A comparative analysis on segmentation of brain MRI using CNN models
1
조회
0
다운로드
4
인용
0
참고문헌
키워드
초록
The segmentation of brain MR images is performed to eliminate unwanted details and to locate relevant objects from the processed images. Many neuro-disorders such as Alzheimer's, multiple sclerosis, epilepsy, etc. can be diagnosed with the aid of brain tissue segmentation. In particular, automated segmentation of deep brain structures such as gray matter (GM), white matter (WM), cerebrospinal fluid (CSF) from magnetic resonance images play a vital role in increasing the rate for the diagnosis of brain-neural disorders. Most of the research on semantic segmentation is focused on improving the accuracy with computationally efficient solutions. In recent years, convolutional networks have shown breakthrough performance in image segmentation and classification. Convolution Neural Networks (CNN), with their ability to self-learn features from the given data, are being applied extensively for medical images analysis. In this paper, we present a comparative study on four popularly used CNN models such as Segnet, U-net, U-Segnet, and M-net for MRI brain tissue segmentation and summarize their observed experimental analysis. The performances are measured based on the Dice coefficient and Jaccard index. Experimental results show that M-net achieves better segmentation accuracy with Dice co-efficient 93% and JI 86%.
Abstract
The segmentation of brain MR images is performed to eliminate unwanted details and to locate relevant objects from the processed images. Many neuro-disorders such as Alzheimer's, multiple sclerosis, epilepsy, etc. can be diagnosed with the aid of brain tissue segmentation. In particular, automated segmentation of deep brain structures such as gray matter (GM), white matter (WM), cerebrospinal fluid (CSF) from magnetic resonance images play a vital role in increasing the rate for the diagnosis of brain-neural disorders. Most of the research on semantic segmentation is focused on improving the accuracy with computationally efficient solutions. In recent years, convolutional networks have shown breakthrough performance in image segmentation and classification. Convolution Neural Networks (CNN), with their ability to self-learn features from the given data, are being applied extensively for medical images analysis. In this paper, we present a comparative study on four popularly used CNN models such as Segnet, U-net, U-Segnet, and M-net for MRI brain tissue segmentation and summarize their observed experimental analysis. The performances are measured based on the Dice coefficient and Jaccard index. Experimental results show that M-net achieves better segmentation accuracy with Dice co-efficient 93% and JI 86%.
참고문헌 (30)
- Y Chudasama, "Functions of frontostriatal systems in cognition : comparative neuropsychopharmacological studies in rats, monkeys and humans", Biological psychology, 73(1), pp.19~38 (2006)
- Brian Patenaude, "A bayesian model of shape and appearance for subcortical brain segmentation", Neuroimage, 56(3), pp.907~922 (2011)
- Liana Apostolova, "Subregional hippocampal atrophy predicts alzheimer's dementia in the cognitively normal", Neurobiology of aging, 31(7), pp.1077~1088 (2010)
- Rolf Heckemann, "Automatic anatomical brain MRI segmentation combining label propagation and decision fusion", NeuroImage, 33(1), pp.115~126 (2006)
- Braak H, "Stageing of Alzheimer's disease-related neurofibrillary changes", Neurobiology of Aging, 16, pp.271~278 (1995)
- Dubois B, "Research criteria for the diagnosis of Alzheimer's disease : revising the NINCDS-ADRDA criteria", Lancet Neurology, 6, pp.734~746 (2007)
- Juan Eugenio Iglesias, "Multi-atlas segmentation of biomedical images : a survey", Medical image analysis, 24(1), pp.205~219 (2015)
- Yann LeCun, "handbook of brain theory and neural networks", pp.255~258 (1998)
- Wenlu Zhang, "Deep convolutional neural networks for multi-modality isointense infant brain image segmentation", NeuroImage, 108(01) (2015)
- Siqi Bao, "Multi-scale structured cnn with label consistency for brain mr image segmentation", Computer Methods in Biomechanics and Biomedical Engineering: Imaging and Visualization, 6, pp.1~5 (2016)
- Mahsa Shakeri, "Sub-cortical brain structure segmentation using f-cnn's", arXiv preprint arXiv:1602.02130, ~ (2016)
- Zeynettin Akkus, "Deep learning for brain MRI segmentation: State of the art and future directions", Journal of Digital Imaging, 30(4), pp.449~459 (2017)
- M. Ranzato, "Unsupervised learning of invariant feature hierarchies with applications to object recognition", Proc. IEEE Conf. Comput. Vis. Pattern Recognit, pp.1~8 (2007)
- H. Noh, "Learning deconvolution network for semantic segmentation", ICCV, pp.1520~1528 (2015)
- S. Zheng, "Conditional random fields as recurrent neural networks", Proceedings of the IEEE International Conference on Computer Vision, pp.1529~1537 (2015)
- D. Eigen, "Predicting depth, surface normals and semantic labels with a common multi-scale convolutional architecture", ICCV, pp.2650~2658 (2015)
- S. Hong, "Decoupled deep neural network for semisupervised semantic segmentation", NIPS, pp.1495~1503 (2015)
- V. Badrinarayanan, "Segnet : A deep convolutional encoder-decoder architecture for image segmentation", arXiv preprint arXiv:1511.00561, ~ (2015)
- Olaf Ronneberger, "U-net: Convolutional networks for biomedical image segmentation", International Conference on Medical Image Computing and Computer-Assisted Intervention, pp.234~241 (2015), Springer
- Kumar, Pulkit, "U-SegNet: Fully Convolutional Neural Network based Automated Brain tissue segmentation Tool", ~ (2018)
- Adiga V S., "Inpainting and Denoising Challenges" (2019), Springer
- F. Abien, "Deep Learning using Rectified Linear Unites (ReLU)", Neural and Evolutionary Computing, 1 (2018)
- Chenetal., "VoxResNet: Deepvoxel-wiseresidualnetworks for volumetric brain segmentation", CoRR, vol. abs, 2016
- D. S. Marcus, "Open Access Series of Imaging Studies (OASIS): cross-sectional MRI data in young, middle aged, nondemented, and demented older adults", J. Cogn. Neurosci (2007)
- L.R. Dice, "Measures of the amount of ecologic association between species", Ecology, 26 (1945)
- William Crum, "Generalized overlap measures for evaluation and validation in medical image analysis", IEEE transactions on medical imaging, 25, pp.1451~1461 (2006)
- Chen-Yu Lee, "Deeply-supervised nets", AISTATS, 2, pp.6~ (2015)
- Rupesh Kumar Srivastava, "Highway networks", arXiv preprint arXiv:1505.00387, ~ (2015)
- I. Nabil, "MultiResUNet : Rethinking the U-Net Architecture for Multimodal Biomedical Image Segmentation", Neural Networks, 121, pp.74~87 (2020)
- Jonathan Long, "Fully Convolutional Networks for Semantic Segmentation", CVPR 2015 (2015)