Volumetric texture features from higher-order images for diagnosis of colon lesions via CT colonography

被引:84
作者
Song, Bowen [1 ,2 ]
Zhang, Guopeng [3 ]
Lu, Hongbing [3 ]
Wang, Huafeng [1 ]
Zhu, Wei [2 ]
Pickhardt, Perry J. [4 ]
Liang, Zhengrong [1 ]
机构
[1] SUNY Stony Brook, Dept Radiol, Stony Brook, NY 11790 USA
[2] SUNY Stony Brook, Dept Appl Math & Stat, Stony Brook, NY 11790 USA
[3] Fourth Mil Med Univ, Dept Biomed Engn, Xian 710032, Shaanxi, Peoples R China
[4] Univ Wisconsin, Sch Med & Publ Hlth, Dept Radiol, Madison, WI 53792 USA
基金
中国国家自然科学基金;
关键词
CT colonography; Colorectal lesions; Texture feature; Textural biomarker; Gradient; Curvature; Computer-aided diagnosis; COMPUTER-AIDED DIAGNOSIS; POLYP DETECTION; SIZE; CLASSIFICATION; SEGMENTATION; COLONOSCOPY; PREVALENCE;
D O I
10.1007/s11548-014-0991-2
中图分类号
R318 [生物医学工程];
学科分类号
100103 [病原生物学];
摘要
Differentiation of colon lesions according to underlying pathology, e.g., neoplastic and non-neoplastic lesions, is of fundamental importance for patient management. Image intensity-based textural features have been recognized as useful biomarker for the differentiation task. In this paper, we introduce texture features from higher-order images, i.e., gradient and curvature images, beyond the intensity image, for that task. Based on the Haralick texture analysis method, we introduce a virtual pathological model to explore the utility of texture features from high-order differentiations, i.e., gradient and curvature, of the image intensity distribution. The texture features were validated on a database consisting of 148 colon lesions, of which 35 are non-neoplastic lesions, using the support vector machine classifier and the merit of area under the curve (AUC) of the receiver operating characteristics. The AUC of classification was improved from 0.74 (using the image intensity alone) to 0.85 (by also considering the gradient and curvature images) in differentiating the neoplastic lesions from non-neoplastic ones, e.g., hyperplastic polyps from tubular adenomas, tubulovillous adenomas and adenocarcinomas. The experimental results demonstrated that texture features from higher-order images can significantly improve the classification accuracy in pathological differentiation of colorectal lesions. The gain in differentiation capability shall increase the potential of computed tomography colonography for colorectal cancer screening by not only detecting polyps but also classifying them for optimal polyp management for the best outcome in personalized medicine.
引用
收藏
页码:1021 / 1031
页数:11
相关论文
共 46 条
[1]
American Cancer Society, 2012, Cancer Facts and Figures 2012
[2]
[Anonymous], 2005, ACR Practical Guideline, V29, P295
[3]
[Anonymous], COMPUTER ROBOT VISIO
[4]
[Anonymous], P NORW C IM PROC PAT
[5]
Texture analysis of medical images [J].
Castellano, G ;
Bonilha, L ;
Li, LM ;
Cendes, F .
CLINICAL RADIOLOGY, 2004, 59 (12) :1061-1069
[6]
LIBSVM: A Library for Support Vector Machines [J].
Chang, Chih-Chung ;
Lin, Chih-Jen .
ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY, 2011, 2 (03)
[7]
USING CANNY CRITERIA TO DERIVE A RECURSIVELY IMPLEMENTED OPTIMAL EDGE DETECTOR [J].
DERICHE, R .
INTERNATIONAL JOURNAL OF COMPUTER VISION, 1987, 1 (02) :167-187
[8]
SCREENING FOR COLORECTAL-CANCER [J].
EDDY, DM .
ANNALS OF INTERNAL MEDICINE, 1990, 113 (05) :373-384
[9]
Engel K., 2006, Real-time volume graphics
[10]
An introduction to ROC analysis [J].
Fawcett, Tom .
PATTERN RECOGNITION LETTERS, 2006, 27 (08) :861-874