Corpus ID: 237572088

Small Lesion Segmentation in Brain MRIs with Subpixel Embedding

  title={Small Lesion Segmentation in Brain MRIs with Subpixel Embedding},
  author={Alex Wong and Allison Chen and Yangchao Wu and Safa Cicek and Alexandre Tiard and Byung-Woo Hong and Stefano Soatto},
  • Alex Wong, Allison Chen, +4 authors Stefano Soatto
  • Published 18 September 2021
  • Computer Science, Engineering
  • ArXiv
We present a method to segment MRI scans of the human brain into ischemic stroke lesion and normal tissues. We propose a neural network architecture in the form of a standard encoder-decoder where predictions are guided by a spatial expansion embedding network. Our embedding network learns features that can resolve detailed structures in the brain without the need for high-resolution training images, which are often unavailable and expensive to acquire. Alternatively, the encoderdecoder learns… Expand

Figures and Tables from this paper


X-Net: Brain Stroke Lesion Segmentation Based on Depthwise Separable Convolution and Long-range Dependencies
A depthwise separable convolution based X-Net that designs a nonlocal operation namely Feature Similarity Module (FSM) to capture long-range dependencies and provides a more effective, dense contextual information extraction and thus facilitates better segmentation. Expand
U-Net: Convolutional Networks for Biomedical Image Segmentation
It is shown that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks. Expand
Multiscale brain MRI super-resolution using deep 3D convolutional networks
This work delves into the relevance of several factors in the performance of the purely convolutional neural network-based techniques for the monomodal super-resolution, and highlights that one single network can efficiently handle multiple arbitrary scaling factors based on a multiscale training approach. Expand
ISLES Challenge: U-Shaped Convolution Neural Network with Dilated Convolution for 3D Stroke Lesion Segmentation
The algorithm for stroke lesion segmentation based on a deep convolutional neural network (CNN) based on U-shaped CNN, which has been applied successfully to other medical image segmentation tasks is proposed. Expand
CLCI-Net: Cross-Level fusion and Context Inference Networks for Lesion Segmentation of Chronic Stroke
A Cross-Level fusion and Context Inference Network (CLCI-Net) for the chronic stroke lesion segmentation from T1-weighted MR images is proposed with the results showing that the network outperforms five state-of-the-art methods. Expand
Multi-level Context Gating of Embedded Collective Knowledge for Medical Image Segmentation
An extension of U-Net for medical image segmentation is proposed, in which it takes full advantages ofU-Net, Squeeze and Excitation block, bi-directional ConvLSTM, and the mechanism of dense convolutions to strengthen feature propagation and encourage feature reuse. Expand
D-UNet: A Dimension-Fusion U Shape Network for Chronic Stroke Lesion Segmentation
A new architecture called dimension-fusion-UNet (D- UNet), which combines 2D and 3D convolution innovatively in the encoding stage is proposed, which achieves a better segmentation performance than 2D networks, while requiring significantly less computation time in comparison to 3D networks. Expand
Single-image super-resolution of brain MR images using overcomplete dictionaries
A sparse-based super-resolution method, adapted for easily including prior knowledge, which couples up high and low frequency information so that a high-resolution version of a low-resolution brain MR image is generated, shown to outperform a recent state-of-the-art algorithm. Expand
Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network
This paper presents the first convolutional neural network capable of real-time SR of 1080p videos on a single K2 GPU and introduces an efficient sub-pixel convolution layer which learns an array of upscaling filters to upscale the final LR feature maps into the HR output. Expand
Brain MRI super-resolution using deep 3D convolutional networks
A three-dimensional convolutional neural network is proposed to generate high-resolution brain image from its input low-resolution (LR) with the help of patches of other HR brain images to demonstrate the need of fitting data and network parameters for 3D brain MRI. Expand