臺灣博碩士論文加值系統

本研究提出一種創新的應用，結合更快的區域卷積神經網絡(Faster Region with Convolution Neural Network, Faster R-CNN)與主動形狀模型(Active Shape Model, ASM)應用於心臟左心室超音波影像之自動分割。由於左心室形狀與外觀變化甚大，但藉由深度學習，我們僅須透過小的訓練集就可以適應不同的心臟相位和案例。
比起其他傳統的分割方法，ASM是一種基於模型的分割方法，其中包含訓練與統計分析。CNN在目標識別方面表現優異，成為許多目標識別挑戰中的首選算法。而Faster R-CNN使用CNN提取影像特徵，改進區域提案方法，與Fast R-CNN檢測網路共享卷積特徵，使得目標檢測和識別近乎同時。
本研究詳細描述了Faster R-CNN與ASM算法，並用於心臟左心室超音波影像。我們測試了醫生提供之臨床數據，結果證實我們的方法在完全自動化挑戰中達到了一定的準確率，且具備非常有競爭力的執行時間。

We introduce an innovative application that combines Faster Region with Convolution Neural Network (Faster R-CNN) and Active Shape Model (ASM) for automated segmentation of the left ventricle of the heart from ultrasound data. This combination is relevant for segmentation problems, where the left ventricle presents large shape and appearance variations, but we can use only small annotated training set to adapt different heart phase and case by deep learning.
Compare to other tradition segmentation methods, ASM is a model-based segmentation methodology which incorporates training and statistical analysis. CNN is excellent in target recognition and it became the preferred algorithm in many target recognition challenge. Faster R-CNN uses the CNN to extract the image features, improves the region proposal method, shares the convolution feature with the Fast R-CNN, and makes the target detection and recognition almost instantaneous.
This study describes the Faster R-CNN and ASM algorithms in detail and used on left ventricular ultrasound images. We test our methodology on the data from clinicians, and our approach achieves the equivalent accuracy in the state-of-the-art results for the fully automated challenge, while having very competitive execution time.

目錄
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機 5
1.3 論文貢獻 9
1.4 論文架構 10
第二章 文獻探討 11
2.1 半自動方法 11
2.2 全自動方法 11
第三章 材料與研究方法 17
3.1 實驗材料 17
3.2 方法流程圖 18
3.3 改進之自適應非等向性擴散濾波 19
3.4 基於更快地區域卷積神經網路的左心室識別 22
3.4.1 預訓練CNN模型 23
3.4.2 訓練RPN 24
3.4.3 訓練Fast R-CNN檢測網路 25
3.4.4 兩個網路共享卷積特徵和聯合調優 25
3.4.5 檢測識別過程 26
3.5 主動形狀模型 27
3.5.1 建立形狀模型 27
3.5.2 建立局部紋理模型 29
3.5.3 特徵點搜尋 29
3.6 評估指標 30
第四章 實驗結果與討論 32
4.1 實驗環境 32
4.2 結果與討論 32
第五章 結論與未來展望 36
5.1 本研究之結論 36
5.2 研究限制 37
5.3 未來展望 37
參考文獻 38

