評估深度卷積神經網路用於檢測和分割Chest X-ray圖像中的鎖骨骨折

簡易檢索 / 詳目顯示

回結果列表

研究生：	劉正邦 Zhang-Bang Liu
論文名稱：	評估深度卷積神經網路用於檢測和分割Chest X-ray圖像中的鎖骨骨折 Evaluating Deep Convolutional Neural Networks for Detection and Segmentation of Fractured Clavicle in Chest X-ray Images
指導教授：	黃輝揚 Adam Huang
口試委員:
學位類別：	碩士 Master
系所名稱：	生醫理工學院 - 生物醫學工程研究所 Graduate Institute of Biomedical Engineering
論文出版年：	2021
畢業學年度：	109
語文別：	英文
論文頁數：	64
中文關鍵詞：	深度學習、實例分割、自動診斷、胸腔X光、鎖骨骨折
外文關鍵詞：	Automatic Diagnosis, Chest X-ray, Clavicle Fracture
相關次數：	點閱：17 下載：0
分享至:	分享至facebook 分享至twitter

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

鎖骨骨折在所有成人骨折中極為常見，在一些緊急的情況下，如果沒有及時診斷和治療，將可能導致氣胸或出血等其他嚴重的併發症。所以能夠在情況較危急的急診室全天候正確有效的識別chest x-ray中危及生命的病例非常重要。然而，較為偏鄉的醫院通常沒有足夠的放射科醫生來全天候的工作以滿足這種即時的需求。另一方面，隨著人工智能(AI)的進步以及高端GPU(AI的關鍵創新)的成本下降，深度學習卷積神經網路的性能變得更加準確，並且目前常應用於各種汽車商業的用途。在這裡，我們假設這種基於GPU的新型AI技術可以轉移到醫學影像分析領域，為偏鄉醫院提供高精度的急性鎖骨骨折自動診斷服務。這項研究的數據集使用了4953張胸腔x-ray影像，其中左側及右側的鎖骨總共有9906根，這些數據集皆經由放射學專家評估和標記各個邊界框和骨折的種類(5287根正常的鎖骨、1068根急性的骨折、1656根舊傷的骨折和 1895根由骨釘固定的鎖骨)。全部的鎖骨圖像被分成6934(70%)張用於訓練 Mask R-CNN 模型和 2972張(30%)用於測試。最後測試的結果，急性病例檢測的召回率為0.781(所有急性病例共有324個案例，而預測為急性病例有253個案例)，檢測的精確度為0.821(預測為急性病例有308個，而有253個案例為急性病例)，最後F1的分數為0.801。

Clavicle fractures are extremely common in adult fractures and some emergency cases may cause other serious complications such as pneumothorax or hemorrhage if not diagnosed and treated timely. It is very important to be able to identify the life threatening cases in chest x-ray films correctly and efficiently around the clock in the emergency department. However, the rural hospitals generally do not have enough staff radiologists to work around the clock to meet this timely demand. On the other hand, with the advances in artificial intelligence (AI) and the cost drop of high end GPU (the key AI innovation), the performance of deep learning convolutional neural networks has become accurate enough and readily available for commercial use in various automobile applications. Here, we hypothesize that this novel GPU-based AI technology can be transferred to the field of medical image analysis for providing high accuracy services in rural hospitals for diagnosing acute clavicle fractures automatically. This study included a dataset of 4953 chest x-ray images with 9906 (left and right) clavicles evaluated and annotated by radiology specialists with bounding boxes and fractural types (5287 normal clavicles, 1068 acute fractures, 1656 chronic fractures, and 1895 fixations). The clavicle images were split into 6934 (70%) for training a Mask R-CNN model and 2972 (30%) for testing. The recall, precision, and F1-score of detection of acute cases in the test results were 0.781 (253 true positive acute calls over a total of 324 true positive acute cases), 0.821 (253 true positive acute calls over a total of 308 positive acute cases), and 0.801 respectively.

