簡易檢索 / 詳目顯示

回結果列表
研究生： 林均澤
Lin, Chin-Tse
論文名稱： 應用於細菌的抗生素藥敏試驗之深度學習顯微影像系統研究
Deep Learning-based Microscopy Imaging System for Antimicrobial Susceptibility Test
指導教授： 張憲彰
Chang, Hsien-Chang
學位類別： 碩士
Master
系所名稱： 工學院 - 生物醫學工程學系
Department of BioMedical Engineering
論文出版年： 2020
畢業學年度： 108
語文別： 英文
論文頁數： 52
中文關鍵詞： 抗生素藥敏試驗 、深度學習 、機器學習 、細菌檢測 、顯微影像
外文關鍵詞： Antimicrobial susceptibility testing (AST), Deep learning, Machine learning, Bacteria detection, Microscopy imaging
相關次數： 點閱：108  下載：1 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 當前的抗生素藥敏試驗通常需要三、四天的檢測時間，並且仰賴大量的人工操作，冗長的檢測時間將會延遲細菌感染在早期的治療決策，嚴重的細菌感染很可能導致敗血症，而敗血症的死亡率，當治療時間每延遲一小時，死亡率便會增加7.6%。因此現今迫切地需要一種更快速的抗生素藥敏試驗解決方法，來提供醫生能更有效地開定精確的抗生素處方。可從先前的研究發現，細菌在抗生素的作用下會產生型態的變化，從而可被應用於判定藥敏試驗的結果。本研究根據此特性提出了一種應用於革蘭氏陰性細菌的抗生素藥物耐受性試驗之深度學習顯微影像系統，我們利用深度學習的演算法，辨識出在抗生素作用下不同細菌型態的變化，並使用機器學習的方法進行迴歸分析，檢討是否可測定出在不同濃度抗生素作用下之針對該細菌的最小抑菌濃度 (MIC)。此系統實際上利用標準菌株大腸桿菌 (ATCC 25922)，在頭孢唑林、頭孢他啶、頭孢吡肟等三種不同代數之頭孢菌素的抗生素濃度組合下進行初步驗證，我們成功於2小時的培養後取得且判讀出不同抗生素濃度的影像，進而判定得知其MIC值，準確度均可以達到95%。本研究將顯微影像系統與軟體整合，與傳統的方法比較，通過簡單的手動操作程序，將試驗整體的時間從3-4天，縮短為4-5小時，並減少了人力成本。

    The current antimicrobial susceptibility testing (AST) method usually takes a few days and labor-intensive, which will delay initial treatment decisions in the early stages of bacterial infections. The severe bacterial infection is very likely to cause sepsis, and the mortality rate of sepsis will increase by 7.6% per hour when effective treatments are delayed. It shows that there is an urgent need for a rapid solution to help doctors prescribe antibiotics more effectively. Previous research shows that that the morphology of bacteria will change under beta-lactam antibiotics treatment. From the characteristic of bacteria, we propose a rapid AST system for Gram-negative bacteria, using deep learning algorithms to identify different bacterial morphology changes. Moreover, using a machine learning method to do the classification automatically determines the minimum inhibitory concentration (MIC) bacteria at different concentration of antibiotics. The system uses the Escherichia coli (ATCC 25922) to conduct preliminary study of cefazolin, ceftazidime and cefepime treatments. The MIC can be determined after 2 hours. The accuracy can reach 95%. Our system integrates the microscopic imaging system and software, which can reduce the detection time from three to four days to 4-5 hours with simple manual operations procedures.

