Abstract
Nowadays, numerous deep learning techniques for segmenting medical images have been proposed, with excellent outcomes and the valuable success of machine learning. However, most models use supervised methods while others use unsupervised methods, with less effective outcomes than supervised learning methods. Therefore, this paper introduces a novel level set loss function for unsupervised tasks and incorporates it with the Active Contour loss for supervised tasks. Besides, since previously introduced deep learning models generate less accurate results due to the intensity inhomogeneity issues and the often presence of low-intensity contrast tissues in infant brain segmentation, we propose a new convolutional neural network model to alleviate this problem. Instead of binary segmentation, our proposed loss entitles our model to segment multiple classes with promising outcomes. The proposed technique is utilized to segment neonatal brain magnetic resonance images into four non-overlap regions. The iSeg-2017 challenge, which offers a collection of neonatal brain magnetic resonance imagesĀ from different sites, is used to evaluate our proposed process. The experiment demonstrates that our new loss function achieves promising results among the 21 participating teams. This illustrates the effectiveness of our technique in multiclass medical image segmentation.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Jahne B, Haubsecker H (2000) Computer vision and applications: a guide for students and practitioners. Elsevier
Zhou Z, Siddiquee MMR, Tajbakhsh N, Liang J (2019) Unet++: redesigning skip connections to exploit multiscale features in image segmentation. IEEE Trans Med Imaging 39(6):1856ā1867
Zhou T, Ruan S, Canu S (2019) A review: deep learning for medical image segmentation using multi-modality fusion. Array 3:100004
Bashar A (2019) Survey on evolving deep learning neural network architectures 1(2):73ā82
Manoharan JS (2021) Study of variants of extreme learning machine (ELM) brands and its performance measure on classification algorithm 3(2), 83ā95
Manoharan S (2020) Early diagnosis of lung cancer with probability of malignancy calculation and automatic segmentation of lung CT scan images 2(4), 175ā186 (2020)
Balasubramaniam V (2021) Artificial intelligence algorithm with SVM classification using dermascopic images for melanoma diagnosis 3(1):34ā42 (2021)
Pham V-T, Tran T-T, Wang P-C, Lo M-T (2020) Tympanic membrane segmentation in otoscopic images based on fully convolutional network with active contour loss. Signal Image Video Process, 1ā9
Ronneberger O, Fischer P, Brox T (2015) U-net: Convolutional networks for biomedical image segmentation. In: International conference on medical image computing and computer-assisted intervention. Springer, pp 234ā241
Tran T-T, Tran T-T, Ninh Q-C, Bui M-D, Pham V-T(2020) Segmentation of left ventricle in short-axis MR images based on fully convolutional network and active contour model. In: 5th international conference on green technology and sustainable development, vol 1284, pp 49ā59
Koresh HJD, Chacko S, Periyanayagi MJPRL (2021) A modified capsule network algorithm for oct corneal image segmentation 143:104ā112
Jadon S (2020) A survey of loss functions for semantic segmentation, arXiv preprint arXiv:2006.14822
LeCun Y, Bottou L, Bengio Y, Haffner P (1998) Gradient-based learning applied to document recognition. Proc IEEE 86(11):2278ā2324
Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. Adv Neural Inf Process Syst 25:1097ā1105
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 770ā778
Pham V-T, Tran T-T, Wang P-C, Chen P-Y, Lo M-T (2021) EAR-UNet: a deep learning-based approach for segmentation of tympanic membranes from otoscopic images. Artif Intell Med 115:1ā12
Zhang W et al (2015) Deep convolutional neural networks for multi-modality isointense infant brain image segmentation. Neuroimage 108:214ā224
Moeskops P, Viergever MA, Mendrik AM, De Vries LS, Benders MJ, IÅ”gum I (2016) Automatic segmentation of MR brain images with a convolutional neural network. IEEE Trans Med Imaging 35(5):1252ā1261
Li C, Huang R, Ding Z, Gatenby JC, Metaxas DN, Gore JC (2011) A level set method for image segmentation in the presence of intensity inhomogeneities with application to MRI. IEEE Trans Image Process 20(7):2007ā2016
Wang L et al (2014) Segmentation of neonatal brain MR images using patch-driven level sets. Neuroimage 84:141ā158
