Preliminary results of computer aided system with the 2nd-generation narrow-band imaging for endoscopic screening of colorectal neoplasms

We propose a novel computer recognition system with magnifying narrow-band imaging (NBI) colonoscopy to predict pathologic diagnosis of colorectal neoplasms. In this work, we propose a system based on the sparse representation of features derived from the Bank of Binarized Statistical Image Features...

Full description

Saved in:
Bibliographic Details
Published in2017 International Conference on Applied System Innovation (ICASI) pp. 854 - 857
Main Authors Wen-Jia Kuo, Li-Yun Wang, Pen-Jen Chen
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.05.2017
Subjects
Binarized Statistical Image Features
Correlation
Imaging
Maximum likelihood detection
Narrow-Band Imaging
Neoplasms
NICE
Nonlinear filters
Support vector machines
SVM
Tumors
Online AccessGet full text

Cover

More Information
Summary:We propose a novel computer recognition system with magnifying narrow-band imaging (NBI) colonoscopy to predict pathologic diagnosis of colorectal neoplasms. In this work, we propose a system based on the sparse representation of features derived from the Bank of Binarized Statistical Image Features (B-BSIF). The BSIF code of a pixel is considered as a local descriptor of the image intensity pattern in the pixel's surroundings. Two-dimensional autocorrelation coefficients were used to represent the interpixel correlation of image for further differentiation. A two-layered support vector machine (SVM) is then used to classify the type of colorectal neoplasm. Finally, the prediction results will be considered to determine the final prediction differential diagnosis result. Experimental results show that the proposed computer-aided system can achieved with high diagnostic performance for predicting the histology of colorectal tumors by using NBI magnifying colonoscopy.
DOI:10.1109/ICASI.2017.7988568