UDC 004.8, 004.048
1 V.M. Glushkov Institute of Cybernetics, National Academy of Sciences of Ukraine, Kyiv, Ukraine
svm@nau.edu.ua
|
2 National Technical University “Igor Sikorsky Kyiv Polytechnic Institute,” Kyiv, Ukraine
kir_ryaz@tk.kpi.ua
|
A THREE-STAGE 2D-3D CONVOLUTIONAL NETWORK ENSEMBLE
FOR SEGMENTATION OF MALIGNANT BRAIN TUMORS ON MRI IMAGES
Abstract. In this paper, the problem of brain tumor binary semantic segmentation from MRI images is solved. The pixel-by-pixel determination of the anomaly region boundary is performed given the presence of noise in the training sample and input data. It is shown that in the case of using 2D models for solving 3D segmentation problems, spatial information between neighboring slices is not considered and not utilized. A new approach for optimizing the processing of 3D medical images using ensemble topologies in three stages is proposed. The first stage involves 2D ensemble processing of images in three dimensions to maximize the diversity criterion and accurately capture the region of interest (ROI). The second stage involves ensemble processing of 3D ROI regions extracted by neural networks with different 3D input block sizes to ensure diversity. In the third stage, the extracted abnormal regions (malignant tumors) from the first and second stages are aggregated by weighted summation and thresholding to obtain the final binary 3D mask of the brain tumor. The proposed approach was tested on the LGG Brain MRI Segmentation Dataset. It is shown that the segmentation accuracy is significantly improved in terms of dice score and mIoU, reducing the use of computationally expensive 3D networks.
Keywords: convolutional neural network, ensemble topology, brain tumor, MRI, 3D neural network.
full text
REFERENCES
- Malhotra P., Gupta S., Koundal D., Zaguia A., Enbeyle W. Deep neural networks for medical image segmentation. Journal of Healthcare Engineering. 2022. Vol. 2022. Article ID 9580991. 15 p. https://doi.org/10.1155/2022/9580991.
- Wang R., Lei T., Cui R., Zhang B., Meng H., Nandi A. Medical image segmentation using deep learning: A survey. IET Image Processing. 2022. Vol. 16, Iss. 5. P. 1243–1267. https://doi.org/10.1049/ipr2.12419.
- Rizwan-i-Haque I., Neubert J. Deep learning approaches to biomedical image segmentation. Informatics in Medicine Unlocked. 2020. Vol. 18. Article 100297. https://doi.org/10.1016/j.imu.2020.100297.
- Muller D., Kramer F. MIScnn: a framework for medical image segmentation with convolutional neural networks and deep learning. BMC Medical Imaging. 2019. Vol. 21. Article 12. https://doi.org/10.1186/s12880-020-00543-7.
- Xin M., Wang Y. Research on image classification model based on deep convolution neural network. EURASIP Journal on Image and Video Processing. 2019. Vol. 2019. Article 40. https://doi.org/10.1186/s13640-019-0417-8.
- Krishna M., Neelima M., Mane H., Matcha V. Image classification using deep learning. International Journal of Engineering & Technology. 2018. Vol. 7, Iss. 7. P. 614–617. https://doi.org/10.14419/ijet.v7i2.7.10892.
- Minaee S., Boykov Yu., Porikli F., Plaza A., Kehtarnavaz A., Terzopoulos D. Image segmentation using deep learning: a survey. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2022. Vol. 44, Iss.7. P. 3523–3542. https://doi.org/10.1109/TPAMI.2021.3059968.
- Islam M.T., Karim Siddique B.M.N., Rahman S., Jabid T. Image recognition with deep learning. Proc. 2018 International Conference on Intelligent Informatics and Biomedical Sciences (ICIIBMS) (21–24 October 2018, Bangkok, Thailand). Bangkok, 2018. P. 106–110. https://doi.org/10.1109/ICIIBMS.2018.8550021.
- Chai J., Zeng H., Li A., Ngai E. Deep learning in computer vision: A critical review of emerging techniques and application scenarios. Machine Learning with Applications. 2021. Vol. 6. Article 100134. https://doi.org/10.1016/j.mlwa.2021.100134.
- Voulodimos A., Doulamis N., Doulamis A., Protopapadakis E., Andina D. Deep learning for computer vision: A brief review. Computational Intelligence and Neuroscience. 2018. Vol. 2018. Article ID 7068349. https://doi.org/10.1155/2018/7068349.
- Long J., Shelhamer E., Darrell T. Fully convolutional networks for semantic segmentation. Proc. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (7–12 June 2015, Boston, MA, USA). Boston, 2015. P. 3431–3440. https://doi.org/10.1109/CVPR.2015.7298965.
- Ronneberger O., Fischer P., Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation. Proc. Medical Image Computing and Computer-Assisted Intervention — MICCAI 2015. MICCAI 2015 (5–9 October 2015, Munich, Germany). Munich, 2015. Navab N., Hornegger J., Wells W., Frangi A. (Eds.). Lecture Notes in Computer Science. 2015. Vol. 9351. P. 234–241. https://doi.org/10.1007/978-3-319-24574-4_28.
- Milletari F., Navab N., Ahmadi S. V-Net: fully convolutional neural networks for volumetric medical image segmentation. Proc. 2016 Fourth International Conference on 3D Vision (3DV) (25–28 October 2016, Stanford, CA, USA). Stanford, 2016. P. 565–571.
