ABSTRACT
The imaging data of Multi-modal brain has been put to a wide use in the identification of epilepsy and tumor lesions because it can provide multi-information about a target (epilepsy lesion, tumor, or tissue). However, most of the existing multi-modal medical image segmentation networks in the existing multi-modal segmentation schemes, on the one hand, adopt the input-level fusion strategy, which directly integrates the multi-modal images in the original input space. But it is hard to preserve the modality-specific properties. On the other hand, the decision-level fusion method with the use of the single network can better exploit the unique information of the corresponding modality, but with the problem that the features learned from a single modality cannot be easily combined with the features of other modality. In this paper, we have proposed Multi-Encoder with Hybrid Lateral Connection Network (MEHLC-Net), a semantic segmentation network based on 2D images with multi encoder structure for multi-modal MRI tumor sub-region segmentation. We use hybrid lateral connections instead of long connections in the U-Net structure to extract features, which can overcome the difficulty of highorder feature fusion caused by multiple encoders. We combine cross-connection with self-attention mechanisms to enhance the accuracy of the network. We evaluate our method on the multimodal data set of BRATS. Compared with the advanced No-New-Net, our model improves the TC of the DSC indicator by 0.13% and the ET indicator by 0.58% respectively.
- Zhao X, Zhang L, Lu H. Automatic polyp segmentation via multi-scale subtraction network[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, Cham, 2021: 120-130.Google Scholar
- Zheng H, Qin Y, Gu Y, Refined Local-imbalance-based Weight for Airway Segmentation in CT[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, Cham, 2021: 410-419.Google Scholar
- Chen H, Dou Q, Yu L, VoxResNet: Deep voxelwise residual networks for brain segmentation from 3D MR images[J]. NeuroImage, 2018, 170: 446-455.Google ScholarCross Ref
- Moeskops P, Viergever M A, Mendrik A M, Automatic segmentation of MR brain images with a convolutional neural network[J]. IEEE transactions on medical imaging, 2016, 35(5): 1252-1261.Google ScholarCross Ref
- Jiang Z, Ding C, Liu M, Two-stage cascaded u-net: 1st place solution to brats challenge 2019 segmentation task[C]//International MICCAI brainlesion workshop. Springer, Cham, 2019: 231-241.Google Scholar
- Zhou T, Ruan S, Canu S. A review: Deep learning for medical image segmentation using multi-modality fusion[J]. Array, 2019, 3: 100004.Google ScholarCross Ref
- Isensee F, Petersen J, Klein A, nnu-net: Self-adapting framework for u-net-based medical image segmentation[J]. arXiv preprint arXiv:1809.10486, 2018.Google Scholar
- Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation[C]//International Conference on Medical image computing and computer-assisted intervention. Springer, Cham, 2015: 234-241.Google Scholar
- Dolz J, Gopinath K, Yuan J, HyperDense-Net: a hyper-densely connected CNN for multi-modal image segmentation[J]. IEEE transactions on medical imaging, 2018, 38(5): 1116-1126.Google Scholar
- Kumar A, Fulham M, Feng D, Co-learning feature fusion maps from PET-CT images of lung cancer[J]. IEEE Transactions on Medical Imaging, 2019, 39(1): 204-217.Google ScholarCross Ref
- Moe Y M, Groendahl A R, Mulstad M, Deep learning for automatic tumour segmentation in PET/CT images of patients with head and neck cancers[J]. arXiv preprint arXiv:1908.00841, 2019.Google Scholar
- Huang G, Liu Z, Van Der Maaten L, Densely connected convolutional networks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 4700-4708.Google Scholar
- Hu J, Shen L, Sun G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 7132-7141.Google Scholar
- Lin T Y, Dollár P, Girshick R, Feature pyramid networks for object detection[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 2117-2125.Google Scholar
- Menze B H, Jakab A, Bauer S, The multimodal brain tumor image segmentation benchmark (BRATS)[J]. IEEE transactions on medical imaging, 2014, 34(10): 1993-2024.Google Scholar
- Bakas S, Akbari H, Sotiras A, Advancing the cancer genome atlas glioma MRI collections with expert segmentation labels and radiomic features[J]. Scientific data, 2017, 4(1): 1-13.Google Scholar
- Bakas S, Reyes M, Jakab A, Identifying the best machine learning algorithms for brain tumor segmentation, progression assessment, and overall survival prediction in the BRATS challenge[J]. arXiv preprint arXiv:1811.02629, 2018.Google Scholar
- S. Bakas, H. Akbari, A. Sotiras , “Segmentation labels for the pre-operative scans of the tcga-gbm collection. the cancer imaging archive,” 2017.Google Scholar
- S. Bakas, H. Akbari, A. Sotiras, M. Bilello, M. Rozycki, J. Kirby, J. Freymann, K. Farahani, and C. Davatzikos, “Segmentation labels for the pre-operative scans of the tcga-gbm collection (2017),” 2017.Google Scholar
- He K, Zhang X, Ren S, Deep residual learning for image recognition[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2016: 770-778.Google Scholar
- Jia H, Xia Y, Cai W, Learning high-resolution and efficient non-local features for brain glioma segmentation in MR images[C]//International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, Cham, 2020: 480-490.Google Scholar
- Myronenko A. 3D MRI brain tumor segmentation using autoencoder regularization[C]//International MICCAI Brainlesion Workshop. Springer, Cham, 2018: 311-320.Google Scholar
Index Terms
- Multi-modal Brain Image Segmentation Based on Multi-Encoder with Hybrid Lateral Connection (MEHLC-Net)
Recommendations
A Novel Brain Tumor Segmentation from Multi-Modality MRI via A Level-Set-Based Model
Segmentation of brain tumor from magnetic resonance imaging is a challenging and time-consuming task due to the unpredictable appearance of tumor tissue in practical applications. In this paper we propose a novel level-set-based model for tumor ...
Review of brain MRI image segmentation methods
Brain image segmentation is one of the most important parts of clinical diagnostic tools. Brain images mostly contain noise, inhomogeneity and sometimes deviation. Therefore, accurate segmentation of brain images is a very difficult task. However, the ...
Segmentation of Brain MRI Image with GVF Snake Model
PCSPA '10: Proceedings of the 2010 First International Conference on Pervasive Computing, Signal Processing and ApplicationsMedical image segmentation is the foundation and research focus in the medical image processing field. In this paper the normalized GVF Snake model combines with traditional edge detection is proposed for the brain MRI image semiautomatic segmentation. ...
Comments