根據世界衛生組織的資料，隨著紫外線每年不斷地增強，皮膚癌的人數也會持續增加。然而皮膚癌演變過程至今仍然沒有可信賴參考的依據，如果能夠提前發現是否罹患皮膚癌，便能提早接受治療。對於醫生或是病理學家來說，標記皮膚病變位置及其種類，有助於醫學上對於皮膚癌的研究。然而標記皮膚病變是相當費時的，因此有自動影像分割的需求。
本篇論文的目的在於開發一個可靠的影像分割技術。相比於機器學習以及傳統影像處理，利用深度學習的方式進行皮膚癌影像分割可以大幅改善分割準確度，然而在邊界處理上，仍然有許多進步的空間。因此，本論文提出了一個新穎的兩階段影像分割網路。第一階段模型初步分割出皮膚病變影像；第二階段模型將初步分割的影像進行邊緣強化，並提升深度網路的強健性和泛化程度。除了兩階段模型外，本方法的特色在於使用了深度監督和空間及通道注意力機制。藉由我們提出的方法，分別在ISIC2017 與ISIC2018 的資料集上，能夠取得0.796 與0.827 的平均Jaccard Index。與先前的方法相比，我們所提出的方法具有最先進的表現。

According to the World Health Organization, the number of skin cancer cases will continue to increase, since the UV radiation level becomes higher every year. Besides, since the reliable registration of the disease has not been achieved, to determine the temporal trends of the disease still has its difficulty. Thus, if the patients can discover early skin cancer, they can have well treatment. Skin lesion image segmentation can help the doctor to diagnose and analyze the skin lesion effectively, but it is time-consuming for doctors and pathologists to manually mark the regions and categories of the skin lesions. Therefore, the need for automatic image segmentation has arisen.
The purpose of this thesis is to develop a reliable image segmentation algorithm that can improve segmentation performance. The skin lesion segmentation methods using deep learning have significant improvements over machine learning and traditional image processing approaches. However, there is still much room for improvement in the boundary information. Therefore, we proposed a novel two-stage network to segment the skin lesion image. The first stage model preliminary segments the skin lesion images, and the second stage model aims to enhance the boundary information. In this way, the robustness and generalization of the network can be improved. The two-stage model features the deep supervision, and the channel and spatial attention mechanism. Our proposed method has trained on the training sets of ISIC2017 and ISIC2018 and obtained an average Jaccard Index of 0.796 and 0.827, respectively on the test set. Our proposed method has achieved state-of-the-art performance compared to the previous methods.

[1] The World Health Organization "Radiation: Ultraviolet (UV) Radiation and Skin Cancer." https://www.who.int/news-room/questions-and-answers/item/radiation-ultraviolet-(uv)-radiation-and-skin-cancer (last accessed 8/31, 2022).
[2] R. L. Siegel, K. D. Miller, S. A. Fedewa, D. J. Ahnen, R. G. Meester, A. Barzi, and A. Jemal, "Colorectal Cancer Statistics" in A Cancer Journal for Clinicians, 2017, vol. 67, no. 3, pp. 177-193.
[3] P. Kharazmi, M. I. AlJasser, H. Lui, Z. J. Wang, and T. K. Lee, "Automated Detection and Segmentation of Vascular Structures of Skin Lesions Seen in Dermoscopy, with an Application to Basal Cell Carcinoma Classification," in IEEE Journal of Biomedical and Health Informatics, 2016, vol. 21, no. 6, pp. 1675-1684.
[4] B. Lei, Z. Xia, F. Jiang, X. Jiang, Z. Ge, Y. Xu, J. Qin, S. Chen, T. Wang, and S. Wang, "Skin Lesion Segmentation via Generative Adversarial Networks with Dual Discriminators," in Medical Image Analysis, 2020, vol. 64, Art. no. 101716.
[5] A. Adegun and S. Viriri, "Deep Learning Techniques for Skin Lesion Analysis and Melanoma Cancer Detection: A Survey of State-of-the-art," in Artificial Intelligence Review, 2021, vol. 54, no. 2, pp. 811-841.