- Seol Y.J., Kim Y.J., Kim Y.S., Cheon Y.W., Kim K.G. A study on 3D deep learning-based automatic diagnosis of nasal fractures. Sensors. 2022. Vol. 22, Iss. 2. Article 506. https://doi.org/10.3390/s22020506.
- Charles R.Q., Su H., Kaichun M., Guibas L.J. PointNet: Deep learning on point sets for 3D classification and segmentation. Proc. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (21–26 July 2017, Honolulu, HI, USA), 2017. P. 77–85. https://doi.org/10.1109/CVPR.2017.16.
- Feng X., Tustison N.J., Patel S.H., Meyer C.H. Brain tumor segmentation using an ensemble of 3D U-Nets and overall survival prediction using radiomic features. Frontiers in Computational Neuro Science. 2020. Vol. 14. Article number 25. https://doi.org/10.3389/fncom.2020.00025.
- Das S., Bose S., Nayak G.K., Saxena S. Deep learning-based ensemble model for brain tumor segmentation using multi-parametric MR scans. Open Computer Science. 2022. Vol. 12, N 1. P. 211–226. https://doi.org/10.1515/comp-2022-0242.
- Zheng H., Zhang Y., Yang L., Liang P., Zhao Z., Wang C., Chen D.Z. A new ensemble learning framework for 3D biomedical image segmentation. Proc. Thirty-Third AAAI Conference on Artificial Intelligence and Thirty-First Innovative Applications of Artificial Intelligence Conference and Ninth AAAI Symposium on Educational Advances in Artificial Intelligence (AAAI’19/IAAI’19/EAAI’19) (27 January – 1 February 2019, Honolulu, Hawaii USA). Honolulu, 2019. Article 725. P. 5909–5916. https://doi.org/10.1609/aaai.v33i01.33015909.
- Cao H., Liu H., Song E., Ma G., Xu X., Jin R., Liu T., Hung Ch.-Ch. Multi-branch ensemble learning architecture based on 3D CNN for false positive reduction in lung nodule detection. IEEE Access. 2019. Vol. 7. P. 67380–67391. https://www.kaggle.com/datasets/mateuszbuda/lgg-mri-segmentation .
- Zhou X., Yamada K., Takayama R., Zhou X., Hara T., Fujita H., Wang S., Kojima T. Performance evaluation of 2D and 3D deep learning approaches for automatic segmentation of multiple organs on CT images. Proc. SPIE 10575, Medical Imaging 2018: Computer-Aided Diagnosis (10–15 February 2018, Houston, Texas, USA). Houston, 2018. Vol. 10575. P. 105752C. https://doi.org/10.1117/12.2295178.
- Srikrishna M., Heckemann R.A., Pereira J.B., Volpe G., Zettergren A., Kern S., Westman E., Skoog I., Schll M. Comparison of two-dimensional- and three-dimensional-based U-Net architectures for brain tissue classification in one-dimensional brain CT. Front. Comput. Neurosci. 2022. Vol. 15. Article number 785244. https://doi.org/10.3389/fncom.2021.785244.
- Stamoulakatos A., Cardona J., Michie C., Andonovic I., Lazaridis P., Bellekens X., Atkinson R., Hossain Md.M., Tachtatzis C. A comparison of the performance of 2D and 3D convolutional neural networks for subsea survey video classification. Proc. OCEANS 2021 San Diego — Porto (20–23 September 2021, San Diego, Portugal). San Diego, 2021. P. 1–10. https://doi.org/10.23919/OCEANS44145.2021.9706125.
- Yang L., Zhang Y., Chen J., Zhang S., Chen D.Z. Suggestive annotation: A deep active learning framework for biomedical image segmentation. In: Medical Image Computing and Computer Assisted Intervention — MICCAI 2017. MICCAI 2017 (11–13 September 2017, Quebec City, QC, Canada). Quebec City, 2017. Lecture Notes in Computer Science. Descoteaux M., Maier-Hein L., Franz A., Jannin P., Collins D., Duchesne S. (Eds.). 2017. Vol. 10435. P. 399–407. https://doi.org/10.1007/978-3-319-66179-7_46.
- Bui T. D., Shin J., Moon T. Skip-connected 3D DenseNet for volumetric infant brain MRI segmentation. Biomed. Signal Process. Control. 2019. Vol. 54, Article 101613. https://doi.org/10.1016/j.bspc.2019.101613.
- Anaraki A.K., Ayati M., Kazemi F. Magnetic resonance imaging-based brain tumor grades classification and grading via convolutional neural networks and genetic algorithms. Biocybernetics and Biomedical Engineering. 2019. Vol. 39, N. 1. P. 63–74.
- Ozyurt F., Sert E., Avci E., Dogantekin E. Brain tumor detection based on a convolutional neural network with neutrosophicexpert maximum fuzzy sure entropy. Measurement. 2019. Vol. 147, Article 106830.
- Ghosh S., Santosh Kc. Tumor segmentation in brain MRI: U-Nets versus feature pyramid network. Proc. 2021 IEEE 34th International Symposium on Computer-Based Medical Systems (CBMS) (1–9 June 2021, Online event). P. 31–36. https://doi.org/10.1109/CBMS52027.2021.00013.
- Li Z., Wang Y., Yu J., Shi Z., Guo Y., Chen L., Mao Y. Low-grade glioma segmentation based on CNN with fully connected CRF. J. Healthc. Eng. 2017. Vol. 2017. Article Gs9283480. https://doi.org/10.1155/2017/9283480.