Parallel Multi-Dimensional LSTM, With Application to Fast Biomedical Volumetric Image Segmentation
Authors:
- Marijn F. StollengaMSMarijn F. Stollenga
- Wonmin ByeonWBWonmin Byeon
- Juergen SchmidhuberJSJuergen Schmidhuber
Preprint
Abstract
Convolutional Neural Networks (CNNs) can be shifted across 2D images or 3D videos to segment them. They have a fixed input size and typically perceive only small local contexts of the pixels to be classified as foreground or background. In contrast, Multi-Dimensional Recurrent NNs (MD-RNNs) can perceive the entire spatio-temporal context of each pixel in a few sweeps through all pixels, especially when the RNN is a Long Short-Term Memory (LSTM). Despite these theoretical advantages, however, unlike CNNs, previous MD-LSTM variants were hard to parallelize on GPUs. Here we re-arrange the traditional cuboid order of computations in MD-LSTM in pyramidal fashion. The resulting PyraMiD-LSTM is easy to parallelize, 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).Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- LINKS: Learning-based multi-source IntegratioN frameworK for Segmentation of infant brain imagesNeuroImage, 2015
- Long short-term memory recurrent neural network architectures for large scale acoustic modeling2014
- Sequence discriminative distributed training of long short-term memory recurrent neural networks2014
- A modular hierarchical approach to 3D electron microscopy image segmentationJournal of Neuroscience Methods, 2014
- An evaluation of four automatic methods of segmenting the subcortical structures in the brainNeuroImage, 2009
- Parallel algorithms for approximation of distance maps on parametric surfacesACM Transactions on Graphics, 2008
- A Novel Connectionist System for Unconstrained Handwriting RecognitionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2008
- Snakes: Active contour modelsInternational Journal of Computer Vision, 1988
- Adaptive histogram equalization and its variationsComputer Vision, Graphics, and Image Processing, 1987
- Measures of the Amount of Ecologic Association Between SpeciesEcology, 1945