A New Look at Gray-level Co-occurrence for Multi-scale Texture Descriptor with Applications to Characterize Colorectal Polyps via Computed Tomographic Colonography

Author(s):  
Weiguo Cao ◽  
Zhengrong Liang ◽  
Marc Pomeroy ◽  
Perry J. Pickhardt ◽  
Matthew A. Barish ◽  
...  
Author(s):  
Weiguo Cao ◽  
Marc J. Pomeroy ◽  
Yongfeng Gao ◽  
Matthew A. Barish ◽  
Almas F. Abbasi ◽  
...  

AbstractTexture features have played an essential role in the field of medical imaging for computer-aided diagnosis. The gray-level co-occurrence matrix (GLCM)-based texture descriptor has emerged to become one of the most successful feature sets for these applications. This study aims to increase the potential of these features by introducing multi-scale analysis into the construction of GLCM texture descriptor. In this study, we first introduce a new parameter - stride, to explore the definition of GLCM. Then we propose three multi-scaling GLCM models according to its three parameters, (1) learning model by multiple displacements, (2) learning model by multiple strides (LMS), and (3) learning model by multiple angles. These models increase the texture information by introducing more texture patterns and mitigate direction sparsity and dense sampling problems presented in the traditional Haralick model. To further analyze the three parameters, we test the three models by performing classification on a dataset of 63 large polyp masses obtained from computed tomography colonoscopy consisting of 32 adenocarcinomas and 31 benign adenomas. Finally, the proposed methods are compared to several typical GLCM-texture descriptors and one deep learning model. LMS obtains the highest performance and enhances the prediction power to 0.9450 with standard deviation 0.0285 by area under the curve of receiver operating characteristics score which is a significant improvement.


2003 ◽  
Vol 125 (2) ◽  
pp. 311-319 ◽  
Author(s):  
C.Daniel Johnson ◽  
William S Harmsen ◽  
Lynn A Wilson ◽  
Robert L Maccarty ◽  
Timothy J Welch ◽  
...  

2009 ◽  
Vol 46 (2) ◽  
pp. 90-96 ◽  
Author(s):  
Odery Ramos Jr. ◽  
César Luiz Boguszewski ◽  
Sandra Teixeira ◽  
Ricardo De Bem ◽  
Benito Parolim ◽  
...  

CONTEXT: Acromegalic patients have better chances to develop colorectal polyps and cancer and, considered a high-risk group, need to undergo frequent screening examinations. Moreover, in acromegalia, the increased bowel length and the intestinal loop complexity can lead to higher levels of technical difficulties and increase the risks of complications at conventional colonoscopy. Computed tomographic colonography, also known as virtual colonoscopy, is an innovative and secure technology which is revolutionizing the diagnosis of colon and rectum neoplasias. OBJECTIVE: To analyze computed tomographic colonography performance for the screening of colorectal polyps in acromegalic patients. METHODS: A prospective study of 21 asymptomatic acromegalic patients, 12 male and 9 female, average age 49, who underwent computed tomographic colonography and conventional colonoscopy. Computed tomographic colonography was performed with a GE Helical Multislice Computed Tomography Apparatus. Conventional colonoscopy was performed in the same day, without previous knowledge of the computed tomographic colonography diagnostics. The study evaluated the capacity of computed tomographic colonography to detect patients with colorectal polyps and identify each colorectal lesion described by the colonoscopy. RESULTS: In two patients (2/21), conventional colonoscopy was incomplete. However, in all patients computed tomographic colonography was complete. In Phase I ("per patient"), computed tomographic colonography diagnosed eight of the nine patients with colorectal polyps and showed 88% sensitivity, 75% specificity and 81% accuracy. In Phase II ("per polyp"), out of the 21 acromegalic patients included in this study, 12 presented normal findings at conventional colonoscopy. A total of 19 polyps were identified in 9 patients. Ten of the 19 polyps were smaller than 10 mm, and 9 were equal to or larger than 10 mm. Computed tomographic colonography identified 7 of the 9 polyps >10 mm described by conventional colonoscopy and only 6 of the 10 small polyps identified at conventional colonoscopy were detected by computed tomographic colonography. The histological analysis of resected lesions revealed 12 tubular adenomas, 6 hyperplastic polyps and 1 colonic tubulo-villous adenoma with an adenocarcinoma focus. CONCLUSION: The authors present the first reports of computed tomographic colonography in the screening of colorectal polyps in acromegalic patients. In this study, computed tomographic colonography was performed without complications and a complete and safe colorectal evaluation was possible in all acromegalic patients. Moreover, computed tomographic colonography presented good sensitivity, specificity and accuracy for the identification of acromegalic patients with polyps of any size and better results in the diagnosis of large polyps, when they were compared to small polypoid lesions.


2008 ◽  
Author(s):  
Odery Ramos Jr ◽  
César Luiz Boguszewski ◽  
Sandra Teixeira ◽  
Anke Boeving ◽  
Ricardo De Bem ◽  
...  

2004 ◽  
Vol 127 (5) ◽  
pp. 1300-1311 ◽  
Author(s):  
Riccardo Iannaccone ◽  
Andrea Laghi ◽  
Carlo Catalano ◽  
Filippo Mangiapane ◽  
Antonietta Lamazza ◽  
...  

2004 ◽  
Vol 3 (2) ◽  
pp. 201-207 ◽  
Author(s):  
Martin Heuschmid ◽  
Oliver Luz ◽  
Juergen F. Schaefer ◽  
Andreas F. Kopp ◽  
Claus D. Claussen ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document