    Abstract I 中文摘要 II 致謝 III Contents IV List of Figures VII List of Tables IX Chapter 1 Introduction 1 1.1 Background 1 1.2 Antimicrobial susceptibility testing 2 1.2.1 Minimal inhibitory concentration (MIC) 2 1.2.2 Traditional AST methods 3 1.2.3 Current AST methods 5 1.2.4 Review of AST development 6 1.3 Mechanism of antibiotics 7 1.3.1 Types and actions of different antibiotics 7 1.3.2 Evolution of Cephalosporin 9 1.3.3 Bacterial morphology under antibiotics 10 1.4 Introduction of deep learning 11 1.4.1 Novel deep learning models 11 1.4.2 Deep learning in the medical field 12 1.4.3 Image processing method for bacteria image 13 1.5 Motivation and Aims 14 Chapter 2 Materials and Methods 16 2.1 Image system design 16 2.1.1 Image parameters 16 2.1.2 Introduction of the image system 17 2.1.3 User interface of the image capture 18 2.1.4 Data collection 19 2.2 AST interpretation 20 2.2.1 Bacteria culture 20 2.2.2 Standard bacteria and their MIC 20 2.3 Culture wells 21 2.3.1 Volume test 21 2.3.2 Chip structure 22 2.4 Data preprocessing 23 2.4.1 Image preprocessing 23 2.4.2 Image labeling 23 2.5 Convolutional neural network 24 2.5.1 Convolution layer 24 2.5.2 Pooling layer 25 2.6 Object detection 26 2.7 YOLO (You only look once) 27 2.7.1 Development of YOLO 27 2.7.2 Introduction of YOLOv3 29 2.7.3 Structure of YOLOv3 30 Chapter 3 Results and Discussion 31 3.1 Evaluating object detection models 31 3.1.1 Intersection over union (IoU) 31 3.1.2 Mean average precision (mAP) 32 3.1.3 K-fold cross-validation 34 3.1.4 Performance of different input sizes 37 3.1.5 Data augmentation 38 3.2 Results of AST report 40 3.2.1 Statistics of bounding box 40 3.2.2 Logistic regression 44 3.2.3 AST results from ML model 45 3.2.4 Results from different β-lactam antibiotic 46 Chapter 4 Conclusion and Prospects 48 Reference 49 Personal Information 52

