and open source under a 3-clause BSD licence, and it is not for use in the diagnosis or treatment of patients. Projects that use it are asked to cite Razeghi et al. (2020), SoftwareX 12:100570.
A deep learning 3D U-Net that reconstructs left atrial shape, volume and surface area from two- and four-chamber views, outperforming the bi-plane area-length method in accuracy and robustness.
An open source solution for electrophysiology data analysis, providing data parsing and analysis tools for electrophysiology research.
OpenEP converts data from proprietary electroanatomic mapping systems into an open format and provides tools to analyse it. It has been developed at King’s College London, Imperial College London, the University of Edinburgh and the University of Birmingham.
Steven Niederer, Cesare Corrado and Caroline Roney of the group are among the authors of the paper that introduced it (Williams et al., 2021, Frontiers in Physiology).
A GPU-accelerated simulator for cardiac electrophysiology, using the finite element method and built on PyTorch.
It runs monodomain and reaction-eikonal simulations on large heart meshes, supports a range of ionic models and conductivity heterogeneity, and produces local activation and repolarisation time maps and 12-lead ECGs. It installs with pip, is released under the MIT licence, and fits into PyTorch and scientific machine learning workflows.
Bei Zhou of the group created it, and the paper that introduced it is by Bei Zhou, Maximilian Balmus, Cesare Corrado, Ludovica Cicci, Shuang Qian and Steven Niederer (Zhou et al., 2026, SoftwareX).
The group publishes its code under OpenHeartDevelopers, where CemrgApp, HeartBuilder and CardioForm are developed alongside the libraries and tools that support them.