Stack Overflow
Stack Overflow

Questions tagged [niftynet]

NiftyNet is a TensorFlow-based open-source convolutional neural networks (CNNs) platform for research in medical image analysis and image-guided therapy.

NiftyNet is a TensorFlow-based open-source convolutional neural networks (CNNs) platform for research in medical image analysis and image-guided therapy.

The code is available via GitLab, or you can quickly get started with the PyPI module available here.

Features

NiftyNet currently supports medical image segmentation and generative adversarial networks. NiftyNet is not intended for clinical use. Other features of NiftyNet include:

  • Easy-to-customise interfaces of network components
  • Sharing networks and pre-trained models
  • Support for 2-D, 2.5-D, 3-D, 4-D inputs
  • Efficient discriminative training with multiple-GPU support
  • Implementation of recent networks (HighRes3DNet, 3D U-net, V-net, DeepMedic)
  • Comprehensive evaluation metrics for medical image segmentation

NiftyNet is released under the Apache License, Version 2.0. Please see the LICENSE file in the NiftyNet source code repository for details.

If you use NiftyNet in your work, please cite Gibson, Li et al. 2017. The NiftyNet platform originated in software developed for Li et. al. 2017.

55 questions
1
vote
1 answer

NiftyNet cannot find multiple labels for multiclass segmentation

I'm running a 3-class segmentation problem, with my Label images being "0(many),1(some),2(some)". But I realized when NiftyNet read the images, it only has 0/1 two values. When the label_normalization is on, the ref.txt file says "label from 0,1 to…
Kai Wang
  • 31
  • 5
1
vote
2 answers

Implement transfer learning on niftynet

I want to implement transfer learning using the Dense V-Net architecture. As I was searching on how to do this, I found that this feature is currently being worked on (How do I implement transfer learning in NiftyNet?). Although from that answer it…
Manuel Concepción Brito
  • 63
  • 7
1
vote
0 answers

Issue with Generalized Dice Loss

I was adapting the generalized dice loss provided by niftynet to use on my own tensorflow network which produces a probability for a single label. In order to use the generalized dice loss, I added this line to the beginning of the file (I am…
bob
  • 21
  • 1
  • 2
1
vote
0 answers

Explanation on Inference Time

Can anyone explain to me why the first window in inference takes considerably longer time compared to following windows when input data is larger than window size? Here's ans example of execution times for my application: enter image description…
Somayyeh
  • 11
  • 3
1
vote
1 answer

NiftyNet volume sampling

I'm using NiftyNet for a segmentation problem. To obtain samples from volumes, there is this option Sample_per_volume. Observation: When I monitored the sample location when setting this to "1", I noticed that instead of only one sample is…
Kai Wang
  • 31
  • 5
1
vote
1 answer

NiftyNet Selective Sampler batches not taken from mix of volumes?

I'm training on three CT volumes using the Selective Sampler to ensure that enough samples are taken around the RoI (due to class imbalance), with some random samples. I'm also augmenting the data by scaling, rotation, and flipping, which takes a…
Ginesu_Kun
  • 31
  • 1
1
vote
1 answer

Niftynet Data augmentation settings

I'm using 'NiftyNet' for my project. For data augmentation, in the config file I noticed there are 'rotation_angle', 'scaling_percentage' and 'random_flipping_axes'. I checked the documents but still don't understand what is the proper setting for…
Kai Wang
  • 31
  • 5
1
vote
1 answer

Sample Weighting Regression Niftynet

Good evening, I have an interrogation about the regression application in Niftynet. Indeed I try to predict CT continuous UH maps from MRI(s). I therefore use a regression CNN (highres3dnet) which converges but as an output I naturally get values…
Paulbd
  • 157
  • 1
  • 9
1
vote
1 answer

How can I get access to intermediate activation maps of the pre-trained models in NiftyNet?

I could download and successfully test brain parcellation demo of NiftyNet package. However, this only gives me the ultimate parcellation result of a pre-trained network, whereas I need to get access to the output of the intermediate layers too.…
Jamshid
  • 41
  • 6
1
vote
1 answer

NiftyNet: indices are out-of-bounds error

I'm just starting to use NiftyNet for medical image segmentation. To get going with the software, I was trying to run the demo that segments images from the Brats Challenge dataset (http://www.braintumorsegmentation.org/). I have downloaded the…
piman314
  • 5,285
  • 23
  • 35
0
votes
0 answers

How can I change planes in NifTI images?

So I am trying to mix two different NifTI datasets, but when I view them in 2D they both have different planes, one of them is inverted also. my question is: 1) How to change plane in NifTI? 2) Also how do I rotate the plane ? thankyou in advance.
olive
  • 21
  • 5
0
votes
1 answer

Conv3D size doesn’t make sense with NIFTI data?

So I am writing custom dataset for medical images, with .nii (NIFTI1 format), but there is a confusion. My dataloader returns the shape torch.Size (1,1,256,256,51) . But NIFTI volumes use anatomical axes, different coordinate system, so it doesn’t…
olive
  • 21
  • 5
0
votes
1 answer

Using NiftyNet for oral cavity photos to classify a tumor

I know very well that I am not asking a specific question. But I am asking for guidance about how to go about doing some work with medical imaging using NiftyNet. This will certainly empower me to do some work and seek help with pointed questions…
kishore
  • 541
  • 1
  • 6
  • 18
0
votes
1 answer

Sharing weights in parallel Convolutional Layer

currently I am developing a new network using NiftyNet and would need some help. I am trying to implement an Autofocus Layer [1] as proposed in the paper. However, at a certain point, the Autofocus Layer needs to calculate K (K=4) parallel…
Andreas
  • 1
  • 1
0
votes
1 answer

How to train data in NiftyNet

I'm trying to train a network using NiftyNet with my own data (CT images and their corresponding labels). I designed the Net class shortly following some other training with similar sample data, all NiftyNet documentation I could find and parameters…
BSC
  • 1
  • 1
1
2
3 4