跳到主要內容

臺灣博碩士論文加值系統

(44.200.86.95) 您好!臺灣時間:2024/05/20 08:11
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:許雅筌
研究生(外文):HSU, YA-CHUAN
論文名稱:基於深度學習實現瘧疾細胞偵測、致病原與感染階段分類方法與分析
論文名稱(外文):The Analysis of Deep Learning-Based Malaria Cell Detection, Pathogen, and Life-Cycle Classification Method
指導教授:陳彥霖陳彥霖引用關係
指導教授(外文):CHEN, YEN-LIN
口試委員:陳彥霖范育成蔣欣翰黃志勝
口試委員(外文):CHEN, YEN-LINFAN, YU-CHENGCHIANG, HSIN-HANHUANG, CHIH-SHENG
口試日期:2022-07-26
學位類別:碩士
校院名稱:國立臺北科技大學
系所名稱:資訊工程系
學門:工程學門
學類:電資工程學類
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:61
中文關鍵詞:瘧疾細胞偵測生命週期分類醫學影像
外文關鍵詞:MalariaCell detectionLife-Cycle classificationMedical image
相關次數:
  • 被引用被引用:0
  • 點閱點閱:234
  • 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
瘧疾是一種經由蚊子傳播,其致病原為瘧原蟲的傳染病,雖然台灣已經被列為瘧疾根除區,每年仍有境外移入病例,依據WHO官方資料,全球在2020年仍有2億4千1百萬人感染瘧疾,且估計約62萬7千人死亡,以非洲區域最為嚴重,瘧疾的標準診斷是由醫事人員目視病患的血液抹片辨識瘧原蟲品種及其感染階段,在缺乏醫療資源的區域,難以在短時間內診斷,因此藉由基於深度學習的影像辨識方法,能夠降低醫療成本,把握黃金治療時間。
公開瘧疾資料集取得有限,其中細胞自然分布具類別不平衡特性,同時血液抹片普遍有低對比度、染色差異及明暗不一致的差異性問題,導致辨識效果不穩定,為了解決上述問題,本論文針對瘧疾細胞特徵及致病原偵測與分類方法分析,並提出資料增強方法及改良的網路架構,訓練具有辨識多種致病原能力的模型,突破相關文獻僅針對單一瘧原蟲感染階段辨識功能,並應用於本文的兩階段辨識系統架構,分別進行細胞偵測與分類,進而提升系統準確度。
Malaria is an infectious disease transmitted by mosquitoes whose causative agent is Plasmodium. Although Taiwan has been listed as a malaria eradication area, there are a few imported cases that were infected with malaria abroad and symptoms show in Taiwan. In 2020, there are still 241 million people over the world infected with malaria while about 627,000 deaths were estimated, and Africa, especially is the most serious region. The standard diagnosis of malaria relies on the visual inspection of medical staff. They identify the malaria species and life-cycle through microscopic blood smears from patients. Due to the lack of medical resources, it is difficult to diagnose the disease within a short time. Therefore, with the image recognition method based on deep learning, medical costs can be reduced and the golden treatment time can be grasped.
Currently, public malaria datasets are rare in which the natural distribution of cells is imbalanced. Moreover, blood smears normally have problems of low contrast, staining differences, and inconsistency between light and dark, which cause unstable recognition. To solve the above problems, we analyzed malaria cell characteristics, pathogen detection, and classification methods to propose a data augmentation methods and improved network architecture to train models with the ability to identify multiple pathogens, breaking through related literature only for a single Plasmodium infection. We designed a two-stage system applied to the proposed method, which performs cell detection and classification respectively. Thereby we improved the accuracy of the system.
摘 要 i
ABSTRACT ii
誌謝 iv
目錄 v
表目錄 viii
圖目錄 ix
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機與目的 3
1.3 論文架構 3
第二章 文獻回顧 5
2.1 深度學習 5
2.1.1 LeNet-5 5
2.1.2 AlexNet 6
2.1.3 VGG 7
2.1.4 ResNet 8
2.1.5 YOLOv4 10
2.2 資料增強方法 11
2.2.1 Mosaic 11
2.2.2 Copy-Paste 11
2.3 加強提取特徵方法 13
2.3.1 Feature Pyramid Networks 13
2.3.2 Attention mechanism 14
2.4 瘧疾辨識相關研究 15
2.4.1 感染細胞與非感染細胞分類 15
2.4.2 細胞感染階段分類 17
第三章 研究方法 20
3.1 資料擴增方法分析 20
3.1.1 Mosaic 20
3.1.2 Copy-Paste 21
3.1.3 資料擴增方法適用性分析 22
3.2 細胞偵測與致病原及感染階段分類方法分析 25
3.2.1 致病原品種 25
3.2.2 感染階段特徵 26
3.2.3 深度學習方法 30
3.3 系統架構 31
3.4 感染細胞偵測方法 32
3.4.1 感染細胞偵測 32
3.4.2 細胞框位置提取 33
3.5 致病原與感染階段分類方法 34
3.5.1 色彩等級變換 34
3.5.2 亮度等級變換 36
3.5.3 對比度特徵強化 37
3.5.4 網路架構改良 38
第四章 實驗結果與分析 41
4.1 實驗環境 41
4.2 資料集說明 42
4.3 實驗方法 45
4.4 系統實驗結果評估 47
4.4.1 單一瘧原蟲模型表現評估 48
4.4.2 兩種瘧原蟲模型表現評估 51
4.4.3 相關文獻比較實驗評估 54
第五章 結論與未來工作 57
5.1 結論 57
5.2 未來工作 57
參考文獻 58
[1]World malaria report 2021. World Health Organization, 2021.
