Acu-Net: A 3D Attention Context U-Net for Multiple Sclerosis Lesion Segmentation

Multiple Sclerosis (MS) lesion segmentation from MR images is important for neuroimaging analysis. MS is diffuse, multifocal, and tend to involve peripheral brain structures such as the white matter, corpus callosum, and brainstem. Recently, U-Net has made great achievements in medical image segmentation area. However, the insufficiently use of context information and feature representation, makes it fail to achieve segmentation of MS lesions accurately. To solve the problem, 3D attention context U-Net (ACU-Net) is proposed for MS lesion segmentation in this paper. The proposed ACU-Net includes 3D spatial attention block, which is used to enrich spatial details and feature representation of lesion in the decoding stage. Furthermore, in the encoding and decoding stage of the network, 3D context guided module is designed for guiding local information and surrounding information. The proposed ACU-Net was evaluated on the ISBI 2015 longitudinal MS lesion segmentation challenge dataset, and it achieved superior performance compared to latest approaches.

[1]  Maria Assunta Rocca,et al.  Multi-branch convolutional neural network for multiple sclerosis lesion segmentation , 2018, NeuroImage.

[2]  Saurabh Jain,et al.  Automatic segmentation and volumetry of multiple sclerosis brain lesions from MR images , 2015, NeuroImage: Clinical.

[3]  Sébastien Ourselin,et al.  Bayesian Model Selection for Pathological Neuroimaging Data Applied to White Matter Lesion Segmentation , 2015, IEEE Transactions on Medical Imaging.

[4]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[5]  Guixia Kang,et al.  Robust 3D Convolutional Neural Network With Boundary Correction for Accurate Brain Tissue Segmentation , 2018, IEEE Access.

[6]  Naimul Mefraz Khan,et al.  A Novel Focal Tversky Loss Function With Improved Attention U-Net for Lesion Segmentation , 2018, 2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019).

[7]  Sheng Tang,et al.  Tree-Structured Kronecker Convolutional Network for Semantic Segmentation , 2018, 2019 IEEE International Conference on Multimedia and Expo (ICME).

[8]  Kun Yu,et al.  DenseASPP for Semantic Segmentation in Street Scenes , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[9]  Snehashis Roy,et al.  Longitudinal multiple sclerosis lesion segmentation: Resource and challenge , 2017, NeuroImage.

[10]  Hairong Zheng,et al.  CLCI-Net: Cross-Level fusion and Context Inference Networks for Lesion Segmentation of Chronic Stroke , 2019, MICCAI.

[11]  Hayit Greenspan,et al.  Longitudinal Multiple Sclerosis Lesion Segmentation Using Multi-view Convolutional Neural Networks , 2016, LABELS/DLMIA@MICCAI.

[12]  Joaquim Salvi,et al.  Multiple Sclerosis Lesion Synthesis in MRI Using an Encoder-Decoder U-NET , 2019, IEEE Access.

[13]  F. Barkhof,et al.  Evidence-based guidelines: MAGNIMS consensus guidelines on the use of MRI in multiple sclerosis—clinical implementation in the diagnostic process , 2015, Nature Reviews Neurology.

[14]  Alex Rovira,et al.  Improving automated multiple sclerosis lesion segmentation with a cascaded 3D convolutional neural network approach , 2017, NeuroImage.

[15]  Sheng Tang,et al.  CGNet: A Light-Weight Context Guided Network for Semantic Segmentation , 2018, IEEE Transactions on Image Processing.

[16]  Thomas Brox,et al.  3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation , 2016, MICCAI.

[17]  Loïc Le Folgoc,et al.  Attention U-Net: Learning Where to Look for the Pancreas , 2018, ArXiv.

[18]  Alex Rovira,et al.  Segmentation of multiple sclerosis lesions in brain MRI: A review of automated approaches , 2012, Inf. Sci..