跳到主要內容

臺灣博碩士論文加值系統

(100.28.0.143) 您好!臺灣時間:2024/07/23 10:49
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:吳承峻
研究生(外文):WU, CHENG-CHUN
論文名稱:自動化偵測胸前放射線影像氣管分岔點
論文名稱(外文):Automated Detection of Carina in Chest X-ray Images
指導教授:許中川許中川引用關係
指導教授(外文):HSU, CHUNG-CHIAN
口試委員:許中川陳重臣王佳文
口試委員(外文):HSU, CHUNG-CHIANCHEN, CHUNG-CHENWANG,CHIA-WEN
口試日期:2022-05-30
學位類別:碩士
校院名稱:國立雲林科技大學
系所名稱:資訊管理系
學門:電算機學門
學類:電算機一般學類
論文種類:學術論文
論文出版年:2022
畢業學年度:110
語文別:中文
論文頁數:37
中文關鍵詞:胸前放射線影像氣管分岔點偵測電腦輔助診斷影像處理
外文關鍵詞:Chest X-ray ImageCarina DetectionComputer-Aided DiagnosisImage Processing
相關次數:
  • 被引用被引用:1
  • 點閱點閱:135
  • 評分評分:
  • 下載下載:14
  • 收藏至我的研究室書目清單書目收藏:0
在放射線胸腔影像上對氣管分岔點的定位確認,是重症照護病房進行檢查與氣管插管矯正的重要評估步驟。快速並且準確的評估正確位置至關重要,及時判斷錯位的支持,可以防止病患的發病率以及死亡率。若能有效地以自動化方式找出氣管內管分岔點位置不僅可以降低醫生的工作負擔,還可以輔助及加速醫護人員的診斷。本研究旨使用深度學習的方法,偵測氣管分岔點位置,提出了一個改進的模型Nested-U2Net,使其在氣管分岔點的分割任務上,有更好的績效表現,並且集成了多種模型,改善單一模型的缺點,實驗結果表明偵測氣管分岔點位置的整體平均誤差距離為0.25公分,偵測誤差落在0.5cm內的準確率為91%,偵測誤差落在1cm內的準確率為98%,證實了所提出的方法是可行的。
The localization and confirmation of the tracheal bifurcation point on the radiographic thoracic image is an important evaluation step for the inspection and tracheal intubation correction in the intensive care unit. Rapid and accurate assessment of the correct location is critical, and timely identification of dislocation supports can prevent morbidity and mortality in patients. If the location of the bifurcation point of the endotracheal tube can be found effectively in an automated way, it can not only reduce the workload of doctors, but also assist and speed up the diagnosis of medical staff. This study aims to use deep learning methods to detect the location of tracheal bifurcation points, propose an improved model Nested-U2Net, which has better performance in the segmentation task of tracheal bifurcation points, and integrate various the model improves the shortcomings of a single model. The experimental results show that the overall average error distance of detecting the position of the tracheal bifurcation point is 0.25 cm, the percentage of the detection error within 0.5 cm is 91%, and the percentage of the detection error within 1 cm is 98%. The result confirms that the proposed method is feasible.
摘要 I
ABSTRACT II
目錄 III
表目錄 IV
圖目錄 V
壹、 緒論 1
1.1 研究背景 1
1.2 研究動機 2
1.3 研究目的 2
貳、 文獻探討 4
2.1 氣管分岔點偵測 4
2.2 醫療影像處理 4
參、 方法 6
3.1 預處理 6
3.2 U-NET++ 7
3.3 U2-NET 8
3.4 NESTED-U2NET 9
3.5 損失函數 13
3.6 K-FOLD CROSS VALIDATION 14
3.7 集成結果 14
3.8 評估標準 16
肆、 實驗 17
4.1 資料集 17
4.2 後處理 17
4.3 實驗結果 18
伍、 結論 28
參考文獻 29