[6] H.-J. Jeong, T.-Y. Kim, H.-G. Hwang, H.-J. Choi, H.-S. Park, and H.-K. Choi, "Comparison of Thresholding Methods for Breast Tumor Cell Segmentation," in Proceedings of the International Workshop on Enterprise Networking and Computing in Healthcare Industry, 2005, pp. 392-395.
[7] A. T. Beuren, R. Janasieivicz, G. Pinheiro, N. Grando, and J. Facon, "Skin Melanoma Segmentation by Borphological Approach," in Proceedings of the International Conference on Advances in Computing, Communications and Informatics, 2012, pp. 972-978.
[8] A. Mouelhi, M. Sayadi, and F. Fnaiech, "Automatic Segmentation of Clustered Breast Cancer Cells Using Watershed and Concave Vertex Graph," in Proceedings of the International Conference on Communications, Computing and Control Applications, 2011, pp. 1-6.
[9] E. Ahn, J. Kim, L. Bi, A. Kumar, C. Li, M. Fulham, and D. D. Feng, "Saliency-Based Lesion Segmentation via Background Detection in Dermoscopic Images," in IEEE Journal of Biomedical and Health Informatics, 2017, vol. 21, no. 6, pp. 1685-1693.
[10] Z.-Q. Zhao, P. Zheng, S.-T. Xu, and X. Wu, "Object Detection with Deep Learning: A Review," in IEEE Transactions on Neural Networks and Learning Systems, 2019, vol. 30, no. 11, pp. 3212-3232.
[11] J. Long, E. Shelhamer, and T. Darrell, "Fully Convolutional Networks for Semantic Segmentation," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2015, pp. 3431-3440.
[12] O. Ronneberger, P. Fischer, and T. Brox, "U-net: Convolutional Networks for Biomedical Image Segmentation," in Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention, 2015, pp. 234-241.
[13] Y. Yuan and Y.-C. Lo, "Improving Dermoscopic Image Segmentation with Enhanced Convolutional-Deconvolutional Networks," in IEEE Journal of Biomedical and Health Informatics, 2017, vol. 23, no. 2, pp. 519-526.
[14] Y. Song, E.-L. Tan, X. Jiang, J.-Z. Cheng, D. Ni, S. Chen, B. Lei, and T. Wang, "Accurate Cervical Cell Segmentation from Overlapping Clumps in Pap Smear Images," in IEEE Transactions on Medical Imaging, 2016, vol. 36, no. 1, pp. 288-300.
[15] R. Wang, S. Chen, C. Ji, and Y. Li, "Cascaded Context Enhancement Network for Automatic Skin Lesion Segmentation," in Expert Systems with Applications, 2022, vol. 201, Art. no. 117069.
[16] K. He, X. Zhang, S. Ren, and J. Sun, "Deep Residual Learning for Image Recognition," in Proceedings of the in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 770-778.
[17] O. Oktay, J. Schlemper, L. Le Folgoc, M. Lee, M. Heinrich, K. Misawa, K. Mori, S. McDonagh, N. Y. Hammerla, and B. Kainz, "Attention U-net: Learning Where to Look for the Pancreas," in Proceedings of Medical Imaging with Deep Learning, 2018. Art. no. 9.
[18] S. Woo, J. Park, J.-Y. Lee, and I. S. Kweon, "Cbam: Convolutional Block Attention Module," in Proceedings of the European Conference on Computer Vision (ECCV), 2018, pp. 3-19.
[19] C.-Y. Lee, S. Xie, P. Gallagher, Z. Zhang, and Z. Tu, "Deeply-Supervised Nets," in Proceedings of the PMLR International Conference on Artificial Intelligence and Statistics, 2015, pp. 562-570.
[20] L. Wang, C.-Y. Lee, Z. Tu, and S. Lazebnik, "Training Deeper Convolutional Networks with Deep Supervision," arXiv preprint arXiv:1505.02496, 2015.
[21] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, "Image Quality Assessment: from Error Visibility to Structural Similarity," in IEEE Transactions on Image Processing, 2004, vol. 13, no. 4, pp. 600-612.
[22] X. Li, X. Sun, Y. Meng, J. Liang, F. Wu, and J. Li, "Dice Loss for Data-Imbalanced NLP Tasks," in Proceedings of the Annual Meeting of the Association for Computational Linguistics, 2019, pp. 465–476.
