臺灣博碩士論文加值系統

醫師在幫病患診斷或者評估手術時，由於院方目前無法提供完整3D模型以及
現今科技未能拍攝出無死角的立體影像，故醫師們在診斷及術前評估時只能依靠
自身經驗針對電腦斷層掃描(Computerized tomography, CT)影像進行辨識，然而
影像是一種二維資訊的表達，並無法給予醫師們三維空間中的準確資訊，只能憑
藉著醫師專業的訓練以及豐富的經驗在腦海中構築出肝臟的立體外型，再透過口
述跟病患解釋病患目前肝臟的情況，然而病患未接受過專業訓練以致無法辨識電
腦斷層(CT)影像亦無法想像出肝臟的立體外型，這往往會造成些許的醫病問題。
因此，本論文提出一套可以自動辨識電腦斷層(CT)影像中肝臟部位並切割及
即時3D建模之系統，利用圖像分割卷積神經網路(Segnet)可自動將影像進行分割
分類之特性分割電腦斷層(CT)影像結果，自動將肝臟區域分割。由於體內各器官
的影像閥值(Threshold)近似，導致無法直接使圖像分割卷積神經網路(Segnet)對
肝臟部分進行完美的分割，因此本論文將圖像分割卷積神經網路(Segnet)對肝臟
部分的結果在進行一次演算法的判別，利用其結果影像產生點雲(Point Cloud)資
料，最後再做點雲的重建產生肝臟的3D Mesh。

Cause of hospital unable to provide a complete 3D model and nowadays technol-
ogy can not produce stereoscopic images without a blind vision, it cause doctors
misjudgment on surgery evaluated or diagnosed. Therefoe doctors can only rely on
their own experience to identify computerized tomography (CT) images during diag-
nosis and preoperative evaluation. However, images are a kind of two-dimensional
information expression, it cannot provide doctors accurate informations in three-
dimensional space. With professional training and extensive experience in the eld
of clinical, the doctors constructs the stereoscopic shape of the liver in their mind,
and then explains the current liver condition of the patient through oral dictation.
But patient has no professional training to identify the computerized tomography
(CT), thus they can't imagine the three-dimensional shape of the liver in their mind,
which causes many problems between doctor and patient.
This paper proposes a system that can automatically identify liver parts and
instant 3D modeling in computed tomography (CT) images. The image segmenta-
tion convolutional neural network (Segnet) can automatically segment and classify
images into computerized tomography (CT) image results. Cause the image thresh-
old of each organ in the body is similar, it is impossible to directly segment the liver
region by the image segmentation convolutional neural network (Segnet). Therefore,
this paper divides the image into a convolutional neural network (Segnet). The re-
sults of the liver part are judged by an algorithm, and the resulting image is used
to generate point cloud data. Finally the point cloud reconstruction is performed
to generate 3D Mesh of the liver.

中文摘要. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii
Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii
目錄. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
圖目錄. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi
符號說明. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
1 緒論. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 研究背景與動機. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 論文貢獻. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 論文架構. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2 相關研究. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.1 電腦斷層掃描影像介紹. . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.2 立體渲染. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.3 卷積神經網路應用. . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
3 系統架構. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
4 資料準備與標記. . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . 10
4.1 影像閥值調整. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
4.2 圖像分割(Graphcut) . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
5 實驗流程與訓練. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
5.1 圖像分割卷積神經網路. . . . . . . . . . . . . . . . . . . . . . . . . . 15
5.2 標記. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
5.3 原始權重測試. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
5.4 判別. . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . 19
6 邊界演算法及點雲表面重建. . . . . . . . . . . . . . . . . . . . . . . . . . 25
6.1 邊界演算法. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
6.2 柏松表面重建演算法(Poisson Surface Reconstruction) . . . . . . . . . . . . 28
7 實驗結果與討論. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
8 結論與未來工作. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
參考文獻. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

[1] Roberto Cipolla Vijay Badrinarayanan, Alex Kendall. Segnet: A deep convo-
lutional encoder-decoder architecture for image segmentation. IEEE Computer
Society, 2017.
[2] HUGUES HOPPE MICHAEL KAZHDAN. Screened poisson surface recon-
struction. ACM Transactions on Graphics (TOG) 2013, 2013.
[3] Haitao Shuai Yue Huang Jianjun Sun Fenglian Gao Lianfen Huang, Minghui-
Weng. Automatic liver segmentation from ct images using single-block linear
detection. BioMed Research International 2016, 2016.
[4] Marie-pierre Jolly Yuri Y. Boykov. Interactive graph cuts for optimal boundary
region segmentation of objects in n-d images. Proceedings of ICCV 2001., 2001.
[5] http://www.mmh.org.tw/taitam/radio/CT.htm.
[6] Pat Hanrahan Robert A. Drebin, Loren Carpenter. Volume rendering. ACM
SIGGRAPH Computer Graphics, 1988.
[7] https://zh.wikipedia.org/wiki/%E7%AB%8B%E4%BD%93%E6%B8%B2%E6%9F%
93.
[8] D. Grangier, L. Bottou, and R. Collobert. Deep convolutional networks for
scene parsing. In ICML Workshop on Deep Learning, 2009.
[9] Magnus Erik Hvass Pedersen. Convolutional neural network. Re-
trived from the World Wide Web: https://github.com/Hvass-Labs/
TensorFlow-Tutorials/blob/master/02_Convolutional_Neural_Network.
ipynb.
[10] Chi-Keung Tang Heung-Yeung Shum Yin Li, Jian Sun. Lazy snapping. ACM
SIGGRAPH 2004, 2004.
[11] Ani1 K. Jain Marie-Pierre Dubuisson. A modied hausdor distance for object
matching. Proceedings of 12th International Conference on Pattern Recognition,
1994.
[12] Pat Hanrahan Robert A. Drebin, Loren Carpenter. Volume rendering. ACM
SIGGRAPH Computer Graphics, 1988.
[13] http://www.iplab.tcu.edu.tw/data/CT/CT_pr.htm.