A global benchmark of algorithms for segmenting the left atrium from late gadolinium-enhanced cardiac magnetic resonance imaging |
| |
| Affiliation: | 1. Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand;2. State Key Lab of Virtual Reality Technology and Systems, Beihang University, Beijing, China;3. School of Electronics Engineering and Computer Science, Peking University, Beijing, China;4. SenseTime Inc, Shenzhen, China;5. Tencent Jarvis Laboratory, Shenzhen, China;6. Pattern Recognition Lab, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany;7. Department of Computer Science and Engineering, The Chinese University of Hong Kong, Hong Kong;8. National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, School of Biomedical Engineering, Health Science Center, Shenzhen University, Shenzhen, China;9. School of Computer Science, Wuhan University, Wuhan, China;10. Shenzhen Key Laboratory of Virtual Reality and Human Interaction Technology, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China;11. EPITA Research and Development Laboratory, Paris, France;12. Department of Computing, Imperial College London, London, United Kingdom;13. School of Naval Architecture, Ocean & Civil Engineering, Shanghai Jiao Tong University, Shanghai, China;14. School of Data Science, Fudan University, Shanghai, China;15. Inria, Université Côte d''Azur, Epione team, Sophia Antipolis, France;p. School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China;q. Department of Electric, Electronic and Information Engineering, University of Bologna, Cesena, Italy;r. Department of Biomedical Engineering, Eindhoven University of Technology, Eindhoven, the Netherlands;s. School of Biomedical Engineering & Imaging Sciences, Kings College London, London, United Kingdom;t. Faculty of Electrical Engineering and Information Technology, University of Magdeburg, Magdeburg, Germany;u. Department of Internal Medicine III, Heidelberg University Hospital, Heidelberg, Germany;v. Biomedical Engineering Center, Fudan University, Shanghai, China;w. Department of Radiology, Leiden University Medical Center, Leiden, the Netherlands;x. Physense, Department of Information and Communication Technologies, Universitat Pompeu Fabra, Barcelona, Spain;y. Department of Bioengineering, Kings College London, London, United Kingdom;1. SJTU-CU ICRC, Department of Engineering Mechanics, School of Naval Architecture, Ocean and Civil Engineering, Shanghai Jiao Tong University, P R China;2. Collaborative Innovation Center for Advanced Ship and Deep-Sea Exploration (CISSE), Shanghai 200240, P R China;3. Department of Statistics, School of Management, Fudan University, China;1. School of Computer Science and Technology, Harbin Institute of Technology, Harbin, 150001, China;2. The Department of Radiology, The Fourth Hospital of Harbin Medical University, Harbin 150001, China;3. School of Physics and Astronomy, University of Manchester, Manchester, UK;4. The Department of Medical Imaging, Western University, London, Canada;5. The Digital Imaging Group of London, London, ON N6A 3K7, Canada |
| |
| Abstract: | Segmentation of medical images, particularly late gadolinium-enhanced magnetic resonance imaging (LGE-MRI) used for visualizing diseased atrial structures, is a crucial first step for ablation treatment of atrial fibrillation. However, direct segmentation of LGE-MRIs is challenging due to the varying intensities caused by contrast agents. Since most clinical studies have relied on manual, labor-intensive approaches, automatic methods are of high interest, particularly optimized machine learning approaches. To address this, we organized the 2018 Left Atrium Segmentation Challenge using 154 3D LGE-MRIs, currently the world's largest atrial LGE-MRI dataset, and associated labels of the left atrium segmented by three medical experts, ultimately attracting the participation of 27 international teams. In this paper, extensive analysis of the submitted algorithms using technical and biological metrics was performed by undergoing subgroup analysis and conducting hyper-parameter analysis, offering an overall picture of the major design choices of convolutional neural networks (CNNs) and practical considerations for achieving state-of-the-art left atrium segmentation. Results show that the top method achieved a Dice score of 93.2% and a mean surface to surface distance of 0.7 mm, significantly outperforming prior state-of-the-art. Particularly, our analysis demonstrated that double sequentially used CNNs, in which a first CNN is used for automatic region-of-interest localization and a subsequent CNN is used for refined regional segmentation, achieved superior results than traditional methods and machine learning approaches containing single CNNs. This large-scale benchmarking study makes a significant step towards much-improved segmentation methods for atrial LGE-MRIs, and will serve as an important benchmark for evaluating and comparing the future works in the field. Furthermore, the findings from this study can potentially be extended to other imaging datasets and modalities, having an impact on the wider medical imaging community. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