[23] D. Zhou, J. Fang, X. Song, C. Guan, J. Yin, Y. Dai, and R. Yang, "IoU Loss for 2D/3D Object Detection," in Proceedings of the IEEE International Conference on 3D Vision (3DV), 2019, pp. 85-94.
[24] N. C. Codella, D. Gutman, M. E. Celebi, B. Helba, M. A. Marchetti, S. W. Dusza, A. Kalloo, K. Liopyris, N. Mishra, and H. Kittler, "Skin Lesion Analysis Toward Melanoma Detection: A Challenge at the 2017 International Symposium on Biomedical Imaging (ISBI), hosted by the International Skin Imaging Collaboration (ISIC)," in Proceedings of the IEEE International Symposium on Biomedical Imaging (ISBI 2018), 2018, pp. 168-172.
[25] N. Codella, V. Rotemberg, P. Tschandl, M. E. Celebi, S. Dusza, D. Gutman, B. Helba, A. Kalloo, K. Liopyris, and M. Marchetti, "Skin Lesion Analysis Toward Melanoma Detection 2018: A Challenge Hosted by the International Skin Imaging Collaboration (ISIC)," arXiv preprint arXiv:1902.03368, 2019.
[26] M. E. Celebi, Q. Wen, S. Hwang, H. Iyatomi, and G. Schaefer, "Lesion Border Detection in Dermoscopy Images Using Ensembles of Thresholding Methods," in Skin Research and Technology, 2013, vol. 19, no. 1, pp. 252-258.
[27] M. Silveira, J. C. Nascimento, J. S. Marques, A. R. Marçal, T. Mendonça, S. Yamauchi, J. Maeda, and J. Rozeira, "Comparison of Segmentation Methods for Melanoma Diagnosis in Dermoscopy Images," in IEEE Journal of Selected Topics in Signal Processing, 2009, vol. 3, no. 1, pp. 35-45.
[28] Z. Ma and J. M. R. S. Tavares, "A Novel Approach to Segment Skin Lesions in Dermoscopic Images Based on A Deformable Model," in IEEE Journal of Biomedical and Health Informatics, 2015, vol. 20, no. 2, pp. 615-623.
[29] A. Gençtav, S. Aksoy, and S. Önder, "Unsupervised Segmentation and Classification of Cervical Cell Images," in Pattern Recognition, 2012, vol. 45, no. 12, pp. 4151-4168.
[30] N. A. M. Isa, S. Sabarudin, U. K. Ngah, and K. Z. Zamli, "Automatic Detection of Breast Tumours from Ultrasound Images Using the Modified Seed Based Region Growing Technique," in Proceedings of the International Conference on Knowledge-Based and Intelligent Information and Engineering Systems, 2005, pp. 138-144.
[31] C. Bergmeir, M. G. Silvente, and J. M. Benítez, "Segmentation of Cervical Cell Nuclei in High-Resolution Microscopic Images: A New Algorithm and A Web-Based Software Framework," in Computer Methods and Programs in Biomedicine, 2012, vol. 107, no. 3, pp. 497-512.
[32] P. Bao, L. Zhang, and X. Wu, "Canny Edge Detection Enhancement by Scale Multiplication," in IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, vol. 27, no. 9, pp. 1485-1490.
[33] R. A. McLaughlin, "Randomized Hough Transform: Better Ellipse Detection," in Proceedings of Digital Processing Applications, 1996, vol. 1, pp. 409-414.
[34] Z. Mirikharaji and G. Hamarneh, "Star Shape Prior in Fully Convolutional Networks for Skin Lesion Segmentation," in Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention, 2018, pp. 737-745.
[35] L. Bi, J. Kim, E. Ahn, A. Kumar, M. Fulham, and D. Feng, "Dermoscopic Image Segmentation via Multistage Fully Convolutional Networks," in IEEE Transactions on Biomedical Engineering, 2017, vol. 64, no. 9, pp. 2065-2074.
[36] H. Li, X. He, F. Zhou, Z. Yu, D. Ni, S. Chen, T. Wang, and B. Lei, "Dense Deconvolutional Network for Skin Lesion Segmentation," in IEEE Journal of Biomedical and Health Informatics, 2018, vol. 23, no. 2, pp. 527-537.