    1. Kumar, A., et al., Duration of Hypotension Before Initiation of Effective Antimicrobial Therapy Is the Critical Determinant of Survival in Human Septic Shock. Critical Care Medicine, 2006. 34(6): p. 1589-1596.
    2. Sandle, T., Pharmaceutical Microbiology, Pharmaceutical Microbiology, Woodhead Publishing: Oxford, 2016. p. 171-183.
    3. Andrews, J.M., Determination of Minimum Inhibitory Concentrations. Journal of Antimicrobial Chemotherapy, 2001. 48(suppl_1): p. 5-16.
    4. Garcia-Garrote, F., et al., Evaluation of a New System, VITEK 2, for Identification and Antimicrobial Susceptibility Testing of Enterococci. Journal of Clinical Microbiology 2000. 38(6): p. 2108-2111.
    5. Horstkotte, M.A., et al., Evaluation of the BD PHOENIX Automated Microbiology System for Detection of Methicillin Resistance in Coagulase-Negative Staphylococci. Journal of Clinical Microbiology, 2004. 42(11): p. 5041-5046.
    6. Jung, J.S., et al., Rapid Detection of Antibiotic Resistance Based on Mass Spectrometry and Stable Isotopes. European Journal of Clinical Microbiology & Infectious Diseases, 2014. 33(6): p. 949-955.
    7. Fluit, A.C., et al., Molecular Detection of Antimicrobial Resistance. Clinical Microbiology Reviews, 2001. 14(4): p. 836-871.
    8. Choi, J., et al., A Rapid Antimicrobial Susceptibility Test Based on Single-cell Morphological Analysis. Science Translational Medicine, 2014. 6(267): p. 267ra174.
    9. Kong, K.F., et al., Beta-lactam Antibiotics: From Antibiosis to Resistance and Bacteriology. Acta Pathologica et Microbiologica Scandinavica, 2010. 118(1): p. 1-36.
    10. Mingeot-Leclercq, M.P., et al., Aminoglycosides: Activity and Resistance. Antimicrobial Agents and Chemotherapy, 1999. 43(4): p. 727-37.
    11. Wong, T., et al., Primary Syphilis: Serological Treatment Response to Doxycycline/Tetracycline Versus Benzathine Penicillin. The American Journal of Medicine, 2008. 121(10): p. 903-908.
    12. Mehlhorn, A. and Brown, D., Safety Concerns with Fluoroquinolones. The Annals of Pharmacotherapy, 2007. 41: p. 1859-1866.
    13. Van Teeseling, M.C.F., et al., Determinants of Bacterial Morphology: From Fundamentals to Possibilities for Antimicrobial Targeting. Frontiers in Microbiology, 2017. 8: p. 1264.
    14. Ribeiro da Cunha, B., et al., Antibiotic Discovery: Where Have We Come From, Where Do We Go? Antibiotics (Basel), 2019. 8(2): p. 45.
    15. Yao, Z., et al., Distinct Single-cell Morphological Dynamics Under Beta-Lactam Antibiotics. Molecular Cell 2012. 48(5): p. 705-712.
    16. Lecun, Y., et al., Gradient-based Learning Applied to Document Recognition. Proceedings of the Institute of Electrical and Electronics Engineers, 1998. 86(11): p. 2278-2324.
    17. Krizhevsky, A., et al., ImageNet Classification with Deep Convolutional Neural Networks. Communications of the Association for Computing Machinery, 2017. 60(6): p. 84-90.
    18. Darmawan, J.B.B. and Mungkasi, S., Performance of Parallel Computation Using CUDA for Solving the One-dimensional Elasticity Equations. Journal of Physics: Conference Series, 2017. 801: p. 012080.
    19. Simonyan, K. and Zisserman, A., Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv e-prints, 2014. arXiv:1409.1556.
    20. He, K., et al., Deep Residual Learning for Image Recognition. arXiv e-prints, 2015. arXiv:1512.03385.
    21. Currie, G., et al., Machine Learning and Deep Learning in Medical Imaging: Intelligent Imaging. Journal of Medical Imaging and Radiation, 2019. 50(4): p. 477-487.
    22. Davenport, T. and Kalakota, R., The Potential for Artificial Intelligence in Healthcare. Future Healthcare Journal, 2019. 6(2): p. 94-98.
    23. Pan, Y., et al., Bacterial Foraging Based Edge Detection for Cell Image Segmentation. in 2015 Annual International Conference of the Institute of Electrical and Electronics Engineers Engineering in Medicine and Biology Society. 2015.
    24. Yi, Q., et al., Direct Antimicrobial Susceptibility Testing of Bloodstream Infection on SlipChip. Biosensors and Bioelectronics, 2019. 135: p. 200-207.
    25. Smith, K.P., et al., Automated Interpretation of Blood Culture Gram Stains by Use of a Deep Convolutional Neural Network. Journal of Clinical Microbiology, 2018. 56(3): p. e01521-17.
    26. Yu, H., et al., Phenotypic Antimicrobial Susceptibility Testing with Deep Learning Video Microscopy. Analytical Chemistry, 2018. 90(10): p. 6314-6322.
    27. Tsai, H.F., et al., Usiigaci: Instance-aware Cell Tracking in Stain-free Phase Contrast Microscopy Enabled by Machine Learning. Softwarex, 2019. 9: p. 230-237.
    28. CLSI. In: Performance Standards for Antimicrobial Susceptibility Testing. 28th ed. Wayne, PA: Clinical and Laboratory Standards Institute; 2018. CLSI supplement M100.
    29. Scherer, D., et al., Evaluation of Pooling Operations in Convolutional Architectures for Object Recognition. in Internet Corporation for Assigned Names and Numbers. 2010.
    30. Redmon, J., et al., You Only Look Once: Unified, Real-Time Object Detection. in 2016 Institute of Electrical and Electronics Engineers Conference on Computer Vision and Pattern Recognition (CVPR). 2016
    31. Redmon, J. and Farhadi, A., YOLO9000: Better, Faster, Stronger. arXiv e-prints, 2016. arXiv:1612.08242.
    32. Redmon, J. and Farhadi, A., YOLOv3: An Incremental Improvement. arXiv e-prints, 2018. arXiv:1804.02767.

    下載圖示 校內：2025-07-27公開
    校外：2025-07-27公開
    QR CODE