4.2. Hướng phát triển
Do bài tốn này khá hạn chế về mặt dữ liệu, tuy thực hiện các biện pháp tăng cường nhưng về bản chất đĩ dữ liệu thực tế vẫn khơng cĩ bổ sung thêm, nên việc bổ sung dữ liệu thật là điều cần thiết. Tìm hiểu thêm các phương pháp xử lý mất cân bằng dữ liệu, và xử lý ảnh y khoa não MRI 3D, tìm cách giảm bớt lượng tài nguyên chi phí tính tốn mỗi lần huấn luyện và kiểm thử. Thay đổi và can thiệp vào hàm mất mát để tăng khả năng giám sát trong quá trình huấn luyện của kiến trúc mạng. Nhằm đưa các thuật tốn này vào hệ thống hỗ trợ phát hiện và chẩn đoán bệnh, giúp nâng cao độ chính xác của việc chẩn đoán mà khơng gây ra các phản ứng xấu lên cơ thể
38
của bệnh nhân kèm theo là sự giảm bớt khối lượng cơng việc và thời gian của các bác sĩ, chuyên gia.
LỜI CẢM ƠN
Em xin cảm ơn thầy Lê Minh Hưng đã giúp đỡ em rất nhiều trong quá trình hồn thành đề tài nghiên cứu khoa học này.
39
TÀI LIỆU THAM KHẢO
[1] Fausto Milletari, Nassir Navab, and Seyed-Ahmad Ahmadi. V-net: Fully convolutional neural networks for volumetric medical image segmentation. In 2016 fourth international conference on 3D vision (3DV), pages 565– 571. IEEE, 2016.
[2] Xudong Wang, Shizhong Han, Yunqiang Chen, Dashan Gao, and Nuno Vasconcelos. Volumetric attention for 3d medical image segmentation and detection. In International conference on medical image computing and computer-assisted intervention, pages 175–184. Springer, 2019.
[3] Benjamin Graham and Laurens van der Maaten. Submanifold sparse convolutional networks. arXiv preprint arXiv:1706.01307, 2017.
[4] Zhuotun Zhu, Yingda Xia, Wei Shen, Elliot Fishman, and Alan Yuille. A 3d coarseto-fine framework for volumetric medical image segmentation. In 2018 International conference on 3D vision (3DV), pages 682–690. IEEE, 2018.
[5] Lyne Tchapmi, Christopher Choy, Iro Armeni, JunYoung Gwak, and Silvio Savarese. Segcloud: Semantic segmentation of 3d point clouds. In 2017 international conference on 3D vision (3DV), pages 537–547. IEEE, 2017.
[6] Hung-Yueh Chiang, Yen-Liang Lin, Yueh-Cheng Liu, and Winston H Hsu. A unified point-based framework for 3d segmentation. In 2019 International Conference on 3D Vision (3DV), pages 155–163. IEEE, 2019.
40
[7] Loic Landrieu and Martin Simonovsky. Large-scale point cloud semantic segmentation with superpoint graphs. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4558–4567, 2018.
[8] Chenfeng Xu, Bichen Wu, Zining Wang, Wei Zhan, Peter Vajda, Kurt Keutzer, and Masayoshi Tomizuka. Squeezesegv3: Spatially-adaptive convolution for efficient point-cloud segmentation. In European Conference on Computer Vision, pages 1–19. Springer, 2020.
[9] Gusi Te, Wei Hu, Amin Zheng, and Zongming Guo. Rgcnn: Regularized graph cnn for point cloud segmentation. In Proceedings of the 26th ACM international conference on Multimedia, pages 746–754, 2018.
[10] David George, Xianghua Xie, and Gary KL Tam. 3d mesh segmentation via multibranch 1d convolutional neural networks. Graphical Models, 96:1–10, 2018.
[11] Truc Le, Giang Bui, and Ye Duan. A multi-view recurrent neural network for 3d mesh segmentation. Computers & Graphics, 66:103–112, 2017.
[12] Rony Kalfarisi, Zheng Yi Wu, and Ken Soh. Crack detection and segmentation using deep learning with 3d reality mesh model for
41
quantitative assessment and integrated visualization. Journal of Computing in Civil Engineering, 34(3):04020010, 2020.
[13] Rui Li and Qingjin Peng. Deep learning-based optimal segmentation of 3d printed product for surface quality improvement and support structure reduction. Journal of Manufacturing Systems, 60:252–264, 2021.
[14] KRIZHEVSKY, Alex; SUTSKEVER, Ilya; HINTON, Geoffrey E. Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems, 2012, 25.
[15] GIRSHICK, Ross, et al. Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition. 2014. p. 580-587.
[16] GIRSHICK, Ross. Fast r-cnn. In: Proceedings of the IEEE international conference on computer vision. 2015. p. 1440-1448.
[17] REN, Shaoqing, et al. Faster r-cnn: Towards real-time object detection with region proposal networks. Advances in neural information processing systems, 2015, 28.
[18] BOCHKOVSKIY, Alexey; WANG, Chien-Yao; LIAO, Hong-Yuan Mark. Yolov4: Optimal speed and accuracy of object detection. arXiv preprint arXiv:2004.10934, 2020.
[19] RONNEBERGER, Olaf; FISCHER, Philipp; BROX, Thomas. U-net: Convolutional networks for biomedical image segmentation. In: International
42
Conference on Medical image computing and computer-assisted intervention. Springer, Cham, 2015. p. 234-241.
[20] ISENSEE, Fabian, et al. nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation. Nature methods, 2021, 18.2: 203-211.
[21] IOFFE, Sergey; SZEGEDY, Christian. Batch normalization: Accelerating deep network training by reducing internal covariate shift. In: International conference on machine learning. PMLR, 2015. p. 448-456. [22] ULYANOV, Dmitry; VEDALDI, Andrea; LEMPITSKY, Victor.
Instance normalization: The missing ingredient for fast stylization. arXiv preprint arXiv:1607.08022, 2016.
[23] VASWANI, Ashish, et al. Attention is all you need. Advances in neural information processing systems, 2017, 30.
[24] HUANG, Gao, et al. Densely connected convolutional networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition. 2017. p. 4700-4708.
[25] YU, Fisher; KOLTUN, Vladlen. Multi-scale context aggregation by dilated convolutions. arXiv preprint arXiv:1511.07122, 2015.