中文摘要    i
Abstract    ii
Table of contents    iv
List of tables    v
List of figures    vi
Chapter 1 Introduction    1
Chapter2 Related Work    5
Section 2.1 Semantic Segmentation    5
Section 2.2 Computer-aided Diagnosis (CAD)    6
Section 2.3 Use deep learning for bone fracture classification    6
Section 2.4 Localization of lesions    7
Chapter3 Data and Annotations    9
Section 3.1 Chest x-ray dataset    9
Section 3.2 Data pre-processing    10
Section 3.3 Clavicle segmentation labeling    13
Section 3.4 Clavicle fracture Annotation    14
Chapter4 Mask R-CNN Architecture    16
Section 4.1 Backbone Network of Mask R-CNN    18
Section 4.2 Region Proposal Network    21
Section 4.3 ROI Heads    25
Chapter5 Mask-RCNN Implementation and experiment design    29
Section 5.1 Mask-RCNN Implementation    29
Section 5.2 Experiment Design    29
Section 5.2.1 Mask RCNN for Clavicle Segmentation (Exp. I)    29
Section 5.2.2 Mask R-CNN for segmenting both normal and fractural clavicles(Exp. II)    31
Section 5.2.3 Mask R-CNN for segmenting acute and chronic fractures(Exp. III)    32
Chapter6 Results    33
Chapter 7 Discussion    43
Chapter8 Conclusion    49


                                

