Revision cc7055c78202813215e6c607255bdfdf6e088dca authored by Konrad Werys on 01 November 2019, 12:51:09 UTC, committed by Konrad Werys on 01 November 2019, 12:51:09 UTC
1 parent 00b20ae
Raw File
TOMATO (Total Mapping Toolbox) is a C++ library for the calculation of parametric maps in cardiac magnetic resonance imaging (MRI). As an open source project, TOMATO allows transparent and standardized cardiac longitudinal relaxation time (T1) mapping in clinical applications. With C++ implementation, TOMATO can easily interface and translate between research software environments like MATLAB, and commercial vendors’ closed-source C++ environments on scanners as well as post-processing software. To complement the core library implementation, a ready-to-use command line tool has been provided. 

It contains ShMOLLI implementation as in [this article](  

*   **Code documentation** in doxygen can be found [here](
*   **Github** repository can be found [here](  

| System                                    |     Status                                                         |
| [DOI][zenodo_link]                        | [![DOI][zenodo_badge]][zenodo_link]                                |
| [OSX/Linux build - Travis][travis_link]   | [![Travis (.org)][travis_badge]][travis_link]                      |
| [Windows build - AppVeyor][appveyor_link] | [![AppVeyor][appveyor_badge]][appveyor_link]                       |
| [Code quality - Codacy][codacy_link]      | [![Codacy Badge][codacy_badge]][codacy_link]                       |
| [Test coverage - Codecov][codecov_link]   | [![codecov][codecov_badge]][codecov_link]                          |
| [Downloads][downloads_link]               | [![GitHub Releases (by Release)][downloads_badge]][downloads_link] |

## Usage

*   App - download the latest version of `TomatoExe` from the [Releases][downloads_link] page and follow the youtube tutorial below
[![Tomato tutorial](](

*   Lib - download the latest version of `TomatoLib` from the [Releases][downloads_link] page

*   Code - if you want to build the code please take a look at the scripts in the `scriptsBuild` and the continuous integrations scripts `.appveyor.yml` and `travis.yml`

## Tomato and Numerical Recipes

Original ShMOLLI (and based on it Tomato) uses code based on Numerical Recipes book. Due to Numerical Recipes licence I cannot share this part of the code online. Please contact me if you would like to use this part of the code. To make up for this limitation we provide a number of alternative fitting procedures.

## Troubleshooting

*   missing msvcp140.dll  
Download <> as suggested in <>

## Software Requirements Specification

1.  Calculate ShMOLLI with exactly the same results as these from the previous implementation.

2.  API for easily choosing
    *   a map calculation algorithm,
    *   a fitting algorithm,
    *   a model function and corresponding energy function.
3.  Compatibility with:
    *   main operating systems (Windows, MacOS, Linux),
    *   C++98 (to include the library in older code).

4.  Handling DICOM files.

5.  Fast calculation. 

## Contributing and code of conduct

Please see []( and [](

## Notes

*   **Continuous integration** was tested [in this repo](
*   **Gitmodules**: after `git clone` run `git submodule update --init --recursive` to get the submodules code.
*   It would be nice to follow [CppCoreGuidlines]( as far as C++98 allows.
*   Maybe I can write a script for github releases with [github releases REST api]( and [curl REST api](
*   LMFIT - somehow windows does not like the static library, so the dynamic is used


*   add acceptance test for 2param
*   maybe rename calculatormolli to calculator3param? 
*   maybe add calculator2param to produce different results?
*   update comments in configuration yaml files
*   re-organise start point calculation
*   output SNR maps
*   make sure SNR and nAmoebaCalls is the same as in Shmolli
*   fix downloadITK_linux_osx
*   separate projects for building tomato dependencies (done for lmfit)
*   possibly I could make all FunctionsT1 object static
*   write user-friendly build scripts (and use them in CI)

back to top