A collection of TensorFlow add-ons for computational MRI.
Project description
TensorFlow MRI is a collection of TensorFlow add-ons for computational MRI.
It contains functionality including:
Estimation of coil sensitivity maps: Walsh’s method, Inati’s fast method and ESPIRiT.
Coil compression using singular value decomposition (SVD).
Image reconstruction operations: basic (FFT, NUFFT), parallel imaging (SENSE, GRAPPA, CG-SENSE) and partial Fourier (zero-filling, homodyne detection, projection onto convex sets).
Calculation of radial and spiral trajectories and sampling densities.
Keras metrics for image quality assessment, classification and segmentation.
Helper operations for array manipulation, image processing and linear algebra.
All operations are performed using a TensorFlow/Keras backend. This has several key advantages:
Seamless integration in machine learning applications.
Runs on heterogeneous systems, with most operations supporting CPU and GPU-accelerated paths.
Code is easy to understand, with most of this package written in Python.
Installation
You can install TensorFlow MRI with pip:
$ pip install tensorflow-mriNote that only Linux is currently supported.
TensorFlow Compatibility
Each TensorFlow MRI release is compiled against a specific version of TensorFlow. Please see the compatibility table below to see what versions of each package you can expect to work together.
TensorFlow MRI |
TensorFlow |
|---|---|
v0.4 |
v2.6 |
v0.5 |
v2.6 |
v0.6 |
v2.6 |
Documentation
Visit the docs for the API reference and examples of usage.
Contributions
If you use this package and something does not work as you expected, please file an issue describing your problem. We will do our best to help.
Contributions are very welcome. Please create a pull request if you would like to make a contribution.
Citation
If you find this software useful in your work, please cite us.
FAQ
When trying to install TensorFlow MRI, I get an error about OpenEXR which includes: `OpenEXR.cpp:36:10: fatal error: ImathBox.h: No such file or directory`. What do I do?
OpenEXR is needed by TensorFlow Graphics, which is a dependency of TensorFlow MRI. This issue can be fixed by installing the OpenEXR library. On Debian/Ubuntu:
$ apt install libopenexr-devRelease history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
Built Distributions
Hashes for tensorflow_mri-0.6.0-cp39-cp39-manylinux_2_12_x86_64.manylinux2010_x86_64.whl
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | eda97145e7bb4c115a3d326d93a9a9147cbfd9c8396372cfcf87f3cc499793f4 |
|
| MD5 | 9295601661ea1727ad7ade6e2fea2815 |
|
| BLAKE2b-256 | c7d9a81414a9b223c2ca96bbafb8addaf5849fe79e6536022200f03d018d7df2 |
Hashes for tensorflow_mri-0.6.0-cp38-cp38-manylinux_2_12_x86_64.manylinux2010_x86_64.whl
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | 437b5b60285b978d228c575ecfdda3d06a0c010f04ddf26ebbe0fc4a834b396b |
|
| MD5 | 01a173e1d358adec82764214cc650152 |
|
| BLAKE2b-256 | 8d90f49af58ea68a211700543c628deda41ebc142f5da4d12018fb054d55259a |
Hashes for tensorflow_mri-0.6.0-cp37-cp37m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | 6c5af60582d5069f668a2addbffced1015b4cbbe60dde337f2ee0bd6fe4e2c3e |
|
| MD5 | 840cde8d2e5ac0b0ac4233beeba30f90 |
|
| BLAKE2b-256 | fedb9089350f9771db2f2fc6bb49779930b62fa375cee94c19dcc98a796ac742 |
Hashes for tensorflow_mri-0.6.0-cp36-cp36m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | a1b40634aae85de57c0ce3341ed1d5733d4c6b95bdb509baf472d2c20ce02dd4 |
|
| MD5 | 4bf9458e31a74c2ec7e7e176596cff41 |
|
| BLAKE2b-256 | 4e17c99d80f9a7c288245436d9cae0c7763e51abe8830fbabdd9129f68cfaecd |
