MIDeepSeg: Minimally interactive segmentation of unseen objects from medical images using deep learning

作     者：Luo, Xiangde Wang, Guotai Song, Tao Zhang, Jingyang Aertsen, Michael Deprest, Jan Ourselin, Sebastien Vercauteren, Tom Zhang, Shaoting 

作者机构：Univ Elect Sci & Technol China Sch Mech & Elect Engn Chengdu Peoples R China SenseTime Res Shanghai Peoples R China Shanghai Jiao Tong Univ Sch Biomed Engn Shanghai Peoples R China Kings Coll London Sch Biomed Engn & Imaging Sci London England Univ Hosp Leuven Dept Radiol Leuven Belgium Univ Hosp Leuven Dept Obstet & Gynaecol Leuven Belgium UCL Inst Womens Hlth London England 

出 版 物：《MEDICAL IMAGE ANALYSIS》 (医学图像分析)

年 卷 期：2021年第72卷

页      面：102102-102102页

核心收录：

学科分类：0831[工学-生物医学工程（可授工学、理学、医学学位）] 0812[工学-计算机科学与技术（可授工学、理学学位）] 1009[医学-特种医学] 10[医学] 

基　　金：National Natural Science Foundations of China [61901084, 81771921] key research and development project of Sichuan province, China [20ZDYF2817] Wellcome Trust [WT101957, 203148/Z/16/Z] Engineering and Physical Sciences Research Council (EPSRC) [NS/A000027/1, NS/A000 049/1] Medtronic/Royal Academy of Engineering Research Chair [RCSRF18194] 

主　　题：Interactive image segmentation Convolutional neural network Geodesic distance Generalization 

摘      要：Segmentation of organs or lesions from medical images plays an essential role in many clinical applications such as diagnosis and treatment planning. Though Convolutional Neural Networks (CNN) have achieved the state-of-the-art performance for automatic segmentation, they are often limited by the lack of clinically acceptable accuracy and robustness in complex cases. Therefore, interactive segmentation is a practical alternative to these methods. However, traditional interactive segmentation methods require a large number of user interactions, and recently proposed CNN-based interactive segmentation methods are limited by poor performance on previously unseen objects. To solve these problems, we propose a novel deep learning-based interactive segmentation method that not only has high efficiency due to only requiring clicks as user inputs but also generalizes well to a range of previously unseen objects. Specifically, we first encode user-provided interior margin points via our proposed exponentialized geodesic distance that enables a CNN to achieve a good initial segmentation result of both previously seen and unseen objects, then we use a novel information fusion method that combines the initial segmentation with only a few additional user clicks to efficiently obtain a refined segmentation. We validated our proposed framework through extensive experiments on 2D and 3D medical image segmentation tasks with a wide range of previously unseen objects that were not present in the training set. Experimental results showed that our proposed framework 1) achieves accurate results with fewer user interactions and less time compared with state-of-the-art interactive frameworks and 2) generalizes well to previously unseen objects. (c) 2021 Elsevier B.V. All rights reserved.