• Corpus ID: 3854002

Parallel Multi-Dimensional LSTM, With Application to Fast Biomedical Volumetric Image Segmentation

  title={Parallel Multi-Dimensional LSTM, With Application to Fast Biomedical Volumetric Image Segmentation},
  author={Marijn F. Stollenga and Wonmin Byeon and Marcus Liwicki and J{\"u}rgen Schmidhuber},
Convolutional Neural Networks (CNNs) can be shifted across 2D images or 3D videos to segment them. [] Key Method Here we re-arrange the traditional cuboid order of computations in MD-LSTM in pyramidal fashion. The resulting PyraMiD-LSTM is easy to parallelise, especially for 3D data such as stacks of brain slice images. PyraMiD-LSTM achieved best known pixel-wise brain image segmentation results on MRBrainS13 (and competitive results on EM-ISBI12).

Figures and Tables from this paper

Fully convolutional structured LSTM networks for joint 4D medical image segmentation

This is the first DL framework with deep integration of FCNs and LSTM for joint 4D segmentation that could be trained end-to-end and achieves promising results with the demonstrated application to longitudinal pediatric magnetic resonance imaging (MRI) segmentation.

Combining Fully Convolutional and Recurrent Neural Networks for 3D Biomedical Image Segmentation

This paper proposes a new DL framework for 3D image segmentation, based on a fully convolutional network (FCN) and a recurrent neural network (RNN), which are responsible for exploiting the intra-slice and inter-slice contexts, respectively.

Image Analysis with Long Short-Term Memory Recurrent Neural Networks

This thesis is to investigate a purely learning-based approach, particularly, Multi-Dimensional LSTM (MD-LSTM) recurrent neural networks to tackle the challenging CV tasks, classification and segmentation on 2D and 3D image data.

Video object segmentation by Multi-Scale Pyramidal Multi-Dimensional LSTM with generated depth context

  • Qiurui WangC. Yuan
  • Computer Science
    2016 IEEE International Conference on Image Processing (ICIP)
  • 2016
A more suitable architecture Multi-Scale Pyramidal Multi-Dimensional Long Short Term Memory (MSPMD-LSTM) is used to reveal the strong relevance within video frames and depth context is extracted and refined to enhance the performance of the model.

RACE-Net: A Recurrent Neural Network for Biomedical Image Segmentation

The RACE-net allows the curve evolution velocities to be learned in an end-to-end manner while minimizing the number of network parameters, computation time, and memory requirements.

VoxResNet: Deep Voxelwise Residual Networks for Volumetric Brain Segmentation

This work unravels the potential of 3D deep learning to advance the recognition performance on volumetric image segmentation and proposes a deep voxelwise residual network, referred as VoxResNet, which borrows the spirit of deep residual learning in 2D image recognition tasks, and is extended into a 3D variant for handlingvolumetric data.

Quantitative Analysis of Patch-Based Fully Convolutional Neural Networks for Tissue Segmentation on Brain Magnetic Resonance Imaging

This study implements eight FCNN architectures inspired by robust state-of-the-art methods on brain segmentation related tasks and uses them within a standard pipeline, and shows that the methods under evaluation can yield top performance on the three data collections.

Improving Deep Pancreas Segmentation in CT and MRI Images via Recurrent Neural Contextual Learning and Direct Loss Function

This work proposes a new convolutional/recurrent neural network architecture to address the contextual learning and segmentation consistency problem and outperforms the state-of-the-art work on CT and MRI pancreas segmentation, respectively.

PreNet: Parallel Recurrent Neural Networks for Image Classification

This paper proposes a hierarchical parallel recurrent neural network (PreNet) to model spatial context for image classification and can achieve the state-of-the-art classification performance, which demonstrates the advantage of PreNet over many comparative CNN structures.

ReSeg: A Recurrent Neural Network for Object Segmentation

The proposed network, called ReSeg, is a structured prediction architecture for images centered around deep recurrent neural networks based on the recently introduced ReNet model for object classification that can challenge the state of the art in object segmentation and may have further applications in structured prediction at large.



Scene labeling with LSTM recurrent neural networks

The approach, which has a much lower computational complexity than prior methods, achieved state-of-the-art performance over the Stanford Background and the SIFT Flow datasets and the ability to visualize feature maps from each layer supports the hypothesis that LSTM networks are overall suited for image processing tasks.

ImageNet classification with deep convolutional neural networks

A large, deep convolutional neural network was trained to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes and employed a recently developed regularization method called "dropout" that proved to be very effective.

Multi-dimensional Recurrent Neural Networks

Multi-dimensional recurrent neural networks are introduced, thereby extending the potential applicability of RNNs to vision, video processing, medical imaging and many other areas, while avoiding the scaling problems that have plagued other multi-dimensional models.

LINKS: Learning-Based Multi-source IntegratioN FrameworK for Segmentation of Infant Brain Images

The proposed random forest technique is employed to effectively integrate features from multi-source images together for tissue segmentation of infant brain images and achieves better performance than other state-of-the-art automated segmentation methods.

Deep Neural Networks Segment Neuronal Membranes in Electron Microscopy Images

This work addresses a central problem of neuroanatomy, namely, the automatic segmentation of neuronal structures depicted in stacks of electron microscopy images, using a special type of deep artificial neural network as a pixel classifier to segment biological neuron membranes.

MRBrainS Challenge: Online Evaluation Framework for Brain Image Segmentation in 3T MRI Scans

The MRBrainS evaluation framework provides an objective and direct comparison of all evaluated algorithms and can aid in selecting the best performing method for the segmentation goal at hand.

Visualizing and Understanding Convolutional Networks

A novel visualization technique is introduced that gives insight into the function of intermediate feature layers and the operation of the classifier in large Convolutional Network models, used in a diagnostic role to find model architectures that outperform Krizhevsky et al on the ImageNet classification benchmark.

Sequence to Sequence Learning with Neural Networks

This paper presents a general end-to-end approach to sequence learning that makes minimal assumptions on the sequence structure, and finds that reversing the order of the words in all source sentences improved the LSTM's performance markedly, because doing so introduced many short term dependencies between the source and the target sentence which made the optimization problem easier.

A modular hierarchical approach to 3D electron microscopy image segmentation

Dropout Improves Recurrent Neural Networks for Handwriting Recognition

It is shown that RNNs with Long Short-Term memory cells can be greatly improved using dropout - a recently proposed regularization method for deep architectures, even when the network mainly consists of recurrent and shared connections.