Mingrui Zhuang1, Zhonghua Chen1,2, Hongkai Wang3,4, Hong Tang1, Jiang He1, Bobo Qin1, Yuxin Yang1, Xiaoxian Jin1, Mengzhu Yu1, Baitao Jin1, Taijing Li1, Lauri Kettunen2. 1. School of Biomedical Engineering, Faculty of Electronic Information and Electrical Engineering, Dalian University of Technology, Dalian, China. 2. Faculty of Information Technology, University of Jyväskylä, Jyväskylä, Finland. 3. School of Biomedical Engineering, Faculty of Electronic Information and Electrical Engineering, Dalian University of Technology, Dalian, China. wang.hongkai@dlut.edu.cn. 4. Liaoning Key Laboratory of Integrated Circuit and Biomedical Electronic System, Dalian, China. wang.hongkai@dlut.edu.cn.
Abstract
PURPOSE: Training deep neural networks usually require a large number of human-annotated data. For organ segmentation from volumetric medical images, human annotation is tedious and inefficient. To save human labour and to accelerate the training process, the strategy of annotation by iterative deep learning recently becomes popular in the research community. However, due to the lack of domain knowledge or efficient human-interaction tools, the current AID methods still suffer from long training time and high annotation burden. METHODS: We develop a contour-based annotation by iterative deep learning (AID) algorithm which uses boundary representation instead of voxel labels to incorporate high-level organ shape knowledge. We propose a contour segmentation network with a multi-scale feature extraction backbone to improve the boundary detection accuracy. We also developed a contour-based human-intervention method to facilitate easy adjustments of organ boundaries. By combining the contour-based segmentation network and the contour-adjustment intervention method, our algorithm achieves fast few-shot learning and efficient human proofreading. RESULTS: For validation, two human operators independently annotated four abdominal organs in computed tomography (CT) images using our method and two compared methods, i.e. a traditional contour-interpolation method and a state-of-the-art (SOTA) convolutional network (CNN) method based on voxel label representation. Compared to these methods, our approach considerably saved annotation time and reduced inter-rater variabilities. Our contour detection network also outperforms the SOTA nnU-Net in producing anatomically plausible organ shape with only a small training set. CONCLUSION: Taking advantage of the boundary shape prior and the contour representation, our method is more efficient, more accurate and less prone to inter-operator variability than the SOTA AID methods for organ segmentation from volumetric medical images. The good shape learning ability and flexible boundary adjustment function make it suitable for fast annotation of organ structures with regular shape.
PURPOSE: Training deep neural networks usually require a large number of human-annotated data. For organ segmentation from volumetric medical images, human annotation is tedious and inefficient. To save human labour and to accelerate the training process, the strategy of annotation by iterative deep learning recently becomes popular in the research community. However, due to the lack of domain knowledge or efficient human-interaction tools, the current AID methods still suffer from long training time and high annotation burden. METHODS: We develop a contour-based annotation by iterative deep learning (AID) algorithm which uses boundary representation instead of voxel labels to incorporate high-level organ shape knowledge. We propose a contour segmentation network with a multi-scale feature extraction backbone to improve the boundary detection accuracy. We also developed a contour-based human-intervention method to facilitate easy adjustments of organ boundaries. By combining the contour-based segmentation network and the contour-adjustment intervention method, our algorithm achieves fast few-shot learning and efficient human proofreading. RESULTS: For validation, two human operators independently annotated four abdominal organs in computed tomography (CT) images using our method and two compared methods, i.e. a traditional contour-interpolation method and a state-of-the-art (SOTA) convolutional network (CNN) method based on voxel label representation. Compared to these methods, our approach considerably saved annotation time and reduced inter-rater variabilities. Our contour detection network also outperforms the SOTA nnU-Net in producing anatomically plausible organ shape with only a small training set. CONCLUSION: Taking advantage of the boundary shape prior and the contour representation, our method is more efficient, more accurate and less prone to inter-operator variability than the SOTA AID methods for organ segmentation from volumetric medical images. The good shape learning ability and flexible boundary adjustment function make it suitable for fast annotation of organ structures with regular shape.
Authors: Ivo Wolf; Marcus Vetter; Ingmar Wegner; Thomas Böttger; Marco Nolden; Max Schöbinger; Mark Hastenteufel; Tobias Kunert; Hans-Peter Meinzer Journal: Med Image Anal Date: 2005-12 Impact factor: 8.545
Authors: Paul A Yushkevich; Artem Pashchinskiy; Ipek Oguz; Suyash Mohan; J Eric Schmitt; Joel M Stein; Dženan Zukić; Jared Vicory; Matthew McCormick; Natalie Yushkevich; Nadav Schwartz; Yang Gao; Guido Gerig Journal: Neuroinformatics Date: 2019-01
Authors: Daniel Haehn; Seymour Knowles-Barley; Mike Roberts; Johanna Beyer; Narayanan Kasthuri; Jeff W Lichtman; Hanspeter Pfister Journal: IEEE Trans Vis Comput Graph Date: 2014-12 Impact factor: 4.579
Authors: Bowen Lin; Shujun Fu; Yuting Lin; Ronny L Rotondo; Weizhang Huang; Harold H Li; Ronald C Chen; Hao Gao Journal: Phys Med Biol Date: 2021-12-02 Impact factor: 4.174
Authors: Guotai Wang; Maria A Zuluaga; Wenqi Li; Rosalind Pratt; Premal A Patel; Michael Aertsen; Tom Doel; Anna L David; Jan Deprest; Sebastien Ourselin; Tom Vercauteren Journal: IEEE Trans Pattern Anal Mach Intell Date: 2018-06-01 Impact factor: 6.226
Authors: Imanol Luengo; Michele C Darrow; Matthew C Spink; Ying Sun; Wei Dai; Cynthia Y He; Wah Chiu; Tony Pridmore; Alun W Ashton; Elizabeth M H Duke; Mark Basham; Andrew P French Journal: J Struct Biol Date: 2017-02-27 Impact factor: 2.867