博碩士論文 105522609 詳細資訊




以作者查詢圖書館館藏 以作者查詢臺灣博碩士 以作者查詢全國書目 勘誤回報 、線上人數:28 、訪客IP:3.15.202.169
姓名 歐莉亞(Nurotul Auliya′)  查詢紙本館藏   畢業系所 資訊工程學系
論文名稱 骨折區檢測
(Fracture Region Detection)
相關論文
★ Single and Multi-Label Environmental Sound Recognition with Gaussian Process★ 波束形成與音訊前處理之嵌入式系統實現
★ 語音合成及語者轉換之應用與設計★ 基於語意之輿情分析系統
★ 高品質口述系統之設計與應用★ 深度學習及加速強健特徵之CT影像跟骨骨折辨識及偵測
★ 基於風格向量空間之個性化協同過濾服裝推薦系統★ RetinaNet應用於人臉偵測
★ 金融商品走勢預測★ 整合深度學習方法預測年齡以及衰老基因之研究
★ 漢語之端到端語音合成研究★ 基於 ARM 架構上的 ORB-SLAM2 的應用與改進
★ 基於深度學習之指數股票型基金趨勢預測★ 探討財經新聞與金融趨勢的相關性
★ 基於卷積神經網路的情緒語音分析★ 運用深度學習方法預測阿茲海默症惡化與腦中風手術存活
檔案 [Endnote RIS 格式]    [Bibtex 格式]    [相關文章]   [文章引用]   [完整記錄]   [館藏目錄]   至系統瀏覽論文 ( 永不開放)
摘要(中) 跟骨是人體足中最大骨,最重要的骨骼。它的作用是保持人體重量和其施加的力量。然而,這種骨骼也是最常受傷的骨。在大多數情況下,發生損傷由於過度的軸向載荷。
跟骨骨折分類用於將受損的跟骨骨分類為每種類型需要不同類型的治療。在幾種不同的分類系統中,Sanders分類是最常用的,因為它基於跟骨骨折的冠狀計算機斷層掃描(CT),顯示跟骨最寬的下表面。然而,儘管其受歡迎程度,但其解釋存在高度的可變性和不一致性,其用戶之間僅具有公平到中等的一致性,這主要是由於經驗的差異造成的。這種不一致是重要的,因為治療中的錯誤可能對患者造成不可挽回的傷害。
目前,醫生手動檢查所選擇的CT掃描圖像,其中包含最完整的跟骨結構,然後根據跟骨骨折分類對其進行分類。這容易出錯,不僅是因為上述問題,還因為有時CT掃描圖像模糊並且沒有顯示清晰的斷裂線。
本論文試圖通過檢測跟骨碎片並顯示骨折線來幫助識別跟骨骨折類型。第一部分是將每個骨碎片與圖像中的其他骨碎片區分開來。第二部分是指出跟骨碎片並顯示其邊緣。
摘要(英) Calcaneus bone is the biggest and the most important bone in human foot. Its role is to hold human weight and other forces that they applied to it. However, this bone is also the most frequently injured tarsal bone. In most cases, the injuries happen because of excessive axial load such as a fall from height or in a motor vehicle accident.
Calcaneus bone fracture classification is used to classify the injured calcaneus bone into types with each type needing a different kind of treatment. Among the several different classification systems, Sanders classification is the most commonly used because it is based on the coronal computed tomography (CT) scan of the calcaneal fracture that shows the widest undersurface of the posterior facet of the calcaneus. However, in spite of its popularity, there is a high degree of variability and inconsistency in its interpretation with only fair to moderate consistency among its users, which is mostly caused by the difference in experience. Such inconsistency is significant because a mistake in treatment may cause irreparable harm to the patient.
Currently, doctors manually examine the chosen CT scan image, which contains the most complete calcaneus structure, then classify it according to calcaneus fracture classification. This is prone to mistakes, not only because of the aforementioned problem but also because sometimes the CT scan image is blurred and does not show clear fracture line.
This thesis attempts to help in identifying the calcaneus fracture type by detecting calcaneus bone fragments and showing the fracture lines.
The first part is to differentiate each bone fragment from the other bone fragments in the image. The second part is to point out the calcaneus bone fragments and show their edges. This system successfully automatically determine which fragments belong to calcaneus bone.