Adams, R., & Bischof, L. (1994). Seeded region growing. IEEE Transactions on pattern analysis and machine intelligence, 16(6), 641-647.
Bookstein, F. L. (1997). Landmark methods for forms without landmarks: morphometrics of group differences in outline shape. Medical image analysis, 1(3), 225-243.
Bosch, J. G., Mitchell, S. C., Lelieveldt, B. P., Nijland, F., Kamp, O., Sonka, M., & Reiber, J. H. (2002). Automatic segmentation of echocardiographic sequences by active appearance motion models. IEEE transactions on medical imaging, 21(11), 1374-1383.
Carneiro, G., & Nascimento, J. C. (2013). Combining multiple dynamic models and deep learning architectures for tracking the left ventricle endocardium in ultrasound data. IEEE transactions on pattern analysis and machine intelligence, 35(11), 2592-2607.
Carneiro, G., Nascimento, J. C., & Freitas, A. (2012). The segmentation of the left ventricle of the heart from ultrasound data using deep learning architectures and derivative-based search methods. IEEE Transactions on Image Processing, 21(3), 968-982.
Chao, S. M., & Tsai, D. M. (2008). An anisotropic diffusion-based defect detection for low-contrast glass substrates. Image and Vision Computing, 26(2), 187-200.
Cootes, T. F., & Taylor, C. J. (1994, October). Using grey-level models to improve active shape model search. In Pattern Recognition, 1994. Vol. 1-Conference A: Computer Vision & Image Processing., Proceedings of the 12th IAPR International Conference on (Vol. 1, pp. 63-67). IEEE.
Cootes, T. F., Taylor, C. J., Cooper, D. H., & Graham, J. (1995). Active shape models-their training and application. Computer vision and image understanding, 61(1), 38-59.
Cootes, T. F., & Taylor, C. J. (2004). Statistical models of appearance for computer vision.
Dalal, N., & Triggs, B. (2005, June). Histograms of oriented gradients for human detection. In Computer Vision and Pattern Recognition, 2005. CVPR 2005. IEEE Computer Society Conference on (Vol. 1, pp. 886-893). IEEE.
Dice, L. R. (1945). Measures of the amount of ecologic association between species. Ecology, 26(3), 297-302.
Dietenbeck, T., Alessandrini, M., Barbosa, D., D’hooge, J., Friboulet, D., & Bernard, O. (2012). Detection of the whole myocardium in 2D-echocardiography for multiple orientations using a geometrically constrained level-set. Medical image analysis, 16(2), 386-401.
Grayburn, P. A., Weissman, N. J., & Zamorano, J. L. (2012). Quantitation of mitral regurgitation. Circulation, 126(16), 2005-2017.
Girshick, R. (2015). Fast r-cnn. In Proceedings of the IEEE International Conference on Computer Vision (pp. 1440-1448).
Girshick, R., Donahue, J., Darrell, T., & Malik, J. (2014). Rich feature hierarchies for accurate object detection and semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 580-587).
Haak, A., Vegas-Sanchez-Ferrero, G., Mulder, H. W., Ren, B., Kirişli, H. A., Metz, C., ... & van Walsum, T. (2015). Segmentation of multiple heart cavities in 3-D transesophageal ultrasound images. IEEE transactions on ultrasonics, ferroelectrics, and frequency control, 62(6), 1179-1189.
Hamou, A. K., & El-Sakka, M. R. (2009). Optical flow active contours with primitive shape priors for echocardiography. EURASIP Journal on Advances in Signal Processing, 2010(1), 836753.
Hansson, M., Brandt, S. S., Lindström, J., Gudmundsson, P., Jujić, A., Malmgren, A., & Cheng, Y. (2014). Segmentation of B-mode cardiac ultrasound data by Bayesian Probability Maps. Medical image analysis, 18(7), 1184-1199.
He, K., Zhang, X., Ren, S., & Sun, J. (2014, September). Spatial pyramid pooling in deep convolutional networks for visual recognition. In European Conference on Computer Vision (pp. 346-361). Springer International Publishing.
Heller, S. J., & Carleton, R. A. (1970). Abnormal left ventricular contraction in patients with mitral stenosis. Circulation, 42(6), 1099-1110.
Hoit, B. D. (2011). Strain and strain rate echocardiography and coronary artery disease. Circulation: Cardiovascular Imaging, 4(2), 179-190.
Huang, X., Dione, D. P., Compas, C. B., Papademetris, X., Lin, B. A., Bregasi, A., ... & Duncan, J. S. (2014). Contour tracking in echocardiographic sequences via sparse representation and dictionary learning. Medical image analysis, 18(2), 253-271.