Chan TF, Vese LA (2000) Image segmentation using level sets and the piecewise-constant Mumford-Shah model, in Tech. Rep. 0014, Computational Applied Math Group, 2000: Citeseer
Shyu K-K, Pham V-T, Tran T-T, Lee P-L (2012) Global and local fuzzy energy based-active contours for image segmentation. Nonlinear Dyn 67(2), 1559ā1578
Tran T-T, Pham V-T, Shyu K-K (2014) Zernike moment and local distribution fitting fuzzy energy-based active contours for image segmentation. SIViP 8(1):11ā25
Chan TF, Vese LA (2001) Active contours without edges. IEEE Trans Image Process 10(2):266ā277
Kim B, Ye JC (2019) MumfordāShah loss functional for image segmentation with deep learning. IEEE Trans Image Process 29:1856ā1866
Badshah N, Chen K, Ali H, Murtaza G (2012) Coefficient of variation based image selective segmentation model using active contours. East Asian J Appl Math 2(2):150ā169
Trinh M-N, Nguyen N-T, Tran T-T, Pham V-T (2021) A Semi-supervised deep learning-based approach with multiphase active contour loss for left ventricle segmentation from CMR images. In: Proceedings of Third International Conference on Sustainable Computing, 2021: Springer
Wang L et al (2019) Benchmark on automatic six-month-old infant brain segmentation algorithms: the iSeg-2017 challenge. IEEE Trans Med Imaging 38(9):2219ā2230
Abadi M et al (2016) Tensorflow: Large-scale machine learning on heterogeneous distributed systems
Harris CR et al (2020) Array programming with NumPy 585(7825):357ā362
Ahmed S (1995) A pooling methodology for coefficient of variation. SankhyÄ: Indian J Stat Ser B, 57ā75
Mora M, Tauber C, Batatia H (2005) Robust level set for heart cavities detection in ultrasound images. In: Computers in cardiology. IEEE, pp 235ā238
Schulze MA, Wu QX (1995) Nonlinear edge-preserving smoothing of synthetic aperture radar images. In: Proceedings of the New Zealand image and vision computingā95 Workshop, pp 28ā29
Qin X, Zhang Z, Huang C, Dehghan M, Zaiane OR, Jagersand M (2020) U2-Net: going deeper with nested U-structure for salient object detection. Pattern Recogn 106:107404
Ramachandran P, Zoph B,. Le QV (2017) Searching for activation functions. arXiv preprint arXiv:1710.05941
Woo S, Park J, Lee J-Y, Kweon IS (2018) Cbam: convolutional block attention module. In: Proceedings of the European conference on computer vision (ECCV), pp 3ā19
Hu J, Shen L, Sun G (2018) Squeeze-and-excitation networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 7132ā7141
Ghiasi G, Lin T-Y, Le QV (2018) Dropblock: a regularization method for convolutional networks. arXiv preprint arXiv:1810.12890
Taha AA, Hanbury A (2015) Metrics for evaluating 3D medical image segmentation: analysis, selection, and tool. BMC Med Imaging 15(1):1ā28
He K, Zhang X, Ren S, Sun J (2015) Delving deep into rectifiers: Surpassing human-level performance on imagenet classification. In: Proceedings of the IEEE international conference on computer vision, p. 1026ā1034
Dozat T (2016) Incorporating nesterov momentum into adam
Wang L et al (2018) Volume-based analysis of 6-month-old infant brain MRI for autism biomarker identification and early diagnosis. In: International conference on medical image computing and computer-assisted intervention. Springer, , pp 411ā419
Li G et al (2014) Measuring the dynamic longitudinal cortex development in infants by reconstruction of temporally consistent cortical surfaces. Neuroimage 90:266ā279
Acknowledgements
This research is funded by the Hanoi University of Science and Technology (HUST) under project number T2021-PC-005. Minh-Nhat Trinh was funded by Vingroup JSC and supported by the Master, PhD Scholarship Programme of Vingroup Innovation Foundation (VINIF), Institute of Big Data, code VINIF. 2021.ThS.33.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
Ā© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Trinh, MN., Pham, VT., Tran, TT. (2023). A Deep Learning-Based Approach with Semi-supervised Level Set Loss for Infant Brain MRI Segmentation. In: Ranganathan, G., Bestak, R., Fernando, X. (eds) Pervasive Computing and Social Networking. Lecture Notes in Networks and Systems, vol 475. Springer, Singapore. https://doi.org/10.1007/978-981-19-2840-6_41
Download citation
DOI: https://doi.org/10.1007/978-981-19-2840-6_41
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-2839-0
Online ISBN: 978-981-19-2840-6
eBook Packages: EngineeringEngineering (R0)