Abstract: To overcome the computational burden of processing three-dimensional (3D) medical scans and the lack of spatial information in two-dimensional (2D) medical scans, a novel segmentation method was proposed that integrates the segmentation results of three densely connected 2D convolutional neural networks (2D-CNNs). In order to combine the low-level features and high-level features, we added densely connected blocks in the network structure design so that the low-level features will not be missed as the network layer increases during the learning process. Further, in order to resolve the problems of the blurred boundary of the glioma edema area, we superimposed and fused the T2-weighted fluid-attenuated inversion recovery (FLAIR) modal image and the T2-weighted (T2) modal image to enhance the edema section. For the loss function of network training, we improved the cross-entropy loss function to effectively avoid network over-fitting. On the Multimodal Brain Tumor Image Segmentation Challenge (BraTS) datasets, our method achieves dice similarity coefficient values of 0.84, 0.82, and 0.83 on the BraTS2018 training; 0.82, 0.85, and 0.83 on the BraTS2018 validation; and 0.81, 0.78, and 0.83 on the BraTS2013 testing in terms of whole tumors, tumor cores, and enhancing cores, respectively. Experimental results showed that the proposed method achieved promising accuracy and fast processing, demonstrating good potential for clinical medicine.

集成多个密集连接二维卷积神经网络（2D-CNNs）分割模型的脑胶质瘤三维分割

[1]BaidU, TalbarS, RaneS, et al., 2020. A novel approach for fully automatic intra-tumor segmentation with 3D U-Net architecture for gliomas. Front Comput Neurosci, 14:10.

[2]BakasS, AkbariH, SotirasA, et al., 2017. Advancing The Cancer Genome Atlas glioma MRI collections with expert segmentation labels and radiomic features. Sci Data, 4:170117.

[3]ChenLL, WuY, DSouzaAM, et al., 2018. MRI tumor segmentation with densely connected 3D CNN. Proceedings Volume 10574, Medical Imaging 2018: Image Processing. SPIE Medical Imaging, 2018, Houston, Texas, USA, 105741F.

[4]DvořákP, MenzeB, 2016. Local structure prediction with convolutional neural networks for multimodal brain tumor segmentation. In: Menze B, Langs G, Montillo A, et al. (Eds.), Medical Computer Vision: Algorithms for Big Data. Springer, Cham, p.59-71.

[5]GoetzM, WeberC, BinczykF, et al., 2015. DALSA: domain adaptation for supervised learning from sparsely annotated MR images. IEEE Trans Med Imaging, 35(1):184-196.

[6]GuoD, WangL, SongT, et al., 2019. Cascaded global context convolutional neural network for brain tumor segmentation. In: Crimi A, Bakas S (Eds.), Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. BrainLes 2019. Lecture Notes in Computer Science, Vol. 11992. Springer, Cham, p.315-326.

[7]HavaeiM, DavyA, Warde-FarleyD, et al., 2017. Brain tumor segmentation with Deep Neural Networks. Med Image Anal, 35:18-31.

[8]HeKM, ZhangXY, RenSQ, et al., 2016. Deep residual learning for image recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, p.770-778.

[9]HuangG, LiuZ, van der MaatenL, et al., 2017. Densely connected convolutional networks. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, p.2261-2269.

[10]HussainS, AnwarSM, MajidM, 2017. Brain tumor segmentation using cascaded deep convolutional neural network. 2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Jeju, Republic of Korea, p.1998-2001.

[11]IslamR, ImranS, AshikuzzamanM, et al., 2020. Detection and classification of brain tumor based on multilevel segmentation with convolutional neural network. J Biomed Sci Eng, 13(4):45-53.

[12]KamnitsasK, LedigC, NewcombeVFJ, et al., 2017. Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation. Med Image Anal, 36:61-78.

[13]KistlerM, BonarettiS, PfahrerM, et al., 2013. The virtual skeleton database: an open access repository for biomedical research and collaboration. J Med Internet Res, 15(11):e245.