1.Badrinarayanan Vijay , Kendall Alex , Cipolla Roberto , (2016). SegNet: A Deep
Convolutional Encoder-Decoder Architecture for Image Segmentation, Computer Vision and Pattern, https://doi.org/10.48550/arXiv.1511.00561
2.Chen, S., Zhang, M., Yao, L., & Xu, W. (2016). Endotracheal tubes positioning detection in adult portable chest radiography for intensive care unit. International journal of computer assisted radiology and surgery, 11(11), 2049-2057.
3.Chen L. , Papandreou G. , Schroff F. , Adam H. (2017) Rethinking Atrous
Convolution for Semantic Image Segmentation, Computer Vision and Pattern Recognition, arXiv:1706.05587
4.Chen L. , Zhu Y. , Papandreou G. , S., Adam H. (2018) Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation, Computer Vision and Pattern Recognition, https://doi.org/10.48550/arXiv.1802.02611
5.Goodman, L. R., Conrardy, P. A., Laing, F., & Singer, M. M. (1976). Radiographic evaluation of endotracheal tube position. American Journal of Roentgenology, 127(3), 433-434.
6.Godoy, M. C., Leitman, B. S., De Groot, P. M., Vlahos, I., & Naidich, D. P. (2012). Chest radiography in the ICU: Part 1, Evaluation of airway, enteric, and pleural tubes. American Journal of Roentgenology, 198(3), 563-571.
7.Huang, M., Ting, Z., and Guo, S. (2013) "Carina detection on chest radiography," International Conference on Orange Technologies (ICOT), pp. 127-130, doi: 10.1109/ICOT.2013.6521174.
8.Koonsanit, K., Thongvigitmanee, S., Pongnapang, N., & Thajchayapong, P. (2017). Image enhancement on digital x-ray images using n-clahe. Paper presented at the 2017 10th Biomedical Engineering International Conference (BMEiCON).
9.Kara, S., Akers, J.Y. & Chang, P.D. (2021). Identification and Localization of Endotracheal Tube on Chest Radiographs Using a Cascaded Convolutional Neural Network Approach. J Digit Imaging 34, 898–904
10.Li Hanchao , Xiong Pengfei , An Jie , Wang Lingxue (2018). Pyramid Attention
Network for Semantic Segmentation, Computer Vision and Pattern
Recognition, https://doi.org/10.48550/arXiv.1805.10180
11.Opitz, D., & Maclin, R. (1999). Popular ensemble methods: An empirical study. Journal of artificial intelligence research, 11, 169-198.
12.Qin Xuebin, Zhang Zichen, Huang Chenyang, Dehghan Masood, Zaiane Osmar R., Jagersand Martin,(2020).U2-Net: Going deeper with nested U-structure for salient object detection,Pattern Recognition,Volume 106,2020,107404,ISSN 0031-3203,https://doi.org/10.1016/j.patcog.2020.107404.
13.Ronneberger, O., Fischer, P., T. (2015) U-net: Convolutional networks for
biomedical image segmentation. International Conference on Medical image computing and computer-assisted intervention, Springer , pp. 234-241
14.Yu D, Wang H, Chen P, Wei Z, (2014).Mixed pooling for convolutional neural
networks, in Rough Sets Knowl. Technol., Cham, Switzerland: Springer, pp. 364-375, 2014.
15.Wen, H., Qi, W., & Shuang, L. (2016). Medical X-ray image enhancement based on wavelet domain homomorphic filtering and CLAHE. Paper presented at the 2016 International Conference on Robots & Intelligent System (ICRIS).
16.Zuiderveld, K. (1994). Contrast limited adaptive histogram equalization. Graphics gems, 474-485.
17.Zhou, Z. H., Wu, J., & Tang, W. (2002). Ensembling neural networks: many could be better than all. Artificial intelligence, 137(1-2), 239-263.
18.Zhou, Z., Siddiquee, M.M.R., Tajbakhsh, N., Liang, J. (2018) Unet++: A nested u-net architecture for medical image segmentationDeep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, Springer (2018), pp. 3-11

QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top