免疫組織化學 (IHC) 是一種廣泛使用標記特定有機物質的方法。其原理為利用抗原和抗體間的專一性結合，使得目標之具有抗原性物質在切片(WSI)中呈色。免疫組織化學染色常被醫生在醫學研究中用於標記特定物質，計算被標記細胞的病理參數，最後觀察其比例、細胞數量、型態參數與癌症腫瘤、治療反應、或疾病感染之關聯性，用以開發疾病診斷的預測方法。本論文的主題乙型肝炎常透過量化免疫組織化學 (IHC) 中的乙型肝炎表面抗原 (HBsAg) 強度來檢測。並且根據以往的研究顯示，HBsAg強度與肝癌復發之間可能存在相關性。然而，現行HBsAg強度是由人為觀察的，不僅費力且主觀。因此，本研究提出了一種用於量化染色強度的自動統計方法。在此方法中結合了卷積神經網絡和影像處理，用以獲得 IHC 染色 WSI 中的細胞染色強度資訊。此外，WSI 中的每個像素都被分類為 0+ ~ 3+ 的染色強度，其中 0+ 表示最小 HBsAg 染色強度，3+ 表示最大 HBsAg 染色強度。本研究使用了約 40 個臨床診斷肝病患者的病理切片( WSI) 所組成的自建數據集以此評估所提出方法的可行性。根據實驗結果表明，所提出的方法同時具有良好的檢測質量與分級性能。總體而言，此方法為自動化量化 IHC 染色 WSI 中 HBsAg 強度的任務提供了一個可靠的解決方案。

Immunohistochemistry (IHC) is a widely used method to mark specific organic substances. The principle is to use a specific combination of antigens and antibodies to make the target antigenic substances appear in the whole slide image (WSI). Immunohistochemical staining is often used in medical research by physicians to mark specific substances, to calculate the pathological parameters of the marked cells, and finally to observe the proportion of expression, cell number, and correlation of the phenotypic parameters with cancer tumors, therapeutic responses, or disease infections in order to develop predictive methods for disease diagnosis. Hepatitis B, the subject of this thesis, is most commonly detected by quantifying the Hepatitis B Surface Antigen(HBsAg) intensity in immunohistochemistry (IHC). Previous studies have shown that there may be a correlation between the HBsAg intensity and liver cancer recurrence. However, the HBsAg intensity is usually observed manually, which is laborious and subjective. Accordingly, this study proposes an automated statistical method for quantifying the staining intensity. In the proposed approach, a Convolution Neural Network and image processing operations are combined to obtain the cell staining intensity information in an IHC-stained WSI. In particular, each cell in the WSI is assigned a grade of 0+ ~ 3+, where 0+ represents the minimum HBsAg intensity and 3+ represents the maximum HBsAg intensity. The feasibility of the proposed approach is evaluated using a self-compiled dataset consisting of around 40 WSIs of real-world patients with clinically-diagnosed liver disease. The results show that the proposed method achieves both a good detection quality and a good classification performance. Overall, the results indicate that the proposed approach provides a promising solution for automating the task of quantifying the HBsAg intensity in IHC-stained WSIs.

[1] World Health Organization, “Global Hepatitis Report 2017,” Geneva;2017
[2] "Cancer",Who.int,2021. Available:https://www.who.int/news-room/fact-sheets/detail/cancer.
[3] L. Zhang, X.-Y. Xie, Y. Chen, N.-L. Ge, R.-X. Chen, Y.-H. Gan, B.-H. Zhang, Y.-H. Wang, and Z.-G. Ren, “Hepatitis B surface antigen predicts recurrence after radiofrequency ablation in patients with low hepatitis B virus loads,” Medicine, vol. 96, no. 52, 2017.
[4] H.-W. Tsai, Y.-J. Lin, P.-W. Lin, H.-C. Wu, K.-H. Hsu, C.-J. Yen, S.-H. Chan, W. Huang, and I.-J. Su, “A clustered ground-glass hepatocyte pattern represents a new prognostic marker for the recurrence of hepatocellular carcinoma after surgery,” Cancer, vol. 117, no. 13, pp. 2951–2960, 2011.
[5] K. Sirinukunwattana, J. P. W. Pluim, H. Chen, X. Qi, P.-A. Heng, Y. B. Guo, L. Y. Wang, B. J. Matuszewski, E. Bruni, U. Sanchez, A. Böhm, O. Ronneberger, B. B. Cheikh, D. Racoceanu, P. Kainz, M. Pfeiffer, M. Urschler, D. R. J. Snead, and N. M. Rajpoot, “Gland segmentation in colon histology images: The glas challenge contest,” Medical Image Analysis, vol. 35, pp. 489–502, 2017.
[6] L. Zhang, X.-Y. Xie, Y. Chen, N.-L. Ge, R.-X. Chen, Y.-H. Gan, B.-H. Zhang, Y.-H. Wang, and Z.-G. Ren, “Hepatitis B surface antigen predicts recurrence after radiofrequency ablation in patients with low hepatitis B virus loads,” Medicine, vol. 96, no. 52, 2017.
[7] Y. Song, H. Chang, H. Huang, and W. Cai, “Supervised Intra-embedding of Fisher Vectors for Histopathology Image Classification,” Medical Image Computing and Computer Assisted Intervention − MICCAI 2017, pp. 99–106, 2017.
