Seeing Is Believing: Video Classification for Computed Tomographic Colonography Using Multiple-Instance Learning

被引:14
作者
Wang, Shijun [1 ]
McKenna, Matthew T. [1 ]
Nguyen, Tan B. [1 ]
Burns, Joseph E. [2 ]
Petrick, Nicholas [3 ]
Sahiner, Berkman [3 ]
Summers, Ronald M. [1 ]
机构
[1] NIH, Bethesda, MD 20892 USA
[2] Univ Calif Irvine, Dept Radiol Sci, Orange, CA 92868 USA
[3] US FDA, Ctr Devices & Radiol Hlth, Silver Spring, MD 20993 USA
关键词
Computed tomographic colonography (CTC); multiple-instance learning; semidefinite programming; video analysis; AIDED DETECTION; CT COLONOGRAPHY; POLYP DETECTION; VIRTUAL COLONOSCOPY; SYSTEM; EVOLUTION; STANDARD;
D O I
10.1109/TMI.2012.2187304
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
In this paper, we present development and testing results for a novel colonic polyp classification method for use as part of a computed tomographic colonography (CTC) computer-aided detection (CAD) system. Inspired by the interpretative methodology of radiologists using 3-D fly-through mode in CTC reading, we have developed an algorithm which utilizes sequences of images (referred to here as videos) for classification of CAD marks. For each CAD mark, we created a video composed of a series of intraluminal, volume-rendered images visualizing the detection from multiple viewpoints. We then framed the video classification question as a multiple-instance learning (MIL) problem. Since a positive (negative) bag may contain negative (positive) instances, which in our case depends on the viewing angles and camera distance to the target, we developed a novel MIL paradigm to accommodate this class of problems. We solved the new MIL problem by maximizing a L2-norm soft margin using semidefinite programming, which can optimize relevant parameters automatically. We tested our method by analyzing a CTC data set obtained from 50 patients from three medical centers. Our proposed method showed significantly better performance compared with several traditional MIL methods.
引用
收藏
页码:1141 / 1153
页数:13
相关论文
共 54 条
[1]  
Ahonen T, 2009, LECT NOTES COMPUT SC, V5575, P61, DOI 10.1007/978-3-642-02230-2_7
[2]  
[Anonymous], P 2009 7 INT WORKSH
[3]  
[Anonymous], 2002, P NEURIPS, DOI DOI 10.5555/2968618.2968690
[4]  
[Anonymous], RADIOLOGY
[5]  
[Anonymous], 2002, ICML
[6]  
[Anonymous], 2000, 17th International Conference on Machine Learning
[7]  
[Anonymous], 2011, ACM T INTEL SYST TEC, DOI DOI 10.1145/1961189.1961199
[8]  
[Anonymous], 2007, Advances in Neural Information Processing Systems
[9]   BilVideo-7: An MPEG-7-Compatible Video Indexing and Retrieval System [J].
Bastan, Muhammet ;
Cam, Hayati ;
Gudukbay, Ugur ;
Ulusoy, Ozgur .
IEEE MULTIMEDIA, 2010, 17 (03) :62-73
[10]  
Bejakovic S, 2009, IEEE INT CONF ROBOT, P3755