關鍵字(中) ★ CT scan images
★ calcaneus bone fracture
★ detection
★ segmentation
關鍵字(英) ★ CT scan images
★ calcaneus bone fracture
★ detection
★ segmentation
論文目次 摘要 i
Abstract ii
Acknowledgments iii
Table of Contents iv
List of Figures vi
List of Tables vii
Chapter 1 Introduction 1
Chapter 2 Literature Study 3
2.1 DICOM Images 3
2.2 Hounsfield Unit 3
2.3 CT Imaging on Calcaneus Bone and Sanders Classification 4
Chapter 3 Methodology and Experiment Setup 7
3.1 Methodology 7
3.1.1 Preprocessing: Threshold Value Decision 7
3.1.2 Obtaining Main Tissue 8
3.1.3 Obtaining Bone Fragments 9
3.1.4 Reassembling Bone Fragments According to Bone 10
3.1.5 Pointing Calcaneus Bone Fragments 10
3.2 Experiment Setup 10
Chapter 4 Experiment and Discussion 13
4.1 Pre-Processing: Threshold Value Decision 13
4.2 Obtaining Main Tissue 14
4.3 Obtaining Bone Fragments 14
4.4 Reassembling Bone Fragments According to Bone 16
4.5 Pointing Calcaneus Bone Fragments and Showing Its Fracture Line 17
4.6 Accuracy Check 18
Chapter 5 Conclusion 20
5.1 Conclusion 20
5.2 Suggestion for Future Works 21
References 22
Appendix A: Details of The number of available cases 23
Appendix B: Qualified Input Images’ Accuracy 26
Appendix C: Original Images 30
Appendix D: Groundtruth Images 44
Appendix E: Detected ‘Calcaneus Bone’ 58
參考文獻 [1] A. Daftary, A. H. Haims and M. R. Baumgaertner, Fractures of the Calcaneus: A review with Emphasis on CT, RadioGraphics, 2005.
[2] R. Soeur, R. Remy, B. and B. , "Fractures of the Calcaneus with Displacement of the Thalamic Portion," Bone & Joint Journal, vol. 57, no. 4, pp. 413-421, 1975.
[3] "Diagnosis: Calcaneus, intraarticular displaced fractures," [Online]. Available: https://www2.aofoundation.org/wps/portal/!ut/p/a0/04_Sj9CPykssy0xPLMnMz0vMAfGjzOKN_A0M3D2DDbz9_UMMDRyDXQ3dw9wMDAx8jfULsh0VAdAsNSU!/?bone=Foot&segment=Calcaneus&soloState=fbox&teaserTitle=&contentUrl=srg/82/01-Diagnosis/08-Diag-Calcaneus_displaced.jsp. [Accessed 21 12 2017].
[4] R. Bhattacharya, U. T. Vassan, P. Finn and A. Port, "Sanders Classification of Fractures of the Os Calsis," The Journal of Bone & Joint Surgery (Br), Vols. 87-B, no. 2, p. 205, 2005.
[5] D. W. Fanning, "Converting CT Data to Hounsfield Units," 10 01 2006. [Online]. Available: http://www.idlcoyote.com/fileio_tips/hounsfield.html. [Accessed 12 2018].
[6] "Wikipedia: Hounsfield scale," [Online]. Available: https://en.wikipedia.org/wiki/Hounsfield_scale. [Accessed 12 2018].
[7] A. T. Davis, P. L. Antony, S. Pani and A. Nisbet, "Assessment of the variation in CT scanner performance (image quality and Hounsfield units) with scan parameters, for image optimisation in radiotherapy treatment planning," Physica Medica, vol. 45, pp. 59-64, 2018.
[8] A. Murphy, M. Skalski and et al., "Sanders CT classification of calcaneal fracture," Radiopaedia, [Online]. Available: https://radiopaedia.org/articles/sanders-ct-classification-of-calcaneal-fracture-2 . [Accessed 21 12 2017].
指導教授 王家慶(Jia-Ching Wang) 審核日期 2018-8-23
推文 facebook   plurk   twitter   funp   google   live   udn   HD   myshare   reddit   netvibes   friend   youpush   delicious   baidu   
網路書籤 Google bookmarks   del.icio.us   hemidemi   myshare   

若有論文相關問題,請聯絡國立中央大學圖書館推廣服務組 TEL:(03)422-7151轉57407,或E-mail聯絡  - 隱私權政策聲明