[8] J. Wang, J. D. MacKenzie, R. Ramachandran, and D. Z. Chen, “Identifying Neutrophils in H&E Staining Histology Tissue Images,” Medical Image Computing and Computer-Assisted Intervention – MICCAI 2014, pp. 73–80, 2014.
[9] R. Duggal, A. Gupta, R. Gupta, and P. Mallick, “SD-Layer: Stain Deconvolutional Layer for CNNs in Medical Microscopic Imaging,” Medical Image Computing and Computer Assisted Intervention − MICCAI 2017, pp. 435–443, 2017.
[10] N.-A. Pham, A. Morrison, J. Schwock, S. Aviel-Ronen, V. Iakovlev, M.-S. Tsao, J. Ho, and D. W. Hedley, “Quantitative image analysis of immunohistochemical stains using a CMYK color model,” Diagnostic Pathology, vol. 2, no. 1, p. 8, 2007.
[11] Kaczmarek E, Górna A, Majewski P. “Techniques of image analysisfor quantitative immunohistochemistry,” Rocz Akad Med Bialymst.2004;49(suppl 1):155-158.
[12] Krizhevsky, A., Sutskever, I., and Hinton, G. E. “Imagenet classification with deep convolutional neural networks,” In Advances in neural information processing systems, pp. 1097–1105, 2012.
[13] Simonyan, K. and Zisserman, A. “Very deep convolutional networks for large-scale image recognition,” arXiv preprint arXiv:1409.1556, 2014.
[14] C. Szegedy, Wei Liu, Yangqing Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A. Rabinovich, “Going deeper with convolutions,” 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015.
[15] K. He, X. Zhang, S. Ren, and J. Sun, “Deep Residual Learning for Image Recognition,” 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016.
[16] C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna, “Rethinking the Inception Architecture for Computer Vision,” 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016.
[17] S. Xie, R. Girshick, P. Dollar, Z. Tu, and K. He, “Aggregated Residual Transformations for Deep Neural Networks,” 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017.
[18] R. Girshick, J. Donahue, T. Darrell, and J. Malik, “Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation,” 2014 IEEE Conference on Computer Vision and Pattern Recognition, 2014.
[19] K. He, X. Zhang, S. Ren, and J. Sun, “Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition,” Computer Vision – ECCV 2014, pp. 346–361, 2014.
[20] R. Girshick, “Fast R-CNN,” 2015 IEEE International Conference on Computer Vision (ICCV), 2015.
[21] S. Ren, K. He, R. Girshick, and J. Sun, “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 6, pp. 1137–1149, 2017.
[22] W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S. Reed, C.-Y. Fu, and A. C. Berg, “SSD: Single Shot MultiBox Detector,” Computer Vision – ECCV 2016, pp. 21–37, 2016. [23] Long, J., Shelhamer, E., & Darrell, T. (2015). Fully convolutional networks for semantic segmentation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 3431-3440).
[24] L.-C. Chen, G. Papandreou, I. Kokkinos, K. Murphy, and A. L. Yuille, “DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 40, no. 4, pp. 834–848, 2018.
[25] V. Badrinarayanan, A. Kendall, and R. Cipolla, “SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 12, pp. 2481–2495, 2017.
[26] J. Dai, K. He, Y. Li, S. Ren, and J. Sun, “Instance-Sensitive Fully Convolutional Networks,” Computer Vision – ECCV 2016, pp. 534–549, 2016.
[27] K. He, G. Gkioxari, P. Dollar, and R. Girshick, “Mask R-CNN,” 2017 IEEE International Conference on Computer Vision (ICCV), 2017.
[28] S. Graham, Q. D. Vu, S. E. Raza, A. Azam, Y. W. Tsang, J. T. Kwak, and N. Rajpoot, “Hover-Net: Simultaneous segmentation and classification of nuclei in multi-tissue histology images,” Medical Image Analysis, vol. 58, p. 101563, 2019.
[29] F. Sheikhzadeh, R. K. Ward, D. van Niekerk, and M. Guillaud, “Automatic labeling of molecular biomarkers of immunohistochemistry images using fully convolutional networks,” PLOS ONE, vol. 13, no. 1, 2018.
[30] A. Lahiani, J. Gildenblat, I. Klaman, N. Navab, and E. Klaiman, “Generalising multistain immunohistochemistry tissue segmentation using end‐to‐end colour deconvolution deep neural networks,” IET Image Processing, vol. 13, no. 7, pp. 1066–1073, 2019.
[31] O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation,” Lecture Notes in Computer Science, pp. 234–241, 2015. [32] T. Chen and C. Chefd’hotel, “Deep Learning Based Automatic Immune Cell Detection for Immunohistochemistry Images,” Machine Learning in Medical Imaging, pp. 17–24, 2014.
[33] X. Tian, Y. Chong, Y. Huang, P. Guo, M. Li, W. Zhang, Z. Du, X. Li, and Y. Hao, “Using Machine Learning Algorithms to Predict Hepatitis B Surface Antigen Seroclearance,” Computational and Mathematical Methods in Medicine, vol. 2019, pp. 1–7, 2019.
[34] T.-Y. Lin, P. Goyal, R. Girshick, K. He, and P. Dollar, “Focal Loss for Dense Object Detection,” 2017 IEEE International Conference on Computer Vision (ICCV), 2017.
[35] Blog.csdn.net, "【OpenCV】HSV color recognition-HSV basic color component range. " Available:https://blog.csdn.net/Taily_Duan/article/details/51506776.