DIRECT: Deep Image REConstruction Toolkit
Installation • Quick Start • Documentation • Model Zoo
DIRECT is a Python, end-to-end pipeline for solving Inverse Problems emerging in Imaging Processing. It is built with PyTorch and stores state-of-the-art Deep Learning imaging inverse problem solvers such as denoising, dealiasing and reconstruction. By defining a base forward linear or non-linear operator, DIRECT can be used for training models for recovering images such as MRIs from partially observed or noisy input data. DIRECT stores inverse problem solvers such as the vSHARP, Learned Primal Dual algorithm, Recurrent Inference Machine and Recurrent Variational Network, which were part of the winning solutions in Facebook & NYUs FastMRI challenge in 2019, the Calgary-Campinas MRI reconstruction challenge at MIDL 2020 and the CMRxRecon challenge 2023. For a full list of the baselines currently implemented in DIRECT see here.
Zero-filled reconstruction, Compressed-Sensing (CS) reconstruction using the BART toolbox, Reconstruction using a RIM model trained with DIRECT
Quick install
DIRECT is published to PyPI as direct-recon (the import package is still direct):
pip install direct-reconFor development, the recommended way is with uv:
git clone https://github.com/NKI-AI/direct.git
cd direct
uv sync # runtime + dev (default groups)See the installation guide for Docker and conda alternatives.
Projects
In the projects folder baseline model configurations are provided for each project.
Baselines and trained models
We provide a set of baseline results and trained models in the DIRECT Model Zoo. Baselines and trained models include the vSHARP, Recurrent Variational Network (RecurrentVarNet), the Recurrent Inference Machine (RIM), the End-to-end Variational Network (VarNet), the Learned Primal Dual Network (LDPNet), the X-Primal Dual Network (XPDNet), the KIKI-Net, the U-Net, the Joint-ICNet, and the AIRS Medical fastmri model (MultiDomainNet).
License and usage
DIRECT is not intended for clinical use. DIRECT is released under the Apache 2.0 License.
Citing DIRECT
If you use DIRECT in your own research, or want to refer to baseline results published in the DIRECT Model Zoo, please use the following BiBTeX entry:
@article{DIRECTTOOLKIT,
doi = {10.21105/joss.04278},
url = {https://doi.org/10.21105/joss.04278},
year = {2022},
publisher = {The Open Journal},
volume = {7},
number = {73},
pages = {4278},
author = {George Yiasemis and Nikita Moriakov and Dimitrios Karkalousos and Matthan Caan and Jonas Teuwen},
title = {DIRECT: Deep Image REConstruction Toolkit},
journal = {Journal of Open Source Software}
}1 maintainer
Release files for direct-recon 2.2.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| direct_recon-2.2.0.tar.gz | 14.2 MB | Details |
Built distributions (wheels)
| File | Reset | |||
|---|---|---|---|---|
| direct_recon-2.2.0-cp313-cp313-win_amd64.whl | CPython 3.13 | CPython 3.13 | Windows x86-64 | Details |
| direct_recon-2.2.0-cp312-cp312-win_amd64.whl | CPython 3.12 | CPython 3.12 | Windows x86-64 | Details |
| direct_recon-2.2.0-cp312-abi3-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl | CPython 3.12 | abi3 | Linux glibc 2.28+ x86-64, Linux glibc 2.24+ x86-64 | Details |
| direct_recon-2.2.0-cp312-abi3-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl | CPython 3.12 | abi3 | Linux glibc 2.28+ ARM64, Linux glibc 2.24+ ARM64 | Details |
| direct_recon-2.2.0-cp312-abi3-macosx_11_0_arm64.whl | CPython 3.12 | abi3 | macOS 11.0+ ARM64 | Details |
| direct_recon-2.2.0-cp312-abi3-macosx_10_13_x86_64.whl | CPython 3.12 | abi3 | macOS 10.13+ x86-64 | Details |
Total release size: 17.6 MB
Release files / direct_recon-2.2.0.tar.gz
| Download URL | direct_recon-2.2.0.tar.gz |
|---|---|
| Size | 14.2 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
dd19bb6254baf5d5eee32d13b3764e2250eaa0cfe8a69d348d4331991f800ccf
|
|
BLAKE2b-256 checksum How to use checksums |
8ea2ae42a897362247922af664b631e1cbf0483ad5da5110c0ce5e6725acf6f4
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Aug 21, 2026.
