Title :
Investigations on ROI selection for liver classification
Author :
Singh, Monika ; Singh, Sushil ; Gupta, Swastik
Author_Institution :
EIED, Thapar Univ., Patiala, India
Abstract :
In medical imaging a Region of Interest (ROI) is a sub part of an image, which contains the vital information related to the diagnosis. Since an ROI is used as a `representative´ of the image, and all further computations and diagnosis depends upon the ROI, therefore, it is very crucial to select an appropriate image area as ROI. Further, the use of an ROI will reduce the computational time to extract the features from sub part rather than from the whole image. This paper investigates the importance of location, shape and size of the ROI for liver tissue characterization. The impact of ROI has been analyzed through texture features extracted from that ROI, and subsequently the experimental analysis for selection of ROI is presented.
Keywords :
biological tissues; feature extraction; feature selection; image classification; image texture; liver; medical image processing; ROI texture feature extraction; image representation; liver tissue characterization; liver tissue classification; medical imaging; region-of-interest selection; Biomedical imaging; Feature extraction; Lesions; Liver diseases; Shape; Ultrasonic imaging;
Conference_Titel :
Electrical and Computer Engineering (CCECE), 2014 IEEE 27th Canadian Conference on
Conference_Location :
Toronto, ON
Print_ISBN :
978-1-4799-3099-9
DOI :
10.1109/CCECE.2014.6900942