[1] Postacchini F, Gumina S, De Santis P, Albo F (2002) Epidemiology of clavicle fractures. Journal of Shoulder and Elbow Surgery 11(5): 452- 456.
[2] Nordqvist A, Petersson C (1994) The incidence of fractures of the clavicle. Clinical orthopaedics and related research (300): 127-132.
[3] Joshua David Harris (2012) Acromioclavicular Joint and Clavicle Injuries.
[4] Holder J, Kolla S, Lehto S (2017) Clavicle Fractures: Allman and Neer Classification. J Adv Radiol Med Image 2(1): 102. doi: 10.15744/2456-5504.2.102.
[5] Cooney DR, Kloss B. Case report: Delayed subclavian vein injury secondary to clavicular malunion. J Emerg Med. 2012 Oct;43(4):648-50. doi: 10.1016/j.jemermed.2010.05.022. Epub 2010 Aug 13. PMID: 20708366.
[6] LeCun, Y., Bengio, Y. & Hinton, G. Deep learning. Nature 521, 436–444 (2015). https://doi.org/10.1038/nature14539
[7] C. Carson, S. Belongie, H. Greenspan and J. Malik, "Blobworld: image segmentation using expectation-maximization and its application to image querying," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, no. 8, pp. 1026-1038, Aug. 2002, doi: 10.1109/TPAMI.2002.1023800.
[8] Minaee S, Boykov Y, Porikli Fatih, et al. Image Segmentation Using Deep Learning: A Survey. arXiv:2001.05566, 2020.
[9] Irem Ulku, Erdem Akagunduz. A Survey on Deep Learning-based Architectures for Semantic Segmentation on 2D images. arXiv:1912.10230 [cs.CV]
[10] Fully Convolutional Networks for Semantic Segmentation. Jonathan Long, Evan Shelhamer, Trevor Darrell; Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015, pp. 3431-3440.
[11] A guide to convolution arithmetic for deep learning. Vincent Dumoulin, Francesco Visin. arXiv:1603.07285v2
[12] U-Net: Convolutional Networks for Biomedical Image Segmentation. Olaf Ronneberger, Philipp Fischer, Thomas Brox. arXiv:1505.04597v1
[13] Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation. Liang-Chieh Chen, Yukun Zhu, George Papandreou, Florian Schroff, Hartwig Adam. arXiv:1802.02611v3
[14] Mask R-CNN. Kaiming He, Georgia Gkioxari, Piotr Dollár, Ross Girshick. arXiv:1703.06870v3
[15] Shiraishi, Junji & Li, Qiang & Appelbaum, Daniel & Doi, Kunio. (2011). Computer-Aided Diagnosis and Artificial Intelligence in Clinical Imaging. Seminars in nuclear medicine. 41. 449-62. 10.1053/j.semnuclmed.2011.06.004.
[16] Teare, P., Fishman, M., Benzaquen, O. et al. Malignancy Detection on Mammography Using Dual Deep Convolutional Neural Networks and Genetically Discovered False Color Input Enhancement. J Digit Imaging 30, 499–505 (2017). https://doi.org/10.1007/s10278-017-9993-2
[17] Lakhani P, Sundaram B. Deep Learning at Chest Radiography: Automated Classification of Pulmonary Tuberculosis by Using Convolutional Neural Networks. Radiology. 2017 Aug;284(2):574-582. doi: 10.1148/radiol.2017162326. Epub 2017 Apr 24. PMID: 28436741.
[18] Gulshan V, Peng L, Coram M, Stumpe MC, Wu D, Narayanaswamy A, Venugopalan S, Widner K, Madams T, Cuadros J, Kim R, Raman R, Nelson PC, Mega JL, Webster DR. Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs. JAMA. 2016 Dec 13;316(22):2402-2410. doi: 10.1001/jama.2016.17216. PMID: 27898976.
[19] Alex Krizhevsky, Ilya Sutskever, and Geoffrey E. Hinton. 2017. ImageNet classification with deep convolutional neural networks. Commun. ACM 60, 6 (June 2017), 84–90. DOI:https://doi.org/10.1145/3065386
[20] Very Deep Convolutional Networks for Large-Scale Image Recognition. Karen Simonyan, Andrew Zisserman. arXiv:1409.1556 [cs.CV].
[21] Deep Residual Learning for Image Recognition. Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun. arXiv:1512.03385 [cs.CV]
[22] Seok Won Chung, Seung Seog Han, Ji Whan Lee, Kyung-Soo Oh, Na Ra Kim, Jong Pil Yoon, Joon Yub Kim, Sung Hoon Moon, Jieun Kwon, Hyo-Jin Lee, Young-Min Noh & Youngjun Kim (2018) Automated detection and classification of the proximal humerus fracture by using deep learning algorithm, Acta Orthopaedica, 89:4, 468 473, DOI: 10.1080/17453674.2018.1453714
[23] You Only Look Once: Unified, Real-Time Object Detection. Joseph Redmon, Santosh Divvala, Ross Girshick, Ali Farhadi. arXiv:1506.02640 [cs.CV]
[24] Rich feature hierarchies for accurate object detection and semantic segmentation. Ross Girshick, Jeff Donahue, Trevor Darrell, Jitendra Malik. arXiv:1311.2524 [cs.CV]
[25] Thian, Yee & Li, Yiting & Jagmohan, Pooja & Sia, David & Chan, Vincent & Tan, Robby. (2019). Convolutional Neural Networks for Automated Fracture Detection and Localization on Wrist Radiographs. Radiology: Artificial Intelligence. 1. e180001. 10.1148/ryai.2019180001.
[26] Chung SW, Han SS, Lee JW, Oh KS, Kim NR, Yoon JP, Kim JY, Moon SH, Kwon J, Lee HJ, Noh YM, Kim Y. Automated detection and classification of the proximal humerus fracture by using deep learning algorithm. Acta Orthop. 2018 Aug;89(4) 468-473. doi:10.1080/17453674.2018.1453714. PMID: 29577791; PMCID: PMC6066766.
[27] Ren, S., He, K., Girshick, R.B., & Sun, J. (2015). Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39, 1137-1149.
[28] Dai, Jifeng & He, Kaiming & Sun, Jian. (2016). Instance-Aware Semantic Segmentation via Multi-task Network Cascades. 3150-3158. 10.1109/CVPR.2016.343.
[29] Y. Li, H. Qi, J. Dai, X. Ji, and Y. Wei. Fully convolutional instance-aware semantic segmentation. In CVPR, 2017.
[30] T.-Y. Lin, P. Dollar, R. Girshick, K. He, B. Hariharan, and ´ S. Belongie. Feature pyramid networks for object detection. In CVPR, 2017.
[31] Jan Hosang, Rodrigo Benenson, Bernt Schiele. Learning non-maximum suppression.
[32] T. Lin, P. Dollár, R. Girshick, K. He, B. Hariharan and S. Belongie, "Feature Pyramid Networks for Object Detection," 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 936-944, doi: 10.1109/CVPR.2017.106.
[33] Jones, R.M., Sharma, A., Hotchkiss, R. et al. Assessment of a deep-learning system for fracture detection in musculoskeletal radiographs. npj Digit. Med. 3, 144 (2020). https://doi.org/10.1038/s41746-020-00352-w

簡易檢索 / 詳目顯示

相關論文