[2]瘧疾預防及治療用藥指引, 4th ed. Taipei: 衛生福利部疾病管制署, 2019.
[3]Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, "Gradient-based learning applied to document recognition," Proceedings of the IEEE, vol. 86, no. 11, pp. 2278-2324, 1998.
[4]A. Krizhevsky, I. Sutskever, and G. E. Hinton, "Imagenet classification with deep convolutional neural networks," Advances in neural information processing systems, vol. 25, 2012.
[5]K. Simonyan and A. Zisserman, "Very deep convolutional networks for large-scale image recognition," arXiv preprint arXiv:1409.1556, 2014.
[6]K. He, X. Zhang, S. Ren, and J. Sun, "Deep residual learning for image recognition," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 770-778.
[7]J. Redmon and A. Farhadi, "YOLOv3: An incremental improvement," arXiv preprint arXiv:1804.02767, 2018.
[8]S. Liu, L. Qi, H. Qin, J. Shi, and J. Jia, "Path aggregation network for instance segmentation," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2018, pp. 8759-8768.
[9]A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao, "YOLOv4: Optimal speed and accuracy of object detection," arXiv preprint arXiv:2004.10934, 2020.
[10]G. Ghiasi et al., "Simple copy-paste is a strong data augmentation method for instance segmentation," in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 2918-2928.
[11]T.-Y. Lin, P. Dollár, R. Girshick, K. He, B. Hariharan, and S. Belongie, "Feature pyramid networks for object detection," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2017, pp. 2117-2125.
[12]S. Woo, J. Park, J.-Y. Lee, and I. S. Kweon, "CBAM: Convolutional block attention module," in Proceedings of the European conference on computer vision (ECCV), 2018, pp. 3-19.
[13]A. Rahman, H. Zunair, T. R. Reme, M. S. Rahman, and M. R. C. Mahdy, "A comparative analysis of deep learning architectures on high variation malaria parasite classification dataset," Tissue and Cell, vol. 69, p. 101473, 2021.
[14]T. R. Talukdar, M. J. Hossain, and T. H. Talukdar, "Malaria detection in Segmented Blood Cell using Convolutional Neural Networks and Canny Edge Detection," arXiv preprint arXiv:2202.10426, 2022
[15]E. Abdulghany and N. Osama, "Classification of Malaria Cell Images with Deep Learning Architectures," 2021, doi: 10.13140/RG.2.2.26387.40484.
[16]S. S. Abbas and T. M. Dijkstra, "Detection and stage classification of Plasmodium falciparum from images of Giemsa stained thin blood films using random forest classifiers," Diagnostic pathology, vol. 15, no. 1, pp. 1-11, 2020.
[17]R. R. Manku, A. Sharma, and A. Panchbhai, "Malaria Detection and Classificaiton," arXiv preprint arXiv:2011.14329, 2020.
[18]J. Hung and A. Carpenter, "Applying faster R-CNN for object detection on malaria images," in Proceedings of the IEEE conference on computer vision and pattern recognition workshops, 2017, pp. 56-61.
[19]S. Yun, D. Han, S. J. Oh, S. Chun, J. Choe, and Y. Yoo, "Cutmix: Regularization strategy to train strong classifiers with localizable features," in Proceedings of the IEEE/CVF international conference on computer vision, 2019, pp. 6023-6032.
[20]M. M. Hossain, M. A. Rahim, A. N. Bahar, and M. M. Rahman, "Automatic malaria disease detection from blood cell images using the variational quantum circuit," Informatics in Medicine Unlocked, vol. 26, p. 100743, 2021.
[21]S. Sato, "Plasmodium—a brief introduction to the parasites causing human malaria and their basic biology," Journal of physiological anthropology, vol. 40, no. 1, pp. 1-13, 2021.
[22]“Centers for Disease Control and Prevention,” July 16, 2020. Accessed on June 29, 2022. [Online]. Available: https://www.cdc.gov/malaria/about/biology/index.html
[23]M. Poostchi, K. Silamut, R. J. Maude, S. Jaeger, and G. Thoma, "Image analysis and machine learning for detecting malaria," Translational Research, vol. 194, pp. 36-55, 2018.
[24]C.-Y. Wang, A. Bochkovskiy, and H.-Y. M. Liao, "YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors," arXiv preprint arXiv:2207.02696, 2022.
[25]S. Ren, K. He, R. Girshick, and J. Sun, "Faster R-CNN: Towards real-time object detection with region proposal networks," Advances in neural information processing systems, vol. 28, 2015.
[26]V. Ljosa, K. L. Sokolnicki, and A. E. Carpenter, “P. vivax (malaria) infected human blood smears (bbbc041v1),” 2012.
[27]C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna, "Rethinking the inception architecture for computer vision," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 2818-2826.
[28]S. Rajaraman et al., "Pre-trained convolutional neural networks as feature extractors toward improved malaria parasite detection in thin blood smear images," PeerJ, vol. 6, p. e4568, 2018.
[29]M. Masud et al., "Leveraging deep learning techniques for malaria parasite detection using mobile application," Wireless Communications and Mobile Computing, vol. 2020, 2020.
[30]A. Vijayalakshmi and B. Rajesh Kanna, "Deep learning approach to detect malaria from microscopic images," Multimedia Tools and Applications, vol. 79, no. 21, pp. 15297-15317, 2020.
[31]Q. A. Arshad et al., "A dataset and benchmark for malaria life-cycle classification in thin blood smear images," Neural Computing and Applications, vol. 34, no. 6, pp. 4473-4485, 2022.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top