[37] C. Shorten and T. M. Khoshgoftaar, "A Survey on Image Data Augmentation for Deep Learning," in Journal of Big Data, 2019, vol. 6, no. 1, pp. 1-48.
[38] Y. Li and L. Shen, "Skin Lesion Analysis Towards Melanoma Detection Using Deep Learning Network," in Sensors, 2018, vol. 18, no. 2, Art. no. 556.
[39] M. Sarker, M. Kamal, H. A. Rashwan, F. Akram, S. F. Banu, A. Saleh, V. K. Singh, F. U. Chowdhury, S. Abdulwahab, and S. Romani, "SLSDeep: Skin Lesion Segmentation Based on Dilated Residual and Pyramid Pooling Networks," in Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention, 2018, pp. 21-29.
[40] A. Srivastava, D. Jha, S. Chanda, U. Pal, H. D. Johansen, D. Johansen, M. A. Riegler, S. Ali, and P. Halvorsen, "MSRF-Net: A Multi-Scale Residual Fusion Network for Biomedical Image Segmentation," in IEEE Journal of Biomedical and Health Informatics, 2021, vol. 26, no. 5, pp. 2252-2263.
[41] G. J. Chowdary and G. Yathisha, "Exploring Dual-Attention Mechanism with Multi-Scale Feature Extraction Scheme for Skin Lesion Segmentation," arXiv preprint arXiv:2111.08708, 2021.
[42] E. D. Cubuk, B. Zoph, J. Shlens, and Q. V. Le, "Randaugment: Practical Automated Data Augmentation with A Reduced Search Space," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPR), 2020, pp. 702-703.
[43] M. Jahanifar, N. Z. Tajeddin, N. A. Koohbanani, A. Gooya, and N. Rajpoot, "Segmentation of Skin Lesions and Their Attributes Using Multi-Scale Convolutional Neural Networks and Domain Specific Augmentations," arXiv preprint arXiv:1809.10243, 2018.
[44] Y. Liu, H. D. Cheng, J. Huang, Y. Zhang, and X. Tang, "An Effective Approach of Lesion Segmentation within the Breast Ultrasound Image Based on the Cellular Automata Principle," in Journal of Digital Imaging, 2012, vol. 25, no. 5, pp. 580-590.
[45] M. Combalia, F. Hueto, S. Puig, J. Malvehy, and V. Vilaplana, "Uncertainty Estimation in Deep Neural Networks for Dermoscopic Image Classification," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPR), 2020, pp. 3211-3220.
[46] X. Qin, Z. Zhang, C. Huang, C. Gao, M. Dehghan, and M. Jagersand, "Basnet: Boundary-aware Salient Object Detection," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 7479-7489.
[47] V. Badrinarayanan, A. Kendall, and R. Cipolla, "Segnet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation," in IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, vol. 39, no. 12, pp. 2481-2495.
[48] Y. Zhu, C. Chen, G. Yan, Y. Guo, and Y. Dong, "AR-Net: Adaptive Attention and Residual Refinement Network for Copy-Move Forgery Detection," in IEEE Transactions on Industrial Informatics, 2020, vol. 16, no. 10, pp. 6714-6723.
[49] S. Ioffe and C. Szegedy, "Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift," in Proceedings of the International Conference on Machine Learning, 2015, pp. 448-456.
[50] M. A. Khan, Y.-D. Zhang, M. Sharif, and T. Akram, "Pixels to Classes: Intelligent Learning Framework for Multiclass Skin Lesion Localization and Classification," in Computers & Electrical Engineering, 2021, vol. 90, Art. no. 106956.
[51] Y. Zhou, B. Wang, L. Huang, S. Cui, and L. Shao, "A Benchmark for Studying Diabetic Retinopathy: Segmentation, Grading, and Transferability," in IEEE Transactions on Medical Imaging, 2020, vol. 40, no. 3, pp. 818-828.
[52] G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, "Densely Connected Convolutional Networks," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 4700-4708.
[53] D. Misra, "Mish: A Self Regularized Non-Monotonic Neural Activation Function," arXiv preprint arXiv:1908.08681, 2019.
