GithubHelp home page GithubHelp logo

js3611 / deep-mri-reconstruction Goto Github PK

View Code? Open in Web Editor NEW
315.0 315.0 103.0 56.29 MB

Deep Cascade of Convolutional Neural Networks for MR Image Reconstruction: Implementation & Demo

License: Other

Python 100.00%
compressed-sensing convolutional-neural-networks deep-learning deep-neural-networks image-reconstruction mri

deep-mri-reconstruction's People

Contributors

cq615 avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

deep-mri-reconstruction's Issues

About test

Hi,
I want to test with single mri image,which script should I use?

Expression of the data consistency layer

I have a question regarding the expression of the output of the data consistency layer in the noisy case.

We have out = (x + v * x_sampled) / (1 + v) both here and here.

But this is done without consideration of whether the sampled point is in sampled region or not. I.e., the formal expression represented by this code would be, instead of eq. 6 of your paper:

x_rec(k) = x_cnn(k) / (1 + v) if k \notin Omega; (x_cnn(k) + v * x_u(k)) / (1 + v) else

I removed the \hat to make it more readable and replaced lambda by v.

I think the correct implementation of eq. 6 should be:

out = x + (v * (x_sampled - mask * x)) / (1 + v)

What do you think?

More sample data to train model from scratch

Hi @js3611 ,

In the paper, you mentioned that "our data is consisting of 10 fully sampled short-axis cardiac cine MR scans. Each scan contains a single slice SSFP acquisition with 30 temporal frames with a 320 ร— 320
mm field of view and 10 mm slice thickness".

But in the data repo, there are only 256x256 x 30 frames.
I assume that this is one of 10 scans. Is it possible to release the rest of 9 scans so that I can train your model from scratch?

Best,

Reference paper for StochasticDnCn

I noticed in the pytorch models this StochasticDnCn model where you stochastically drop blocks of conv + data consistency (I guess during training).

I wanted to know if there was a paper where the rationale behind this model is detailed (with potentially results).

lasagne.layers 'prelu'

can't execute ''from lasagne.layers import prelu''
in your requirement txt, you said lasagne==v0.2, but I can just find v0.1

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.