[14]KrishnaN, KhalanderMR, ShettyN, et al., 2019. Segmentation and detection of glioma using deep learning. In: Chiplunkar N, Fukao T (Eds.), Advances in Artificial Intelligence and Data Engineering. Advances in Intelligent Systems and Computing, Vol. 1133. Springer, Singapore, p.109-120.

[15]LiQN, GaoZF, WangQY, et al., 2018. Glioma segmentation with a unified algorithm in multimodal MRI images. IEEE Access, 6:9543-9553.

[16]LiYH, JiaFC, QinJ, 2016. Brain tumor segmentation from multimodal magnetic resonance images via sparse representation. Artif Intell Med, 73:1-13.

[17]McKinleyR, MeierR, WiestR, 2018. Ensembles of densely-connected CNNs with label-uncertainty for brain tumor segmentation. In: Crimi A, Bakas S, Kuijf H, et al. (Eds.), Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. BrainLes 2018. Lecture Notes in Computer Science, Vol. 11384. Springer, Cham, p.456-465.

[18]MengqiaoW, JieY, YiC, et al., 2017. The multimodal brain tumor image segmentation based on convolutional neural networks. 2017 2nd IEEE International Conference on Computational Intelligence and Applications (ICCIA). Springer, Cham, p.336-339.

[19]MenzeBH, JakabA, BauerS, et al., 2015. The multimodal brain tumor image segmentation benchmark (BRATS). IEEE Trans Med Imaging, 34(10):1993-2024.

[20]MohanG, SubashiniMM, 2018. MRI based medical image analysis: survey on brain tumor grade classification. Biomed Signal Proc Control, 39:139-161.

[21]NairV, HintonGE, 2010. Rectified linear units improve restricted Boltzmann machines. Proceedings of the 27th International Conference on Machine Learning (ICML-10), Haifa, Israel, p.807-814.

[22]PanMY, ShiYH, SongZJ, 2020. Segmentation of gliomas based on a double-pathway residual convolution neural network using multi-modality information. J Med Imaging Health Informatics, 10(11):2784-2794.

[23]SimonyanK, ZissermanA, 2015. Very deep convolutional networks for large-scale image recognition. Proceedings of the 3rd International Conference on Learning Representations. Springer, Cham, p.6.https://arxiv.org/abs/1409.1556

[24]UdupaP, VishwakarmaS, 2016. A survey of MRI segmentation techniques for brain tumor studies. Bonfring Int J Adv Image Proc, 6(3):22-27.

[25]UrbanG, BendszusM, HamprechtFA, et al., 2014. Multi-modal brain tumor segmentation using deep convolutional neural networks. Proceedings MICCAI-BRATS, Springer, Cham, p.31-35.

[26]ZhaoLY, JiaKB, 2015. Deep feature learning with discrimination mechanism for brain tumor segmentation and diagnosis. 2015 International Conference on Intelligent Information Hiding and Multimedia Signal Processing (IIH-MSP). Adelaide, SA, Australia, p.306-309.

[27]ZhaoXM, WuYH, SongGD, et al., 2018. A deep learning model integrating FCNNs and CRFs for brain tumor segmentation. Med Image Anal, 43:98-111.

[28]ZhouCH, ChenSC, DingCX, et al., 2018. Learning contextual and attentive information for brain tumor segmentation. In: Crimi A, Bakas S, Kuijf H, et al. (Eds.), Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. BrainLes 2018. Lecture Notes in Computer Science, Vol. 11384, Springer, Cham, p.497-507.

[29]ZhouZX, HeZS, ShiMF, et al., 2020. 3D dense connectivity network with atrous convolutional feature pyramid for brain tumor segmentation in magnetic resonance imaging of human heads. Comput Biol Med, 121:103766.

[30]ZhugeY, KrauzeAV, NingH, et al., 2017. Brain tumor segmentation using holistically nested neural networks in MRI images. Med Phys, 44(10):5234-5243.

[31]ZikicD, IoannouY, BrownM, et al., 2014. Segmentation of brain tumor tissues with convolutional neural networks. Proceedings MICCAI Workshop on Multimodal Brain Tumor Segmentation Challenge. Springer, Cham, p.36-39.