[54] T. L. B. Khanh, D.-P. Dao, N.-H. Ho, H.-J. Yang, E.-T. Baek, G. Lee, S.-H. Kim, and S. B. Yoo, "Enhancing U-Net with Spatial-Channel Attention Gate for Abnormal Tissue Segmentation in Medical Imaging," in Applied Sciences, 2020, vol. 10, no. 17, Art. no. 5729.
[55] Z. Deng, X. Hu, L. Zhu, X. Xu, J. Qin, G. Han, and P.-A. Heng, "R3net: Recurrent Residual Refinement Network for Saliency Detection," in Proceedings of the International Joint Conference on Artificial Intelligence, 2018, pp. 684-690.
[56] K. He, X. Zhang, S. Ren, and J. Sun, "Delving Deep into Rectifiers: Surpassing Human-Level Performance on Imagenet Classification," in Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2015, pp. 1026-1034.
[57] F. Zhuang, Z. Qi, K. Duan, D. Xi, Y. Zhu, H. Zhu, H. Xiong, and Q. He, "A Comprehensive Survey on Transfer Learning," in Proceedings of the IEEE, 2020, vol. 109, no. 1, pp. 43-76.
[58] I. Bello, B. Zoph, V. Vasudevan, and Q. V. Le, "Neural Optimizer Search with Reinforcement Learning," in Proceedings of the International Conference on Machine Learning, 2017, pp. 459-468.
[59] D. P. Kingma and J. Ba, "Adam: A Method for Stochastic Optimization," in arXiv preprint arXiv:1412.6980, 2014.
[60] N. Qian, "On the Momentum Term in Gradient Descent Learning Algorithms," in Neural Networks, 1999, vol. 12, no. 1, pp. 145-151.
[61] J. Duchi, E. Hazan, and Y. Singer, "Adaptive Subgradient Methods for Online Learning and Stochastic Optimization," in Journal of Machine Learning Research, 2011, vol. 12, no. 7, pp. 2121–2159.
[62] P. Tschandl, C. Rosendahl, and H. Kittler, "The HAM10000 Dataset, a Large Collection of Multi-Source Dermatoscopic Images of Common Pigmented Skin Lesions," in Scientific data, 2018, vol. 5, no. 1, pp. 1-9.
[63] M. Benčević, I. Galić, M. Habijan, and D. Babin, "Training on Polar Image Transformations Improves Biomedical Image Segmentation," in IEEE Access, 2021, vol. 9, pp. 133365-133375.
[64] M. Berseth, "ISIC 2017-Skin Lesion Analysis towards Melanoma Detection," arXiv preprint arXiv:1703.00523, 2017.
[65] K. Abhishek and G. Hamarneh, "Matthews Correlation Coefficient Loss for Deep Convolutional Networks: Application to Skin Lesion Segmentation," in Proceedings of IEEE International Symposium on Biomedical Imaging (ISBI), 2021: IEEE, pp. 225-229.
[66] L. Bi, J. Kim, E. Ahn, and D. Feng, "Automatic Skin Lesion Analysis Using Large-Scale Dermoscopy Images and Deep Residual Networks," arXiv preprint arXiv:1703.04197, 2017.
[67] Z. Zhou, M. M. R. Siddiquee, N. Tajbakhsh, and J. Liang, "Unet++: A Nested U-net Architecture for Medical Image Segmentation," in Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, 2018, pp. 3-11.
[68] L.-C. Chen, G. Papandreou, F. Schroff, and H. Adam, "Rethinking Atrous Convolution for Semantic Image Segmentation," arXiv preprint arXiv:1706.05587, 2017.
[69] M. Yang, K. Yu, C. Zhang, Z. Li, and K. Yang, " DenseASPP for Semantic Segmentation in Street Scenes," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018, pp. 3684-3692.
[70] R. Gu, G. Wang, T. Song, R. Huang, M. Aertsen, J. Deprest, S. Ourselin, T. Vercauteren, and S. Zhang, "CA-Net: Comprehensive Attention Convolutional Neural Networks for Explainable Medical Image Segmentation," in IEEE Transactions on Medical Imaging, 2020, vol. 40, no. 2, pp. 699-711.
[71] R. Azad, M. Asadi-Aghbolaghi, M. Fathy, and S. Escalera, "Bi-Directional Conv- LSTM U-Net with Densley Connected Convolutions," in Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops(ICCV), 2019, pp. 0-0.