Huttenlocher, D. P., Klanderman, G. A., & Rucklidge, W. J. (1993). Comparing images using the Hausdorff distance. IEEE Transactions on pattern analysis and machine intelligence, 15(9), 850-863.
Kass, M., Witkin, A., & Terzopoulos, D. (1988). Snakes: Active contour models. International journal of computer vision, 1(4), 321-331.
Keraudren, K., Oktay, O., Shi, W., Hajnal, J. V., & Rueckert, D. (2014). Endocardial 3d ultrasound segmentation using autocontext random forests. In Proc. MICCAI CETUS (pp. 41-48).
Landgren, M., Overgaard, N. C., & Heyden, A. (2013, March). Segmentation of the left heart ventricle in ultrasound images using a region based snake. In SPIE Medical Imaging (pp. 866945-866945). International Society for Optics and Photonics.
Meyer, F., & Beucher, S. (1990). Morphological segmentation. Journal of visual communication and image representation, 1(1), 21-46.
Milletari, F., Yigitsoy, M., & Navab, N. (2014). Left ventricle segmentation in cardiac ultrasound using hough-forests with implicit shape and appearance priors.
Dalley, A. F., & Moore, K. L. (1999). Clinically oriented anatomy.
Ng, P. C., & Henikoff, S. (2003). SIFT: Predicting amino acid changes that affect protein function. Nucleic acids research, 31(13), 3812-3814.
Noble, J. A., & Boukerroui, D. (2006). Ultrasound image segmentation: a survey. IEEE Transactions on medical imaging, 25(8), 987-1010.
Perona, P. & Malik, J.(1990). Scale-Space and Edge Detection Using Anisotropic Diffusion. IEEE Trans. on Pattern Analysis and Machine Intelligence. 12(7), 629–639.
Qin, X., Cong, Z., & Fei, B. (2013). Automatic segmentation of right ventricular ultrasound images using sparse matrix transform and a level set. Physics in medicine and biology, 58(21), 7609.
Ramani, G. V., Uber, P. A., & Mehra, M. R. (2010, February). Chronic heart failure: contemporary diagnosis and management. In Mayo Clinic Proceedings (Vol. 85, No. 2, pp. 180-195). Elsevier.
Ren, S., He, K., Girshick, R., & Sun, J. (2015). Faster r-cnn: Towards real-time object detection with region proposal networks. In Advances in neural information processing systems (pp. 91-99).
Sardanelli, F., Quarenghi, M., Di Leo, G., Boccaccini, L., & Schiavi, A. (2008). Segmentation of cardiac cine MR images of left and right ventricles: interactive semiautomated methods and manual contouring by two readers with different education and experience. Journal of Magnetic Resonance Imaging, 27(4), 785-792.
Silva, I., Sanches, J., & Almeida, A. G. Toward a fully automatic left ventricle segmentation using cine-MR images.
Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556.
Starr, C., Evers, C., & Starr, L. (2015). Biology today and tomorrow with physiology. Cengage Learning.
Van Ginneken, B., Frangi, A. F., Staal, J. J., ter Haar Romeny, B. M., & Viergever, M. A. (2002). Active shape model segmentation with optimal features. IEEE transactions on medical imaging, 21(8), 924-933.
Wang, Z., & Zhang, D. (1999). Progressive switching median filter for the removal of impulse noise from highly corrupted images. IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing, 46(1), 78-80.
Wold, S., Esbensen, K., & Geladi, P. (1987). Principal component analysis. Chemometrics and intelligent laboratory systems, 2(1-3), 37-52.
Zuiderveld, K. (1994, August). Contrast limited adaptive histogram equalization. In Graphics gems IV (pp. 474-485). Academic Press Professional, Inc..
Zeiler, M. D., & Fergus, R. (2014, September). Visualizing and understanding convolutional networks. In European conference on computer vision (pp. 818-833). Springer International Publishing.

仉曉紅、田家凱、李曉燕（民 96）。應變和應變率對早期發現糖尿病性心肌病的應用價值。中華超聲影像學雜誌，16（1），33-35。
牛海燕、高宇、張江霞、黃曉玲、丁桂春、王建華（民 101）。二維應變評價原發性高血壓患者左心室收縮功能。中華醫學超聲雜誌，9（9），14-17。
朱文暉（民 101）。三維斑點追蹤技術評價高血壓患者左室旋轉及扭轉。中南大學學報（醫學版），38（3），231-236。
許曼音（民 92）。糖尿病學。上海：上海科學技術出版社。
楊白予、李國傑、朱向明、張霞、胡國兵（民 102）。應變率成像評價2型糖尿病患者左室局部舒張功能的臨床研究。實用醫學影像雜誌，14（3），205-207。
蔣文平（民 94）。室上性快速心律失常治療指南。中華心血管病雜誌，33（1），2-15。
衛生福利部統計處（民 96-105）。【年度死因統計】。未出版之統計數據。
羅偉權、黃志勇、孫健（民 102）。超聲二維應變成像在冠心病心肌收縮運動異常中的應用。當代醫學，19（4），7-8。