Transparency logRelease files / direct_recon-2.2.0-cp313-cp313-win_amd64.whl
| Download URL | direct_recon-2.2.0-cp313-cp313-win_amd64.whl |
|---|---|
| Size | 574.6 kB |
| Tags | CPython 3.13 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
e3b301e572bdb10a5d90a60bc04f59298d2eb16676a88e3f5d44b145e6fe416c
|
|
BLAKE2b-256 checksum How to use checksums |
3d3174b3d7308fbe0a5bccb19010acc4891074f83452dbf3eff3ad31d1a2012e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Aug 21, 2026.
Transparency logRelease files / direct_recon-2.2.0-cp312-cp312-win_amd64.whl
| Download URL | direct_recon-2.2.0-cp312-cp312-win_amd64.whl |
|---|---|
| Size | 574.6 kB |
| Tags | CPython 3.12 Windows x86-64 |
|
SHA-256 checksum How to use checksums |
4bd490e5f7dc1a529977e564f171636d9b2fb39d69bda135fcfffb6ed8929945
|
|
BLAKE2b-256 checksum How to use checksums |
1002b97096ac54a5d5be71a282459744f73ff8ea7d2f5a3bbbdf08dbb7b7bc0b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Aug 21, 2026.
Transparency logRelease files / direct_recon-2.2.0-cp312-abi3-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl
| Download URL | direct_recon-2.2.0-cp312-abi3-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl |
|---|---|
| Size | 577.7 kB |
| Tags | CPython 3.12 Linux glibc 2.24+ x86-64 Linux glibc 2.28+ x86-64 abi3 |
|
SHA-256 checksum How to use checksums |
670b397aefec01ab9b855e3ddccc412f82371782aa5ed47be56190accde9ec49
|
|
BLAKE2b-256 checksum How to use checksums |
d2295444376ecbc2216751cff328c88eff610593b9a1800d70871b1f592100b7
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Aug 21, 2026.
Transparency logRelease files / direct_recon-2.2.0-cp312-abi3-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl
| Download URL | direct_recon-2.2.0-cp312-abi3-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl |
|---|---|
| Size | 578.9 kB |
| Tags | CPython 3.12 Linux glibc 2.24+ ARM64 Linux glibc 2.28+ ARM64 abi3 |
|
SHA-256 checksum How to use checksums |
9ea4975df88f8077e99d4838e8bc9265bca276234ff56b7891b28de6945c7d24
|
|
BLAKE2b-256 checksum How to use checksums |
c67bfbd723c93d6aa39598c62e9667dba74c7e547054bccf58d6e9301e629cd3
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Aug 21, 2026.
Transparency logRelease files / direct_recon-2.2.0-cp312-abi3-macosx_11_0_arm64.whl
| Download URL | direct_recon-2.2.0-cp312-abi3-macosx_11_0_arm64.whl |
|---|---|
| Size | 556.2 kB |
| Tags | CPython 3.12 abi3 macOS 11.0+ ARM64 |
|
SHA-256 checksum How to use checksums |
4ce316f043aa3e302bf8f58533f3db1a3a7306dcac1c4b177b5683701bb282b4
|
|
BLAKE2b-256 checksum How to use checksums |
06aee58184622f9004dd29f9b79689c2f3d609a202033e884371add7b4ebf534
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Aug 21, 2026.
Transparency logRelease files / direct_recon-2.2.0-cp312-abi3-macosx_10_13_x86_64.whl
| Download URL | direct_recon-2.2.0-cp312-abi3-macosx_10_13_x86_64.whl |
|---|---|
| Size | 564.7 kB |
| Tags | CPython 3.12 abi3 macOS 10.13+ x86-64 |
|
SHA-256 checksum How to use checksums |
3341eb393111d98c5d0048373bcd7783c9730e0ac74ba45bd0403bd7788076a2
|
|
BLAKE2b-256 checksum How to use checksums |
c4d7beda6dad2f169314947303ffba65af5b9c963321e79b9d73f23332f2744b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Aug 21, 2026.
Transparency log
